Citation
Masculinity and whiteness in public organizations : managing institutional race and gender in federal agencies

Material Information

Title:
Masculinity and whiteness in public organizations : managing institutional race and gender in federal agencies
Creator:
Heckler, Nuri
Place of Publication:
Denver, CO
Publisher:
University of Colorado Denver
Publication Date:
Language:
English

Thesis/Dissertation Information

Degree:
Doctorate ( Doctor of philosophy)
Degree Grantor:
University of Colorado Denver
Degree Divisions:
School of Public Affairs, CU Denver
Degree Disciplines:
Public affairs
Committee Chair:
Ronquillo, John
Committee Members:
Schaible, Lonnie
Stivers, Camilla

Notes

Abstract:
This study examines the institutions of Whiteness and Masculinity and their influence on public organizations. Whiteness and Masculinity are institutions, collections of rules, norms, and culture that benefit whites and men. These institutions are under-researched because they are epistemologically veiled behind color- and gender-blindness. Nevertheless, they disproportionately influence public organizations including the federal government that has been dominated by white men in all three branches, and historically studied by mostly white men scholars. This study unveils Masculinity and Whiteness by identifying performances of these institutions that are likely to appear in public organizations and using those performances to explore two research questions. 1) What effects do Whiteness and Masculinity have on public employees’ work life? 2) How do public organizations, public managers, and public administrators perform Whiteness and Masculinity. The research was conducted in two stages. In the first stage, publicly-available data was analyzed using clustered fixed effects on white/men/white men job satisfaction, management satisfaction, and public service motivation when a larger proportion of their managers are also whites/men/white men. The quantitative data are then analyzed using machine learning tools to explore relationships between myriad independent variables, and satisfaction and public service motivation for whites/men/white men compared to their people-of-color/women colleagues. In the next stage, the quantitative data identify work units that are typical of the federal government for in-depth interviews to conduct institutional ethnography. The findings indicate that many of the performances of Masculinity and Whiteness documented in the literature are present in federal work units, and that new performances have evolved that are meaningfully impacting federal service. Examples include the myth of a generation gap that is justifying inaction especially among younger federal workers, and a complex relationship between color-blindness, alternative work arrangements, and camaraderie. From these data, a map of institutional Whiteness and Masculinity is developed that is leveraged to make suggestions for better managing the race and gender institutions dominant in most public organizations.

Record Information

Source Institution:
University of Colorado Denver
Holding Location:
Auraria Library
Rights Management:
Copyright Nuri Heckler. Permission granted to University of Colorado Denver to digitize and display this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.

Downloads

This item has the following downloads:


Full Text
MASCULINITY AND WHITENESS IN PUBLIC ORGANIZATIONS:
MANAGING INSTITUTIONAL RACE AND GENDER IN FEDERAL AGENCIES
by
Nuri Heckler
B.A., City College of New York, 2005 JD., University of Denver, 2008
A Dissertation submitted to the Faculty of the Graduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Public Affairs
2019


11
© 2019
NURI HECKLER
ALL RIGHTS RESERVED


Ill
This dissertation for the Doctor of Philosophy degree by Nuri Heckler has been approved for the Public Affairs Program by
Mary Ellen Guy, Chair John Ronquillo Lonnie Schaible Camilla Stivers
Date: May 18,2019


Heckler, Nuri (Ph.D., Public Affairs Program)
Whiteness and Masculinity in Public Organizations: Managing Institutional Race and Gender in Federal Agencies
Dissertation directed by Professor Mary Ellen Guy
ABSTRACT
This study examines the institutions of Whiteness and Masculinity and their influence on public organizations. Whiteness and Masculinity are institutions, collections of rules, norms, and culture that benefit whites and men. These institutions are under-researched because they are epistemologically veiled behind color- and gender-blindness. Nevertheless, they disproportionately influence public organizations including the federal government that has been dominated by white men in all three branches, and historically studied by mostly white men scholars. This study unveils Masculinity and Whiteness by identifying performances of these institutions that are likely to appear in public organizations and using those performances to explore two research questions. 1) What effects do Whiteness and Masculinity have on public employees’ work life? 2) How do public organizations, public managers, and public administrators perform Whiteness and Masculinity. The research was conducted in two stages. In the first stage, publicly-available data was analyzed using clustered fixed effects on white/men/white men job satisfaction, management satisfaction, and public service motivation when a larger proportion of their managers are also whites/men/white men. The quantitative data are then analyzed using machine learning tools to explore relationships between myriad independent variables, and satisfaction and public service motivation for whites/men/white men compared to their people-of-color/women colleagues. In the next stage, the quantitative data identify work units that are typical of the federal government for in-depth interviews to conduct


V
institutional ethnography. The findings indicate that many of the performances of Masculinity and Whiteness documented in the literature are present in federal work units, and that new performances have evolved that are meaningfully impacting federal service. Examples include the myth of a generation gap that is justifying inaction especially among younger federal workers, and a complex relationship between color-blindness, alternative work arrangements, and camaraderie. From these data, a map of institutional Whiteness and Masculinity is developed that is leveraged to make suggestions for better managing the race and gender institutions dominant in most public organizations.
The form and content of this abstract are approved. I recommend its publication.
Approved: Mary Ellen Guy


This dissertation is dedicated to Micky, Luca, and Pippin who serve as my examples of meaningful and authentic engagement with justice and equality. May this work share some
our love with the universe.


ACKNOWLEDGEMENTS
Without Dr. Mary E. Guy, public administration research and practice would be far less interesting and just. Her contributions to this dissertation literally changed my life and made every part of this research possible. Her model of courage and uncompromising good humor in the face of serious issues is reflected in diminished form in this dissertation.
I also need to thank Dr. Cheryl Matias whose leadership, mentorship, and radical mothering challenged me to better understand myself as a white man.
Many of my law professors influenced this dissertation, including Robert Chang, Catherine Smith, Rashmi Goel, Andy Reid, and the late Ann Scales who I believe is still having fun and raising hell.
Of the many mentors that I have had at CU Denver, Dr. John Ronquillo, Dr. Lonnie Schaible, Dr. Seba Bishu, Dr. Tanya Heikkila, Dr. Todd Ely, and Dr. Jessica Sowa left their thumbprints all over this research. Each of these experts imparted wisdom and knowledge, and I would never have been introduced to any of them without the investment that the School of Public Affairs made in my training. Thank you all.
Thank you also to the anonymous informants. Your willingness to participate and your openness was both unexpected and refreshing. I wish I could get to know all of you better.
I also need to thank Dr. Camilla Stivers and Dr. Brandi Blessett who modeled careers and ways of thinking about public administration that made me feel I could bring my true self to my research and pursue it conscientiously and without apology. Along with the Critical Perspectives CIG at ARNOVA, they convinced me that academia was a life that could be lived authentically driven by hard work, intelligence, and mercy.
Last but not least, thank you Mom. I am who I am because of you.


TABLE OF CONTENTS
CHAPTER
I- The Personal is Public Administration................................................1
Epistemological and Empirical Gaps...................................................2
Marking Whiteness and Masculinity....................................................5
Whiteness and Masculinity Research in Public Administration.........................11
Conclusion: Research on Whiteness and Masculinity in an Institutional Framework.....18
II- Institutional Whiteness and Masculinity: A Theory of Intersectionality..............20
The Institutional Framework.........................................................21
Public Organizations................................................................25
Emotionally Bounded Individuals.....................................................27
Whiteness and Masculinity...........................................................28
Society-level Whiteness and Masculinity.............................................30
Individual-level Whiteness and Masculinity..........................................37
Organizational-level Whiteness and Masculinity......................................46
Whiteness and Masculinity in Public Service Motivation and Job Satisfaction.........53
Conclusion..........................................................................57
III- RESEARCH DESIGN AND METHODOLOGY....................................................58
Stage One: Race and Gender in the Federal Employee Viewpoint Survey.................58
Hypotheses: Race and Gender Affect Job Satisfaction and Public Service Motivation...60
Data: the FEVS......................................................................60
Independent Variables: Race and Gender Congruence
62


ix
Dependent Variables: PSM and Job Satisfaction......................................64
Control Variables..................................................................65
Analysis...........................................................................67
Stage-One Limitations..............................................................71
Using Stage-One Findings to Situate Stage Two Findings.............................73
Stage two: Institutional Ethnography of Federal Offices............................74
Case Selection: ‘Typical’ Organizations in the Federal Government..................76
Data Collection: Interviews and Social Media Observation...........................77
Stage Two Analysis: Coding, Hermeneutics, Mapping, and Ethnomethodology............80
Ethical Considerations.............................................................83
Stage Two Limitations..............................................................84
Conclusion.........................................................................86
IV. QUANTITATIVE FINDINGS...........................................................87
Summary Statistics: Disentangling Masculinity and Whiteness........................87
Job and Management Satisfaction....................................................88
Public Service Motivation..........................................................90
Additional Independent Variables...................................................91
Individual-level Logit Analysis....................................................94
Decision Trees: Machine Learning about Whiteness and Masculinity..................101
Conclusion.......................................................................113
V. QUALITATIVE FINDINGS.............................................................114
Diversity Management, Color-blindness, and Alternative Work Arrangements..........116
Masculinity, (Whiteness), and Abstract Objectivity................................128


X
Other Performances of Whiteness and Masculinity...................................134
Support for Propositions..........................................................151
Conclusion........................................................................154
VI. CONCLUSION........................................................................155
Managing Whiteness and Masculinity in Public Organizations........................155
Mapping Whiteness and Masculinity in Federal Agencies.............................156
Using the Map to Diagnose and Manage Whiteness and Masculinity...............160
Managing Linkages ................................................................161
Managing Performances.............................................................164
Managing Complexes................................................................165
Implications of Institutional Masculinity and Whiteness for Research..............170
Conclusion: Discovering and Pulling Social Justice Levers in Public Organizations.171
REFERENCES............................................................................173
APPENDIX
195


XI
LIST OF TABLES
TABLE
2.1: Society-level performances of Masculinity and Whiteness...............................31
2.2: Individual-level performances of Whiteness and Masculinity............................38
2.3: Organizational-level performances of Whiteness and Masculinity........................47
3.1: Stage one hypotheses...................................................................61
3.2: Similarities between the FEVS and the federal workforce generally.....................61
3.3: Independent variables summary statistics..............................................63
3.4: Dependent variables summary statistics.................................................66
3.5: Control variables summary statistics...................................................66
3.6: Analysis strategies for overcoming limitations in the FEVS dataset....................68
3.7: Stage two propositions.................................................................75
4.1: Unit clustered categorical predicted means after controlling for time, age, military status,
and tenure in federal government work......................................................89
4.2: Unit clustered categorical predicted means after controlling for time, age, military status,
and tenure in federal government work.......................................................89
4.3: Individual-level logit regressions on Job Satisfaction.................................96
4.4: Individual-level logit regressions on management satisfaction..........................98
4.5: Individual-level logit regressions on PSM.............................................100
4.6: Table of confirmation or rejection of hypotheses......................................102
5.1: Level of support of propositions......................................................152


xii
LIST OF FIGURES
FIGURE
2.1: Model of institutional effects of Whiteness and Masculinity on public organizations.22
2.2: Masculinity and Whiteness in the institutional framework............................24
3.1: Model of the relationship between stage one and stage two research..................59
4.1: Descriptions of example units derived from CART.....................................105
6.1: Map of Whiteness and Masculinity Performances in Federal Work Units.................157
Al: Unit-level ANOVA regression tree on Comparative White Job Satisfaction...............195
A2: Unit-level ANOVA regression tree on Comparative White Management Satisfaction........196
A3: Unit-level ANOVA regression tree on Comparative White PSM............................197
A4: Unit-level ANOVA regression tree on Comparative Men’s Job Satisfaction...............198
A5: Unit-level ANOVA regression tree on Comparative Men’s Management Satisfaction........199
A6: Unit-level ANOVA regression tree on Comparative Men’s PSM............................200
A7: Unit-level ANOVA regression tree on Comparative White Men’s Job Satisfaction.........201
A8: Unit-level ANOVA regression tree on Comparative White Men’s Management
Satisfaction.............................................................................202
A9: Unit-level ANOVA regression tree on Comparative White Men’s PSM.....................203


1
CHAPTER 1
THE PERSONAL IS PUBLIC ADMINISTRATION
Feminist scholars argue “the personal is political” (Hanisch, 2006), and critical race scholars argue “the political is personal” (E. O. O’Brien, 2003). Following these cues, this dissertation argues that the personal is public administration. Behaviors of race and gender learned in childhood homes, primary and secondary education, and adult relationships influence how public administrators do their jobs. For an outsized number of public administrators, Whiteness and Masculinity are their primary race and gender influences.
This dissertation focuses on a significant gap surrounding Whiteness and Masculinity in public administration research. The gap is highlighted by considering two findings side-by-side. (1) Representative bureaucracy studies indicate that a public servants’ race and gender affect organizational outcomes (Selden, 1997). (2) The white racial identity and the masculine gender identity are the most represented in many if not most American public organizations (Guy & Newman, 2005). This study’s overarching research question fits in this gap. How do Whiteness and Masculinity affect public organizations?
For public administration scholars, Whiteness and Masculinity are particularly salient and problematic. Representative bureaucracy scholars find that teacher gender affects how well girls perform on math tests in public schools (Reiser, Wilkins, Meier, & Holland, 2002), police officer gender affects willingness of victims to report domestic violence (Theobald & Haider-Markel, 2009), and spokesperson gender affects willingness of citizens to cooperate with municipal recycling programs (Riccucci, Van Ryzin, & Li, 2015). Teacher race affects whether children of color are referred to gifted and talented programs (Elhoweris, Mutua, Alsheikh, & Holloway,


2
2005) or referred to the principal’s office for discipline (Pigott & Cowen, 2000). Welfare officer race affects interactions with recipients (Watkins-Hayes, 2011), and physician race affects the belief that patients are telling the truth about reported levels of pain (Meghani, Byun, & Gallagher, 2012). If public administration is a reflection of institutions (Wamsley et al., 1990) influenced by white men like the founders of the nation (Hamilton, Madison, & Jay, 1894; Rohr, 1990), the white men who originally set out to self-consciously study American public administration (Hoffman, 2002; W. Wilson, 1887), and 44 of 45 American presidents, then understanding Whiteness’ and Masculinity’s influence is necessary for understanding public organizations.
This chapter describes the blind spot around Whiteness and Masculinity in public administration and public organization research drawing from Mills’ (1997) concept of the epistemological racial contract, and nervousness as a mechanism of that epistemology of ignorance (Gooden, 2014; Mills, 2007). The next section begins illuminating that blind spot by elucidating Whiteness and Masculinity. The last section outlines the implications of Masculinity and Whiteness research for research on social equity, representative bureaucracy, diversity management, and institutional organizational theory. The chapter’s conclusion is a summary of the study’s primary argument that most public organizations consciously or unconsciously reflect Whiteness and Masculinity.
Epistemological and Empirical Gaps
There is a gap in thinking, an epistemological gap around Whiteness and Masculinity, which in turn causes an empirical gap, a gap in knowledge and understanding. Take some examples from the representative bureaucracy literature. Reiser et al. (2002) find that girls improve math test scores when in contact with women who are successful in math. If the study


3
was focused on Masculinity, the findings could just as easily indicate that something about the way men teach math is hurting the potential math scores of their girl students. Similarly, Selden (1997) finds that bureaucrats of color are more likely than white bureaucrats to provide subsidized loans to farmers of color. Focusing on Whiteness, the finding could just as easily have been that white bureaucrats make loan determinations that discriminate against people of color (people of color). Representative bureaucracy scholarship has made substantial contributions in empirical findings about how people of color and women impact the public organizations in which they work. However, because of epistemological gaps around Masculinity and Whiteness, little is known about the impact of the race and gender that dominate many if not most public organizations.
Epistemological gaps are features of Whiteness and Masculinity rather than academic oversight. Mills (1997) argues that race is a political, moral, and epistemological contract. The epistemological contract is an agreement to “misinterpret reality”, “to see the world wrongly, but with the assurance that this set of mistaken perceptions will be validated...” by society (Mills, 1997, p. 34). To protect this misinterpretation of reality, scholars find that race is a “nervous area of government” (Gooden, 2014), that is “unspeakable” (W. Ayers, 2004), “unthinkable” (Foldy & Buckley, 2014), and “invisible” (Sue, 2004). Of all the racial identities, Whiteness is the most susceptible to misinterpretation (Mills, 1997; Yancy, 2008). Whiteness is commonly conceived only as “non-Othemess” (Fanon, 1963; Yancy, 2008), a neutral blank slate against which to perceive other races as inferior (Mills, 2007). The epistemological contract makes discussion of Whiteness taboo (Carbado & Gulati, 2013), and contributes to the gap surrounding Whiteness as an academic study (Zuberi & Bonilla-Silva, 2008).
The epistemological gap surrounding Masculinity is similar to the gap surrounding


4
Whiteness. Collinson and Hearn (1996) argue there is a “strange silence” (p. 1) on the topic of Masculinity and the workplace. This silence “cannot be explained by either carelessness or conspiracy: the silence around these issues is built into the very process of their reproduction” (p. vii). Ferguson (1984) argues that “ideologically invisible” (p. 16) production of thought maintains the patriarchic bureaucracy, while Stivers (1993) argues that the “apparent neutrality” (p. 5) of masculine public administration maintains its legitimacy. Like Mills (1997), these scholars find that misinterpretation is a key feature of Masculinity, indicating a gender epistemological contract working alongside the racial contract.
If there are racial and gender epistemological contracts, then social scientists are signatories to these agreements. Zuberi and Bonilla-Silva (2008) edit a volume dedicated to understanding how the assumptions of most social science methods are based on a logic of white racial superiority. In their chapter, Zuberi and Bonilla-Silva (2008, pp. 330-331) point out that most analyses of race and gender focus on women, blacks, and workers as problems injected into an experimental laboratory that is assumed raceless and genderless. Zuberi (2000) details the history of sociology from the perspective of a quantitative sociologist of color, finding that many of the basic statistical regression methods used today were deemed reliable on field tests proving the basic tenet of eugenic science that white people are superior. Similarly, Pini and Pease (2013) edit a volume critiquing the masculinity of modern methods and masculine scientific theorizing. Pini and Pease (2013) find that “real-men” researchers reinforce “hegemonic masculinity” in their findings. Relatedly, Stivers (2000) argues that researchers have mostly abandoned questions and methods pioneered by women to the detriment of both women and public administration research. Meanwhile, Gooden (2014) argues that the study of race is “stifled” (p. 3) by nervousness, and Alexander and Stivers (2010) contend that “scholarship in public


5
administration reflects assumptions of professional neutrality and innocence, and a lack of awareness” of how race impacts administration. These authors call for deliberate coloring and gendering of social science research.
Just as a student of biology must stain a slide with iodine before examining a microscopic cell, Whiteness and Masculinity researchers must deliberately color their world view. Du Bois (1980 [1898]) laments that sociology focuses on how the problems of black America impact the nation’s “white inhabitants” and argues for a “critical” “point of view” (p. 77). A century later, critical whiteness studies (CWS) and multiple masculinities (MM) scholars shift perspectives by explicitly problematizing Whiteness and Masculinity (D. L. Collinson & Hearn, 1996; Harris, 1993). Like staining a microscopic slide, this shift in perspective is a deliberate attempt to mark the unmarked Masculinity and Whiteness that pervade daily life (Frankenberg, 2001; D. E.
Smith, 1987). This marking of phenomena is an important step toward filling the gaps surrounding Whiteness and Masculinity.
Marking Masculinity and Whiteness
To mark Whiteness and Masculinity as affirmative notions rather than the absence of otherness means starting with clear definitions (Bar On, 1993; Frankenberg, 2001; Goertz, 2006). To that end, Whiteness is a collection of performances that reinforce white supremacy. Masculinity is a collection of performances that reinforce patriarchy. These definitions are simple, of course, and the rest of this section will deconstruct what is meant by collection of performances and by white supremacy and patriarchy.
Whiteness and Masculinity are collections of performances (De Beauvoir, 2009; Du Bois, 1935). Some scholars use the term “social construction” to describe Whiteness (Frankenberg, 2001) and Masculinity (Carrigan, Connell, & Lee, 1985), and that is also an acceptable way of


6
thinking about them. However, this study emphasizes De Beauvoir’s (2009) and Du Bois’ (1935) concepts of gender and race as constant and ongoing collections of performances (Frankenberg, 2001), rather than a construction that once created remains relatively static. The other reason to think of Masculinity and Whiteness as collections of performances is because no individual performance alone can maintain patriarchy and white supremacy, but many performances combined become self-reinforcing (Sherman, 1996). Thus power recreates itself by associating seemingly unrelated performances like supervisorial authority with lighter skin tone, or expertise with golfing and larger shoulders. White men, then, are not born white and men, but made into white men through years of trained and reinforced race and gender performances (De Beauvoir, 2009; Foucault, 1977; Pascoe, 2007; Thandeka, 1999).
The Whiteness and Masculinity collections of performances are individual, organizational, and society wide. Within an individual white man’s mind, the psychology of Whiteness and Masculinity starts at birth. In the first six months, caregivers spend less time making eye contact with babies wearing blue than with babies wearing pink, and are more likely to let babies in masculine colors “cry it out”, and more quick to cuddle babies wearing feminine colors (Eliot, 2009). These findings are particularly significant in a society where the first question commonly asked of a parent is, “Boy or girl?” (Eliot, 2009). Males cry slightly more than females in infancy partially due to the fact that they are, on average, smaller and less fully formed at birth, but by five-years-old males have learned to be boys and are significantly less likely than girls to cry in experimental conditions (Eliot, 2009). By high school many boys showily express callousness and willingness to act violently for fear that they will be shamed as homosexual or inadequately masculine (Pascoe, 2007). Due to ongoing racial segregation in housing and schooling (Edsal & Edsal, 1991; Orfield, Frankenberg, Ee, & Kuscera, 2014),


7
phenotypically white babies are unlikely to encounter a significant number of people of color for the first decade of their lives (Helms, 1990). When white children encounter people of color, Thandeka (1999) argues that their curiosity is often rebuffed by the adults they trust, leading them to avoid speaking about race and feel nervous when the topic comes up (Matias, 2016).
As adults, white men are more likely to have implicit biases that benefit whites and men (Banaji & Greenwald, 2013), while speaking and acting in ways that implicitly and/or explicitly prioritize white and masculine experiences (Bonilla-Silva, 2012; Cooper, 2000). By adulthood, men are less likely to express emotion of any kind, except anger (Bancroft, 2002; Katz, 2013). Whites too are less likely to show emotion, except for happiness, especially in the workplace (Roediger, 1999). Being a white American means having an upper-middle class Midwest suburban demeanor, accent, and appearance including dress and lighter skin tone (small-w whiteness) (Carbado & Gulati, 2013). Men similarly are expected to be less emotional, more rational, more focused on work than family, and have physical attributes including wearing pants and suits rather than skirts and dresses (small-m masculinity) (Burris, 1996; Glauber, 2008; Pascoe, 2007). These prescriptions of dress, behavior, demeanor, and psychology are part of collections of performances that ultimately sum up to Masculinity and Whiteness in individuals.
These individual performances happen within organizations that are also performing Whiteness and Masculinity. The level of analysis that is of primary concern to this study is the organization. The literature on Masculinity in organizations exposes many aspects of management and bureaucracy as innately masculine and even patriarchic (D. L. Collinson & Hearn, 1996). Multiple Masculinity (MM) organizational theorists cite Kanter (1977) as their intellectual foremother, who finds that men maintain power without having to act individually, but through organizational culture and structures. As a recent example, Henderikse el al. (2017)


8
find that organizational culture is more important for achieving gender parity in organizational staffing than even human resources policies. Organizational cultures reinforce Masculinity when creating social opportunities focused around masculine recreational activities (Kanter, 1977) like golf (Cabrera, 2016) or entertaining investors at a gentlemen’s club (Weiss, 2014). Kanter (1977) points out that secretaries, who are mostly women, are rewarded for their fealty to their bosses who are mostly men and discouraged from maneuvering to protect themselves or seek advancement. This division continues in the present day with a new polarization of gendered occupations (Burris, 1996), to the extent that men are unwilling to train for jobs considered feminine even in an economy where they are more plentiful than traditionally masculine jobs (Miller, 2017). Interestingly, men of color are far more likely than white men to take on those traditionally pink collar jobs (Miller, 2017).
Like Masculinity, Whiteness is a workplace norm in many if not most organizations. Whiteness traits such as American mid-west, suburban speech, behavior, dress, and mannerisms are norms in most workplaces (Foldy & Buckley, 2014; Lipsitz, 2009). Any transgression from these established workplace norms by people of color coworkers is viewed as deviant. Carbado and Gulati (2013) give the example of co-workers sharing stories about watching the sunrise after a long night in the office. White employees are assumed to be dedicated and hardworking, while people of color employees must defend themselves against accusations of laziness, bad planning, or even throwing a party in the office (Carbado & Gulati, 2013). Another example of this is Ali, Yamada, and Mahmood’s (2015) finding that Muslim American women feel less job satisfaction when they wear hijab than when they do not, even after accounting for religiosity. Directly addressing the organization as a moderator of the impact of race, Kochan et al. (2003) find that most of the effects of racial diversity found in the business literature are moderated by


9
organizational factors such as culture, communication, and job satisfaction. Individuals perform Whiteness and Masculinity as soloists, and simultaneously for and with organizations as big bands.
Society also performs constantly evolving Masculinity and Whiteness. Sociological literature refers to these collections of performances as institutions (W. R. Scott, 2005). These performance collections are similar to critical concepts of the hegemonic discourse (Foucault, 1977), and Mills’ (1997) racial contract. For all members of American society assumptions that Whiteness and Masculinity performances are collected in certain configurations guide individual and organizational actions. Much of the CWS and MM literature is from this perspective. CWS scholars find that Whiteness slowly evolved from the 1600s through a series of legal, economic, and social movements (Roediger, 1999). For example, the Serviceman’s Readjustment Act (1944), known as the GI Bill of Rights, provided benefits to put veterans of World War II back to work, giving ethnic whites, like Jews and Italians who were substantially more likely to receive honorable discharges than blacks, the leg up that they needed to move from a pseudo-Whiteness to full participation in the dominant racial category (Brodkin, 1994). More recently, Obama’s presidency and color-blindness combined to reproduce Whiteness creating optimal circumstances for a backlash against people of color (Bonilla-Silva & Dietrich, 2011).
Meanwhile, American Masculinity evolves to protect the power of men by polarizing the workforce into ‘expert’ professions, and feminine ‘non-expert’ professions (Burris, 1996). As business leaders extol fluid organizational structures for technical experts of software engineers who are mostly men (Cooper, 2000), the mostly women clerical and frontline staff are subject to higher levels of micromanagement with the advent of technologies that make Bentham’s Panopticon seem unobtrusive (Burris, 1996). This new technocratic Masculinity (Burris, 1996)


10
or ‘nerd Masculinity’ (Cooper, 2000) replaces older performances of Godliness, physical strength, and rational behavior to establish power, all traits associated with men.
White supremacy and patriarchy are the allocation of power to people who most convincingly perform Whiteness and Masculinity respectively. Power refers to the directing of resources by controlling the behavior of others (Foucault, 1977; Marx, 2004). The power of Whiteness and Masculinity is often a conservative power to restrict understanding and discussion of race and gender oppression structures, and therefore stop progress before it starts (Bachrach & Baratz, 1962; Lindstead, 2000; Mills, 2007). This conservative power is how segregation became the norm in the South over the course of many decades of so-called "separate and equal" jurisprudence (Du Bois, 1980). More than simply maintaining the status quo, this conservative power slowly erodes dissent as dominant players find ways to recreate the status quo to benefit themselves (D. Bell, 1980).
When conservative power fails, some turn to aggressive power to maintain the position of the powerful (Foucault, 1977). This happened when the Brown v. Board of Education decision brought a serious critique of the separate but equal doctrine to the center of American politics (D. Bell, 1980). No longer able to maintain the status quo, many whites turned to an aggressive response including church bombings, murders, and brutal policing of blacks in cities across the country (D. Bell, 1980). These two faces of power (Bachrach & Baratz, 1962) keep people from deviating from Whiteness and Masculinity through conservative discipline, while providing an aggressive punishment when deviance does occur (Foucault, 1977). These forces help maintain advantages for men and whites.
Some may object to the terms white supremacy and patriarchy because they are linked to anachronistic bigotry. This study chooses these terms for two reasons. The first is that historical


11
bigotry did not disappear to be replaced by the modern allocation of power to benefit whites and men. Modem white supremacy and patriarchy is a continuation of an American history of oppression (Bonilla-Silva, 2012; Stivers, 1993). The second reason is that the terms accurately depict an ongoing empirical reality where men and whites continue to dominate most social, political, and religious life in America (Lipsitz, 2009; Status of women in the States, 2016).
Whiteness and Masculinity work in tandem to allocate power to white men. Crenshaw (1991) coins the term intersectionality to describe how race and gender intersect to reproduce racism and sexism against women of color who are victims of battery and rape. Referring to a similar concept, hooks (1984) argues that feminist researchers have struggled because “they do not understand fully the inter-relatedness of sex, race, and class oppression or refuse to take this inter-relatedness seriously” (p. 14). These studies show that Whiteness and Masculinity interact in complicated ways that may contribute to one or both concepts. To date, little is known about how Whiteness and Masculinity interact and conflict.
Whiteness and Masculinity Research in Public Administration
Most of the research around race and gender in public administration focuses attention on those who are harmed by Whiteness and Masculinity, rather than the phenomena doing the harm. Social equity (Gooden, 2014), representative bureaucracy (Selden, 1997), and diversity management (Riccucci, 2002) studies mostly highlight the problems that arise when women and people of color interact with public administration either as bureaucrats in or clients of the system. These insights are invaluable for casting light on race and gender in public administration. This study’s research on Whiteness and Masculinity can complement these programs by exposing the mechanisms of power that create discriminatory outcomes. This study also positions itself within institutional research on the power of society and organizations to


12
control the choices of individuals (W. R. Scott, 2005), and makes contributions to that literature by detailing specific performances that reproduce power.
Social equity. Research on Masculinity and Whiteness potentially influences a great deal of social equity research. Social equity is defined as a more flexible measure than equality, “allowing for equivalency while not demanding exact sameness” (Guy & Mccandless, 2012, p. S5), and understanding equality in a context of previous injustice. Gooden (2015) argues the current era is the fifth “epoch” (p. 211) of social equity research, and Whiteness and Masculinity research significantly informs four out of five of these epochs.
The first two epochs ask the questions “What is the context for equality?” and “Who is ‘we’?” (Gooden, 2015, p. 211). If, as is argued in this study, Whiteness and Masculinity influences every aspect of life and government in America, both concepts are an important part of the context for equality. Masculinity and Whiteness research can contribute to understanding the context of social equality as masculine and white. While Gooden (2015, p. 211) argues that the context for social equity is both “formal and deeply entrenched”(p. 211), Kanter (1977) and Ferguson (1984) find that the role of masculinity in organizations is just as deeply entrenched and both formal and informal. Harris (1993) finds that Whiteness is enforced partially through a set of formally recognized property rights that undermine race equality, while Roediger (1999), Brodkin (1994), and Ignatiev (1995) document the American history of Whiteness through the evolving definition of who is white.
Similarly, understanding how Whiteness and Masculinity reproduce power by separating “we” from “they” illuminates what social equity scholars mean by we. In particular, Gooden (2015) argues that public administrators themselves, and those who research public administration, seek to better understand who “we” are. This task requires understanding the


13
racial and gender identities of whites and men who make up an over-representative portion of public administration managers and researchers.
Whiteness and Masculinity research has some catching up to do when it comes to understanding the context and identity of social equity scholarship on the cutting edge of Gooden’s (2015) final two epochs. Gooden’s (2015) fourth epoch seeks to understand why inequity persists. Of all the big questions of social equity, this is the one that Masculinity and Whiteness studies address most directly. As defined above, Whiteness and Masculinity are collections of performances that reproduce and maintain white supremacy and patriarchy (D. Collinson & Hearn, 2003; Ignatiev, 1997). Inequity persists because we have yet to dismantle social institutions like Whiteness and Masculinity. This leads directly to Gooden’s (2015) contemporary epoch which looks at how accountability for social equity can be achieved. This study addresses this by identifying the individual performances that maintain and reproduce white supremacy and patriarchy, thereby teeing up future research and experimentation on developing accountability mechanisms for these performances.
A rich understanding of Whiteness and Masculinity in public organizations contributes to the social equity research agenda by overcoming nervousness to reveal how the personal lives of administrators in their homes, their childhood, and their relationships contributes to the maintenance of social inequity. Referring to race as a “nervous area of government,” Gooden (2014, p. 4) argues “.. .evaluation is unlikely to occur in a serious way if organizations are fundamentally too uncomfortable to directly engage in the topic.” As described above, this nervousness can be understood as a mechanism of power that maintains the racial epistemological contract (Bachrach & Baratz, 1962; Mills, 1997). Martin (2000) argues that the contract extends to gender in the form of taboos that reinforce patriarchy in organizations.


14
Understanding nervousness as a conservative power mechanism indicates that nervousness will be more limiting around discussions of Whiteness and Masculinity, because to maintain power these collections of performances must continue to appear as natural and inevitable consequences of small-w whiteness and small-m masculinity (Burris, 1996; Marx, 2004; Mills, 1997). Public administrators learn nervousness in their private lives from their closest friends and family, even when they do not condone prejudice (Bonilla-Silva, 2015). Therefore, changing management, leadership, and the organization means focusing on the private life as a factor impacting public administration. This dissertation is an attempt to “directly engage in the topic” (Gooden, 2014, p. 4), taking nervousness head on by identifying what Whiteness and Masculinity performances reproduce inequality, and how to observe those performances in public organizations.
Representative bureaucracy and diversity management Like this dissertation, much of representative bureaucracy and diversity public management research assumes that the personal lives of public administrators, namely their race, gender, and other identities, influence how services are provided. Representative bureaucracy assumes that government operates differently when public administrators demographically reflect the people they serve (Selden, 1997). One thing that representative bureaucracy gains from research of Whiteness and Masculinity is a path to pragmatic solutions short of replacing white and man bureaucrats with people of color and women. This is an intractable problem for representative bureaucracy scholars ever since the original texts lamented that public administrators could never fully represent the poor and unemployed because they, by definition, have jobs and income (Kingsley, 1944). While the modern era has witnessed a dramatic increase in the diversity of public administration, representativeness is still a long way off (Guy & Newman, 2005). Representative bureaucracy findings provide few tools for public managers looking for solutions short of


15
replacing large portions of their white men employees with women and people of color. This is particularly troubling in light of a recent finding that human resources policies had no impact on creating gender equality after controlling for factors like organizational culture, diversity knowledge, and leadership (Henderikse et al., 2017). Research on Whiteness and Masculinity can bridge the gap between representative bureaucracy and diversity management by providing solutions related to the performances that create the discriminatory outcomes discovered by representative bureaucracy researchers.
Whiteness and Masculinity research illuminates how seemingly neutral behaviors in organizations have unintended discriminatory outcomes. Researchers find that work hours can reproduce white supremacy (Carbado & Gulati, 2013) and patriarchy (Cooper, 2000), staffing choices are polarized by race and gender (Burris, 1996), and fluid organizational structures help maintain power in upper-level managers still dominated by white men (Burris, 1996; Cooper, 2000). The causes of discrimination in many organizations are less often bigotry and more often unconscious bias (Banaji & Greenwald, 2013) or unintended consequences (Yancy, 2008). While researchers focus on managing diversity in the workplace (Riccucci, 2002), these performances are found to be more common in workforces without any diversity whatsoever (Carter, Helms, & Juby, 2004). For diversity management and representative bureaucracy, Whiteness and Masculinity are levers for understanding and changing what is happening in the organizations that the increasingly diverse workforce is entering (Riccucci, 1997), thereby addressing the question of how to recruit diversity into organizations that are not yet diverse.
Another contribution to representative bureaucracy is Whiteness and Masculinity studies’ help making sense of passive, active, and symbolic representation. Representative bureaucracy scholars find that simply hiring people of color and women is not necessarily sufficient to create


16
representative outcomes (Riccucci & Van Ryzin, 2017). The act of including people of color and women in the bureaucracy is called passive representation, while outcomes that are less discriminatory signal that there is some kind of active representation (Riccucci & Van Ryzin, 2017). If simply seeing a people of color or a woman in a role is enough to help a beneficiary trust the bureaucracy, that is called symbolic bureaucracy (Van Ryzin, Riccucci, & Li, 2016). Conceptualizing race and gender as performances, as do Du Bois (1903) and De Beauvoir (2009) simplifies some of the differences between passive, active, and symbolic representation. If organizations are normally performing in ways that conform to Whiteness and Masculinity, when people of color or women enter such an organization their performances will change the way the organization operates and simultaneously conform to the organization. However, this passive representation will not translate as well to active representation if the new people of color and women organizational members are trained to perform the Whiteness and Masculinity dominant in most organizations.
The conceptualization of race and gender as collections of performances is a challenge to the idea that representation can ever be truly passive. A basic assumption of this study is that race and gender are not firm demographic identities, but performances. Based on this assumption, scholars find that people of color and women perform Whiteness and Masculinity as a strategic attempt to get ahead, and/or as a result of oppression (Hanisch, 2006; Neville, Coleman, Falconer, & Holmes, 2005). When people of color and women enter organizations performing Whiteness and Masculinity, they perform race and gender, but not necessarily the race and gender researchers assume they will perform. For example, Carbado and Gulati (2013) discuss the pressure on people of color to conform to displays of Whiteness in the workplace, while Martin (1990) and Cooper (2000) find that women struggle to conform to Masculinity.


17
What has been called passive representation, then, is the active representation of Whiteness and Masculinity. But this is not necessarily bad news for achieving a representative bureaucracy. Just as people of color and women can perform Whiteness and Masculinity, whites and men can potentially dismantle Whiteness and Masculinity when performances are identified and changed. Findings from contemporary sources indicate that many whites and men wish to help end discrimination (NBC News/Wall Street Journal Survey, 2016; Stepler, 2016). CWS and MM research provide tools for acting on that desire.
Institutional public organization theory. By theorizing Whiteness and Masculinity as institutions, this study will contribute to institutional theory by exploring the extent to which organizational environments consist of distinct overlapping institutions. Unpacking that sentence, organizational institutional theory originated from attempts to understand how the world around an organization, the environment in which it sits influences the organization itself (Lawrence & Lorsch, 1967; W. R. Scott, 2001). This resulted in a number of studies finding that organizations are influenced by their environments in myriad ways (e.g. Bozeman, 1987; DiMaggio & Powell, 1983; Meyer & Rowan, 1977; Ostrom, 2005; Roy, 1997; Stivers, 2000; Uzzi, 1999). The foci of those studies were on how changes in rules impact institutions in ways that change many organizations at once (Roy, 1997; Uzzi, 1999), or on how the entire collection of institutions, the institutional arrangement taken as a whole operated on a small subset of organizations (Ostrom, 2005). Looking at Whiteness and Masculinity as distinct but related makes it possible to understand performances as parts of embedded institutions. This understanding enables researchers to identify embedded Whiteness and Masculinity distinctly from one another, and determine better how power is acting on individuals in organizations as compared to other similar organizations.


18
This study also contributes by incorporating insights from CWS and MM research into institutional theory. If Whiteness and Masculinity can be conceptualized as institutions (Sherman, 1996), then CWS, MM, and related researchers have been analyzing these institutions for more than a century. One example of potential contribution is Derrick Bell’s (1980) article that marked the beginning of critical race theory. In that article, D. Bell (1980) describes the decline in school integration from the time of Brown v. Board of Education (1954) through the maintenance of conservative power. This leads D. Bell (1980) to argue that racism is permanent and evolving, reflecting the arguments of institutional scholars who find that institutions maintain power through evolutions that can be traced using historical data (Dobbin, 2005). The insights of CWS and MM researchers into the embeddedness and power of race and gender in organizations and individuals represents a contribution to institutional thinking that stands to inform analyses of other institutions such as capitalism (Weber, 1972) and consumerism (Campbell, 1987). In other words, Whiteness and Masculinity are just two sources of performances that limit individual behavior, and focusing on how these two embed in organizations will help scholars understand how other institutions influence public organizations and other action arenas.
Conclusion: Research on Whiteness and Masculinity in an Institutional Framework
This dissertation’s main argument is that when public administrators go to work, they bring with them Whiteness and Masculinity from the most personal sectors of their lives including unconscious biases learned in the cradle, the classroom, and intimate relationships. This argument has the potential to impact representative bureaucracy research, social equity theory, diversity management, and institutional organizational studies. The rest of this study fills in this argument, starting in chapter two with the development of a theory of Whiteness and


19
Masculinity in public organizations.
The third chapter outlines a method for gathering and analyzing data on Whiteness and Masculinity performances in the federal government, and tracking their impact in organizational structure and culture, in the behavior of managers, and in the actions of street-level bureaucrats. Chapter Four reports the findings of a quantitative study using the Federal Employee Viewpoint Survey (2016) to analyze the impact of Whiteness and Masculinity on whites’, men’s, and white men’s satisfaction and public service motivation. In Chapter Five, the qualitative findings elucidates how white men make sense of their race and gender while performing public service. In the concluding chapter, an institutional map reveals strategies for public servants to manage Masculinity and Whiteness in public organizations.


20
CHAPTER TWO
INSTITUTIONAL WHITENESS AND MASCULINITY: A THEORY OF
INTERSECTIONALITY
Chapter One describes race and gender as personal behaviors initially learned in infancy when individuals are identified as “boy” and/or “white.” This chapter starts from the opposite pole by arguing that gender and race are global geological forces that originate in the cretaceous period when dinosaurs roamed the shores of North America. Some of those shores were covered in plankton that floated in the coast off what is now the southern United States (McClain, 2012). These plankton left deep rich soils in a crescent from the coast of Virginia, through Atlanta to Bolivar County on the banks of the Mississippi. 65 million years later, people from Africa were kidnapped and forced to cultivate cotton grown in the rich soils the plankton left behind. In 2016, Donald Trump’s electoral sweep of the deep south was qualified by this same crescent of counties along that ancient coastline where black voters who descended from those enslaved people voted for the democratic candidate (CNN, 2016). This example illustrates how the personal is impacted by large forces (Roberts, 2014), and also how factors like race are fecund variables for explaining large-scale human behavior.
Gender, race, and public organizations are best understood holistically. “In a holistic view it is impossible to begin from either psychology or economics and explain everything in a one-way causation. Instead, the relations of society must be seen as a whole” (Sherman, 1996). Said another way, race and gender emanate simultaneously from microscopy of a single individual, and from the panorama of society. Performances at all levels influence and are influenced by rules, norms, and cultures that dictate meaning for social life, what sociologists call institutions (W. R. Scott, 2001). These institutions create a framework that affects every


21
aspect of public organizational life, and individuals in public organizations. Whiteness and Masculinity are two such institutions. To analyze these institutions, this study will investigate two related secondary research questions. (1) What effects do Whiteness and Masculinity performances have on public employees’ work life? (2) How do public organizations, public managers, and public administrators perform Whiteness and Masculinity?
To answer these research questions, this chapter develops a theory of Masculinity and Whiteness in public organizations (see Figure 2.1/ The first section develops the institutional framework of which Masculinity and Whiteness are two institutions. The next section substantiates a number of performances that are influenced by Whiteness and Masculinity at the society, individual, and organizational levels, laying out propositions for study in the process. At the end of that section, the literature on public service motivation and job satisfaction establishes dependent variables for quantitative testing of the impact of Masculinity and Whiteness in public organizations. The last section of the chapter is a short conclusion.
White Men in the Institutional Framework The Institutional Framework
Whiteness and Masculinity are two institutions in the institutional framework surrounding public organizations. Institutions are collections of rules, norms, and cultures that dictate meaning for social life (W. R. Scott, 2001). Rules refer to formal rules that are passed through deliberative bodies or carry the force of law (Ostrom, 2005; W. R. Scott, 2001). Norms are shared understanding about the best or most correct way of doing things (Meyer & Rowan, 1977; Ostrom, 2000; Rousseau, 1896). Norms include ethical and moral beliefs that are shared by a society, such as the protestant ethic valuing hard work (Weber, 2004) or the romantic ethic valuing beauty and aesthetics (Campbell, 1987). Norms also include what Ostrom (1990, p. 72)


22
Organizational Individual
Society Level Uve| Level
Figure 2.7: Model of institutional effects of Whiteness and Masculinity on public organizations


23
calls “rules-in-use,” informal rules that control behavior, but with no formal recognition. Together, norms and rules make claims about how individuals should act when they are sharing resources in a society.
Culture is a lens for making sense of the world in a particular time and place (LeCompte & Schensul, 2010). Weick (1995) calls culture a synonym for paradigm (p. 120), an interpretation of reality that helps diverse people share their understanding of the world. Thus, culture is a kind of Rosetta stone by which individuals understand one another’s beliefs through shared artifacts. Just as the Rosetta stone includes three translations of a unique text, culture is the expression of unique values through multiple stories, symbols, statements of belief, holidays, media, and other experiences where things, actions, and words (performances) take on emotional meaning (Schein, 2010). When these performances take on emotional meaning, they have value so organizations work to guard them (Weick, 1995). Culture, rules, and norms make up the institutions that hold organizations and societies together (W. R. Scott, 2001).
Whiteness and Masculinity are two institutions that operate in an institutional framework surrounding public organizations (see Figure 2.2) (Sherman, 1996). Other institutions in the framework include capitalism, bureaucracy (Weber, 1972), consumerism (Campbell, 1987), technocracy (Burris, 1993), Blackness (Fanon, 1963), Femininity (Bartky, 1990), and many others. An institution embeds in an organization when a performance from that institution occurs in the organization (Granovetter, 1985). For example, the institution of consumerism includes a norm for aesthetic beauty that may be performed by constructing a sleek open floorplan in a computer store, or an eclectic treasure hunt in an antique shop (Campbell, 1987). These aesthetics are performances of consumerism that are embedded as artifacts in the computer store and the antique shop. At first, these performances appear disparate, but when taken in their


24
Institutional Framework
Figure 2.2. Masculinity and Whiteness in an institutional framework.


25
institutional context they can both be understood as bound together by the institution of consumerism (Campbell, 1987).
Institutions and organizations change. Similar to Darwinian adaptation, organizations inherit institutional traits or performances (Blyth, Hodgson, Lewis, & Steinmo, 2011; Fiirstenberg, 2016). For example, managers working in corner offices on higher floors is a performance of the institution of capitalism (Schein, 2010). However, adhocracy is replacing bureaucracy as an overlapping institution in many modem capitalist organizations (Burris, 1993; Mintzberg, 1983). Adhocracy is defined by symbols of equality (Mintzberg, 1979), at least among expert classes (Burris, 1996). This emphasis on equality has led to the design of capitalist offices that no longer adhere to previous spatial norms (Cooper, 2000; Schein, 2010). Assigning corner offices to managers is still a performance of capitalism, but so too is a newer open floorplan (De Croon, Sluiter, Kuijer, & Frings-Dresen, 2005). If, over time, the new open floorplans become more common than cellular floorplans with reserved corner offices, that would be an example of the institution of capitalism evolving (Fiirstenberg, 2016). Like capitalism, Whiteness and Masculinity evolve (Brodkin, 1994; Ignatiev, 1995; Roediger, 1999). Importantly, institutional evolution is often circular and regressive (Davis, Diekmann, & Tinsley, 1994; DiMaggio & Powell, 1983; Fiirstenberg, 2016; Sherman, 1996). Many scholars have argued that current institutional trends in public organization are both circular (e.g. Goodsell, 1983; Light, 1997; Lynn Jr., 2006; O’Leary & Vij, 2012; J. Q. Wilson, 1989) and regressive (Eikenberry, 2009; Ingraham, 1993; B. Sandberg, 2016; Stivers, 2008).
Public Organizations
Public organizations range a wide spectrum from socially responsible business (Bartley, Koos, Samel, Setrini, & Summers, 2015) and social enterprise (Kerlin, 2006), through


26
community nonprofit organizations (Never & Westberg, 2016) and social welfare nonprofits (Romzek & Johnston, 2005), to traditional government agencies (Weber, 1972). What unites these organizations as being public is that they have public inputs and/or public outputs.
The input that delineates public from private organizations is political authority (Bozeman, 1987), while Moulton calls the public organizational output realized public value (Moulton, 2009). While scholars point out that both political authority and realized public value can be outputs as well as inputs, thinking of them this way helps clarify the definition of publicness. Bozeman (1987) separates organizational inputs into economic and political authority that are used to create goods and services. Economic authority is the ability to barter and trade for resources, while political authority is legitimacy brought to an organization by popular or democratic means (Lindblom, 1977). One recent example of the use of political authority is socially responsible businesses’ use of ethical credentials to motivate customers to buy their products (Bartley et al., 2015). A company like Patagonia uses their ethically sourced manufacturing model to encourage customers to choose their products over other products that may be less expensive (Bartley et al., 2015). Therefore, a socially responsible business is able to convert a political input into added economic output. Patagonia’s realized economic value is the outdoor clothing they manufacture and their profit, and the realized publicness is the high paying jobs and reduced environmental impact. Moulton (2009) argues that the realized publicness output is as important as political authority in changing the character of an organization along the spectrum from most private to most public. What is less known is how these public inputs and outputs impact race and gender performances within individuals associated with public organizations along the spectrum.
Public organizations are particularly important contributors to Whiteness and Masculinity


27
even as they are constrained by Whiteness and Masculinity. As described in further detail below, public organizations created modern Whiteness, and reinforce both Whiteness and Masculinity through laws, regulations, and the distribution of services based on the social construction of individual citizens and beneficiaries (Heckler, 2017; Mackinnon, 1989; Roediger, 1999; Schneider & Ingram, 1993). Meanwhile, as described above, white men have outsized influence on almost every aspect of public administration from the founding of the country up to today (Witt, 2006). As public organizations stretch further into the private sphere, the influence of Whiteness and Masculinity grows because the private sector has a larger proportion of white men, and white men dominate the executive apexes of private corporations, especially in Fortune 500 companies that are more likely to be run by white men today than in recent decades (Berman, 2015; Guy & Newman, 2005; Lindzon, 2016). Perhaps most importantly, public organizations are uniquely positioned to affect social equity (Gooden, 2014), and Whiteness and Masculinity constrain efforts of equity-minded public servants.
Emotionally Bounded Individuals
The more emotional investment culture can create in the performances valued by an institution, the more power the institution has over individuals within it (Weick, 1995). Because values are created by emotional investment, the bounded emotionality model of the individual proposed by Mumby and Putnam (1992) provides a useful concept for making sense of how individuals interact with institutions. Emotionally bounded individuals are driven initially by their relationships to others and their emotional experiences. “Bounded, then, refers to an individual being able to recognize another person’s subjectivity, a state that is necessary for producing understanding or interrelatedness” (Mumby & Putnam, 1992, p. 474). This emotionality is bounded by “intersubjective limitations” (Mumby & Putnam, 1992, p. 474),


28
which can be understood as the pressures placed on the individual by relationships including institutions, organizations, friendships, and familial ties. The boundedly emotional individual uses their “physical, mental and emotional resources” (Mumby & Putnam, 1992, p. 475) to create shared experiences and share their values with others. Thus, the balance between individual values and group values is not solely a self-serving calculus, but a question of giving what they are able while at the same time investing in something bigger than themselves (Mumby & Putnam, 1992).
Bounded emotionality recognizes as problematic the basic psychological assumptions of many of the bounded rationality theories of public administration. Before the boundedly rational actor can select a method of achieving goals and objectives based on the best available information (B. D. Jones, 2003), the goals and objectives themselves have already been set by emotionality in an institutional framework (Mumby & Putnam, 1992). These values are determined by boundedly emotional individuals influenced by institutions like Whiteness and Masculinity.
Whiteness and Masculinity
Whiteness and Masculinity are the institutions that maintain and reinforce white supremacy and patriarchy. Not all white men are the same, but Whiteness and Masculinity have institutional effects that can be observed through the regularity of white and masculine performances in which they engage. Multiple Masculinities (MM) scholars point out that men are unique and stress that scholars should avoid essentializing (Boyle, 2002), and Critical Whiteness Studies scholars argue the same is true of whites (Allen, 2008). Thus, white men are not homogenous, but Whiteness and Masculinity are hegemonic institutions that influence white men and others (Connell & Messerschmidt, 2005). These ideals are hegemonic because they are


29
both unattainable and aggressively enforced (Pascoe, 2007; Thandeka, 1999; Wise, 2011). All men and white people perform their Masculinity and Whiteness to different degrees along a spectrum, facing ostracism, and threats of violence for those men and whites who do not adequately perform (Allen, 2008; Ignatiev & Garvey, 1996; Majors & Billson, 1992; Pascoe, 2007).
Whiteness and Masculinity are almost always impossible to attain, and literature documents how feelings of inadequacy related to this catch-22 become rage, shame, violence, and tears (e.g. Andersen, 2016; Bancroft, 2002; Bonilla-Silva, 2003b; Cooper, 2000; Katz, 2013; Matias, 2016; Thandeka, 1999). One dramatic example is Case and Deaton’s (2017) finding that self-induced mortality is increasing among white men due to drugs, alcohol, and suicide. While economists like Case and Deaton (2015, 2017) cite Murray (2012) to argue that economic inequality is the primary driver of white self-harm, their own research acknowledges that all other racial (except Asian) and gender groups earn less and have lower workforce participation, but yet are not seeing the same increases in self-induced death rates. The explanation preferred by critical race and gender theorists is that the ideals of Whiteness and Masculinity are so unattainable and aggressively policed that white men who fail to approximate them suffer from their allegiance to their race and gender (Allen, 2008; Majors & Billson, 1992). Where Hermstein and Murray (1994) argue financial hardship is the result of cultural inferiority combined with biological “dysgenic pressures” (see also Murray, 2012; Vance, 2016), critical race and gender scholars argue that the concepts of superiority and inferiority themselves require redefinition (Butler, 1990; Fanon, 1963). Given the power of this hegemony over white men, understanding behaviors in individuals requires understanding system-level performances.


30
Society-level Whiteness and Masculinity
Society-level performances of Whiteness and Masculinity fall into three main categories; economic, cultural, and political (see Table 2.1). Starting with the political, Du Bois (1903) argues that Congress granted the franchise to black men in an effort to maintain moral authority without having to continue the expensive Freedman’s Bureau programs that would have created substantially greater racial equality. In the latter half of the 20th Century, the Republican party became the home for whites who resisted black and brown demands for equality (Edsal & Edsal, 1991). As a result, Edsal and Edsal (1991) find that whites vote consistently as a block in a way that indicates an aversion to black and brown people. Continuing this trend, white men and women voted against voters of color (and Jews) in every Presidential election so far this century (CNN, 2004, 2008, 2012, 2016; Kuhn, 2009).
Masculinity too performs in the electorate at the society level. Men make up over 75 percent of state legislatures, and higher proportions of U.S. Congress, Senate, governors and mayors (Center for American Women and Politics, 2017). Over 94 percent of state legislators are white, men, or both. Just as black suffrage was enacted as a concession to Whiteness (Du Bois, 1903), women’s suffrage was granted as a concession to Masculinity. Despite the aims of women suffragists who started the Women’s Peace Party to agitate for peace in World War I, Woodrow Wilson introduced women’s suffrage as part of the war effort (Goldstein, 2001). Research finds that men are still more “pro-force” (Silverman, 1987, p. 190) in how they wish to achieve foreign policy objectives. Today, the gap between how men and women vote is larger than ever, but not as large as the racial voting gap (CNN, 2016). Both gaps are important for public organizations because politicians selected by these racialized and gendered voting processes set the tone for public service, and create laws that are artifacts of Whiteness and Masculinity.


Table 2.1.
Society-level performances of Masculinity and Whiteness.
Media Whites and men are portrayed as noble and innocent.
Stereotype Whites and men employees are assumed to conform to characteristics stereotypically associated with whites and men.
Norms Norms that protect Whiteness and Masculinity.
Noblesse The claim that men are obligated to protect innocents by asserting power over them.
Innocence The claim that whites or white culture is more morally justifiable.
Wealth Whites have more per capita wealth than all other racial groups.
Income Men earn more money than women, and whites earn more than most other racial groups.
Sacrificial Zones Whites are less likely to live in areas with environmental hazards.
Housing Whites live in neighborhoods that are mostly white.
Schooling Whites are more likely to go to schools that are overwhelmingly white.
Capitalism The belief that wealth, income, and job advantages of white men are the result of merit.
Consumerism The belief that a person is more interesting and desirable to be around if that person owns interesting and desirable things.
Representation Whites and men are more often elected officials.
Laws Laws that protect Whiteness and Masculinity.
Example
A federal employee admires a white man character from a movie that saves a woman by inflicting violence on men of color.
A white man from Colorado is asked about the best camping spots near Washington, D.C. before a woman of color who grew up in the region.
A white man hunter displays a gun in his car window without fear that he will be stopped by police.
Justifying restricting contraception by asserting the need to protect babies; justifying xenophobia by asserting the need to stop "rapists."
Justifying restricting contraception by asserting the superiority of the Christian moral code; justifying xenophobia by asserting the need to stop "radical Islam."
A white employee's grandparents retire and are "granny nanny."
Jobs filled by white men employees pay more than jobs with more women and employees of color.
A white employee lives in a neighborhood with many parks for his
children.
White employees live near each other and carpool in everyday with the
white boss.
A white employee went to school with the boss who is also white.
Employees assume that a white man's promotion is exclusively the result of that persons intelligence and talent.
The boss wants to visit the house of a white man who owns a classic car
and some land.
94% of state legislators are whites, men, or both.
A white man uses the 14th Amendment to argue that Affirmative Action is unconstitutional.


32
The legal structure of the United States was originally based on assumptions that white men were full citizens, women existed in relation to them, blacks were their property, and all other races were their adversaries (Harris, 1993; Scales, 2006). Legal scholars argue that Masculinity continues to be maintained through laws that benefit men, such as MacKinnon’s (1989) scathing critique of how rape laws contribute to a culture of sexual abuse against women. Similarly, Scales (2013) argues that gender-blind Masculinity prevents open debate of militarism, and contributes to the ever-increasing power of the President.
Whiteness was originally a legal construction with the introduction of laws classifying the children of black women slaves as chattel, therefore creating the original multi generational race-based slavery in the Virginia Colony (Roediger, 1999). In modem law, Harris (1993) documents how Whiteness itself is treated as property through legal color-blindness. Even in the case of affirmative action in university settings American law forbids race consciousness when used to right historical wrongs, but permits it when used for diversity that benefits white students (Harris, 1993; Regents of the Univ. of Calif, v. Bakke., 1978). Similarly, Lipsitz (2009) documents the process through which white homeowners benefit as a result of color-blindness in federal policymaking in the 20th century. Even today, color-blindness in a housing policy promulgated by the Obama administration is likely to benefit mostly white homeowners in still-segregated housing communities (Heckler, 2017). Like elected officials, laws dictate the daily life of public administrators (Bertelli & Lynn, Jr., 2006), and these critiques should make public administration scholars cautious about using law to provide foundations for the field that are neither racist nor sexist (Rohr, 1990; Rosenbloom & Piotrowski, 2005).
Even as political and legal performances continue, society creates and propagates a culture of Whiteness and Masculinity. Recall that culture is the process of creating meaning and


33
value (Weick, 1995). Media and art create meaning around Whiteness and Masculinity (Katz, 2013; Vera & Gordon, 2003), such as common stereotypes that have been well-documented in the psychological literature, and that will be described further in the section on white and masculine psychology below.
In addition to creating and reinforcing stereotypes, media communicate images of men as having a duty to unilateral action and strength by force (Gilligan, 1996, 2003). Gilligan (2003, p. 1167) explains, “In all cultures and all eras of history, most homicides, suicides, wars, and even
so called unintentional injuries and deaths..are committed by, and suffered by men.” He
argues that the desire for masculine noblesse oblige, and shame at not achieving it, is the fundamental driving force behind this call to violence. Katz (2013) and Scales (2009, 2013) similarly find that media and law create rhetorical connections between Masculinity, militarism, and noblesse. These media combine masculine noblesse with methods of organization common in the military, such as technocratic pragmatism (Burris, 1996; Ferguson, 1984) and top-down bureaucracy (Katz, 2013; Scales, 2013). Stivers (1993) argues these emphases on top-down leadership and deference to expertise are more congruent with the societal “image” of men than that of women, and finds this places women public servants in dilemmas for which there are no easy solutions.
If Masculinity values noblesse, Whiteness culture inflates the value of innocence. To Mills (1997), the racial moral contract originates with the connection between lowercase-w whiteness and innocence through the labeling of geography with moral value. Thus, Africa is constructed as the “dark continent” occupied with people who do not share European innocence (Mills, 1997, p. 50). This labelling continues through racialized geographical titles like the foreignness of the barrio, the desperation of the ghetto, and innocence of the segregated suburbs


34
(Leonardo & Hunter, 2009). Baldwin (1962) refers to this innocence when he laments that it is not racism itself, but the claim of innocence that “constitutes the crime” (p. 15). This white innocence contributes to Whiteness in public service. The image of the well-meaning white teacher naively but resourcefully rescuing classrooms full of black and brown students is a common cultural construct (Matias, 2013; Solorzano, 1997). This storyline has profound impact on the jobs of teachers who come to be trusted by their students, putting them in a position to reinforce links between innocence and Whiteness for their students (Leonardo & Boas, 2013).
The outsized value of innocence and noblesse for white men combine with laws, rules, and media to create Whiteness and Masculinity as valuable property creating wealth, income, health, and other outcomes for whites and men (Harris, 1993; Mackinnon, 1989). For example white people are more likely to go to a school that is less than 25 percent black or brown than at any time since 1968 (Orfield et al., 2014), and housing continues to be racially segregated (Lipsitz, 2009), supporting Edsal and Edsal’s (1991) finding that, like voting, white housing and schooling choices are partially motivated by racial aversion. Men continue to earn greater income than women, and whites earn more than all other racial groups except Asian Americans (B.L.S., 2017). Meanwhile, the white wealth gap continues unabated as whites have more average wealth than all other groups including Asian Americans (Oliver & Shapiro, 2006). Whites have longer lifespans than brown and black people (Case & Deaton, 2017), while studies show that shorter lifespans are positively correlated with masculine behavior even among females (Gilligan, 2003; Lippa, Martin, & Friedman, 2000). One contributor to lower white mortality is that whites are significantly less likely to live in areas that have environmental hazards, and race is a stronger predictor of these “sacrificial zones” than poverty (Lipsitz, 2009). Taken together, Masculinity and Whiteness have profound economic impacts.


35
Very little has been written about the intersection between Whiteness and Masculinity at the society-level, but some guidance can be gleaned from analysis of other intersections. For example, critical race scholars document that white women teachers resort to crying in efforts to gain sympathy and claim they are the victims of conversations exposing their racism, thereby using feminine vulnerability to enact their Whiteness (Matias, 2016). Meanwhile, Majors and Billson (1992) find that some black men sacrifice heavily to uphold images of masculine noblesse in the face of racial discrimination. Put in the context of Gilligan’s (2003) understanding that noblesse is a defensive mechanism against feelings of shame, the sacrifice black men make for their noblesse can be understood as an attempt to maintain Masculinity despite not having access to Whiteness. For white men, this is likely to mean that feelings of deficit, or shame at failing to attain ideal Whiteness could lead to more aggressive efforts to approximate Masculinity, while shame at lacking Masculinity could lead white men to place more emphasis on racial divisions.
Contemporary commentaries including Vance (2016) and Andersen (2016) document violent aggression among mostly men in white communities resulting from perceived deficits, indicating that the same phenomenon at play among some black men is happening among white men turning to masculine noblesse to compensate for feelings of deficit in white innocence. Less documented is whether white men who have feelings of deficit in masculine noblesse turn to white innocence to compensate. Could this lead white men in public service to position themselves as martyrs in the same way that white women teachers play the victim to silence critiques from people of color (Leonardo & Boas, 2013; Matias, 2016)? Could it create men who use their white innocence to inoculate controlling behaviors toward those over whom they have power in the same way as spousal abusers (Bancroft, 2002)? There is not enough empirical


36
literature on the intersection of Whiteness and Masculinity to venture a guess.
The literature indicates that white men need both adversaries and objects of value, and the intersection of Whiteness and Masculinity provide both. Whiteness perceives the white race as the absence of color, a race that is identified by not being the other (Fanon, 1963). Therefore, Whiteness cannot exist without black, brown, yellow, and red bodies (Fanon, 1963). Masculine noblesse also relies on others. Noblesse requires an adversary to conquer and a feminine prize to win (Katz, 2013; Mackinnon, 1989; Pascoe, 2007). By valuing white people as innocent (Mills, 1997), Whiteness helps white men justify the targeting of people of color, whose imagined “darkness” and lack of innocence makes them deserving of punishment (Mills, 1997; Schneider & Ingram, 1993). Similarly, this situation serves Whiteness by providing objects for white men to protect so they can justify the use of force without disrupting their white innocence. The two institutions then, like so many other overlapping institutions in the framework, are interdependent and cannot be understood without simultaneously considering the effects of both.
Society-level propositions. These institutional performances create pressures for public organizations to conform to Whiteness and Masculinity (DiMaggio & Powell, 1983), and support the study’s first propositions. Public organizations and public servants understand that Masculinity and Whiteness are the societal norm, and they will subtly and overtly make reference to those institutions in their communications and actions to justify their work and guide their behavior.
Societal Proposition #1- Public organizations and public servants refer to society-level Whiteness and Masculinity to resolve conflict.
Another proposition derives from DiMaggio and Powells (1982) concept of isomorphism that organizations in a society must conform to the institutional pressures of the organizations


37
around them to obtain resources.
Societal Proposition #2- Public organizations and public servants refer to society-level Whiteness and Masculinity to maintain their status with other organizations, public servants, and individuals in society.
Individual-level Whiteness and Masculinity
White men train in Whiteness and Masculinity performances from the moment they are born and continue to enact those performances until they are buried under a headstone reading “Beloved Husband and Father.” In public organizations individuals face Masculinity valuing aggressive toughness and technical rationality. Masculinity prescribes a “tough guise” (Katz, 2013), a rejection of so-called feminine emotionality in favor of rationality, calculation, and aggression (see Table 2.2). Katz’s (2013) tough guise and Majors’ and Billson’s (1992) “cool pose” communicate a willingness and ability to act aggressively by showing only emotions that communicate power. Researchers find that this aggressiveness is causally linked to the propensity to act violently and criminally (Bancroft, 2002; Gilligan, 2003; Katz, Heisterkamp, & Fleming, 2011; Klinesmith, Kasser, & McAndrew, 2006). In professional white men, the tough guise mixes with white innocence to become the image of the noble man who hides his personal connections and emotions behind a veil of competence (Kendall, 2000). This “nerd masculinity” (Kendall, 2000) adopts toughness, not through displays of physical aggression, but through all-
night work sessions that redefine “poor planning......as a test of will, a test of manhood for a
team of engineers (men)” (Cooper, 2000, p. 385). These men spending long nights away their families dismiss private-life concerns to avoid workplace taboos, or leverage those concerns to prove their willingness to sacrifice family for work (Cooper, 2000; J. Martin, 1990).


Table 2.2:
Individual-level performances of Whiteness and Masculinity.
Performance Observable
Dress Manner of clothing that resembles 19th century suits and casual attire worn by white men.
Noblesse The claim that men are obligated to protect innocents by asserting power over them.
Innocence The claim that whites or white culture is more justifiable.
Mannerisms Stance, stride, gestures, and posture imitating 1950's white men.
Speech "Proper" grammar and speech associated with the American Midwest.
Discriminatory framing Focusing on avoiding short-term losses in a way that benefits white men.
Intuitive Assessments A statement or artifact indicating a careless judgement that fallaciously validates a stereotype benefitting white men.
Stereotype Promise Blindness Unconsciously or consciously acting to conform with a stereotype about white men. Pretending or aspiring to not notice race and/gender. A statement or artifact that takes attention away from
Displacement discriminatory behaviors by arguing for a different more morally defensible objective.
Microaggressions A statement or artifact asserting power for white men based on group status or identity. A statement or artifact expressing a false belief that white
Minimization men have only minimal advantages, or fewer advantages than historically.
Projection Accusing women and POC of discrimination against white men.
Technocracy Statement or artifact supporting doing things the most efficient way to justify benefits to white men.
Examples
Polos, pant suits, slacks, snug tee-shirts, loafers, dress shoes.
Voting for an unconstitutional travel ban from Muslim nations based on the need to protect innocent Americans from terrorism.
Claiming that racism is a problem of other people; claiming that black culture is a major cause of poverty.
A "powerful" handshake with a twist that reflects power; smooth purposeful gait; a wide sitting posture.
Grammar corrections, complements on "well-spoken" people of color, strong beliefs about the proper use of certain words.
Hiring a white man as the "best candidate for the job" in the short-term despite caring about equality.
Assuming incorrectly that a doctor is a white man; assuming that a woman cannot or does not want to go to a sports bar.
Working harder to prove oneself when perceived to be competing against women and POC.
Claiming not to let race or gender influence decision-making.
Banning hijab and niqab to protect Muslim women. Promoting a mustache competition to include more POC.
A large poster depicting an Indigenous tribal caricature as a sports mascot; "Where are your really from?"
"My ancestors were Irish slaves, and they worked their way out.";
"Women expect men to pay for their meals, but don't want to earn less
than men."
Accusing a woman of color of discrimination against white men.
Heightened concern for white masculine dress, speech, and mannerisms; focus on efficiency to justify racism and sexism.
CtJ
00


39
The tough guise combines with individual-level Whiteness performances to give white men two layers of benefits. White racial performances include suburban mid-west vernacular, dress, and mannerisms (Carbado & Gulati, 2013). The clothing that is considered normal in the workplace more closely resembles clothing and hairstyles that are easily associated with white people than people of color, meaning that professional attire and hair goes unnoticed for white people
(Carbado & Gulati, 2013), in much the same way that the appearance of men is less commentworthy than the appearance of women generally (Dellinger & Williams, 1997; Weitz, 2001). Similarly, professionally recommended mannerisms and dress for women include “leaning in” to conversations in the manner of men, and pantsuits that on men are simply called suits (Chamberlain, 2017; S. Sandberg, 2013).
Masculinity and Whiteness both value hard work, but while Masculinity values aggressive boastful martyrism to express noblesse, Whiteness values agreeable polite gentility to express innocence. Modern American Whiteness evolved out of the move from the traditional European agrarian societies to industrial factories (Roediger, 1999). 19th Century European Americans had advantages over blacks even when they first set foot in the United States (Guglielmo, 2003), and each wave of immigrants achieved full Whiteness by participating in the economics of the newly forming spirit of capitalism and the industrial workforce (Brodkin, 1994; Ignatiev, 1995; Roediger, 1999; Weber, 2005). White business owners engaged in the industrial revolution valued punctuality and agreeableness over trade skills and self-regulation important in old world commerce (Roediger, 1999). European immigrants adopted these polite performances as a cloak against anti-immigrant sentiment, a means of proving that new immigrants were as white as the established U.S. power structure (Ignatiev, 1995; Roediger,


40
1999). Whiteness politesse and agreeableness, then, are defensive mechanisms for the newly white, reminding Guglielmo (2003, p. 41) of an Italian coal worker covered in soot who ran from a 19th century race riot shouting, “I’m white! I’m white!” Brodkin (1994) argues that whites continued to perform that capitalist version of gentility even as the Jews entered Whiteness in the middle of the 20th century1, and Bonilla-Silva (2003a) argues that many Latinos are attempting to make the same pivot to white agreeableness and insouciance in the 21st.
There is not much known about how dutiful masculine noblesse and innocent white gentility intersect in white men public servants. Both images focus on exerting great amounts of effort on behalf of employers to serve the interests of organizations, but many potential conflicts are suggested by the combination of aggressiveness and gentility. It is possible that white men managers use Whiteness performances to smooth over workplace conflicts, while bringing out Masculine performances to make authoritative decisions, thereby effectively embedding both institutions in ways that improve organizational effectiveness and efficiency, while hurting equity.
Psychology of Whiteness and Masculinity. Like all people, white men have imperfect perceptions of reality. Kahneman and Tversky (1984) explain that thinking happens at two levels simultaneously. The intuitive system reacts quickly and automatically to associate ideas with other ideas and feelings (Kahneman, 2003). This part of the brain is quick to make conclusions, slow to learn, and emotional (Kahneman, 2003). This intuitive system is almost effortless. It takes substantially more effort (and calories) to use the deliberative system, as is indicated by
1 This move by Jews bewilders Baldwin (1998, pp. 179-180) who asserts that American Jews agreed to “inform their children that Black women, Black men and Black children had no human integrity,” in exchange for “abandoning their children to the things white men could buy.” Jews, therefore, sold their morality in exchange for latent suburban hostility instead of outright antisemitism. Contrarily, Jews can claim that they are “not q-white.” Jews consistently vote with people of color in major elections (CNN, 2016), and Jews are still targeted with vandalism and violence at their homes, businesses, and places of worship.


41
increased heart rate, pupil dilation, and increased respirations (Kahneman, 2011). However, the deliberative system is more flexible and controllable, enabling individuals to reject initial assessments that are often biased (Kahneman, 2011). To reject a biased initial assessment is like swimming up-river by using a system that takes a substantial amount of effort to reject the conclusions of a system that takes almost none (Kahneman, 2011). When a human being processes reality, if the intuitive system gets it wrong, it is likely to stay wrong unless that person has the appropriate information and puts in the effort to overcome that incorrect perception.
When it comes to race and gender, the intuitive system often gets it wrong. Another pair of psychologists, Banaji and Greenwald (2013) have spent their career documenting intuitive assessments based on stereotypes. Their findings are both stunning and exactly what scholars of race and gender expect. It takes a fraction of a second less to group flowers and sunshine with pictures of white children than pictures of black children. It takes a fraction of a second more to associate nurses with men and doctors with women (Banaji & Greenwald, 2013). Greenwald and Banaji (1995) find that Americans’ intuitive systems create presumptions for fallacious stereotypes benefiting whites and men, and other researchers find that people act upon these presumptions even after admitting they are false (Nelson, Acker, & Manis, 1996).
Intuitive assessments impact how public servants think about others, and how they think about themselves. Stereotype threat research indicates that when a person is aware of a stereotype that is salient to their identity, they unconsciously conform to that stereotype (Steele, 2010). White men told their athletic ability is being compared to black men perform worse in mini-golf than white men who are not told they are being compared to black men (Steele, 2010). Similarly, white men perform worse in calculus classes when they share a room with large numbers of Asian men (J. L. Smith & White, 2002). Men perform worse on tests of reading body


42
language when they are told it is a “social sensitivity test” than when told it is an “information processing” test (Koenig & Eagly, 2005). One small (n = 33) experiment finds some (p < 0. 15) relationship between higher gender identification in men and improved math performance (Schmader, 2002). Also, Danso and Esses (2001) find that whites score higher on intellectual tests when the experimenter is black. By isolating the relationship of proctor to test taker, this experiment supports the hypothesis that whites “work especially hard to maintain the perception of White intellectual superiority” (Danso & Esses, 2001, p. 159). These studies suggest that white men benefit from a stereotype promise based in an intuitive belief in their own superiority.
The work required to maintain perceptions of superiority suggests that Whiteness and Masculinity performances require exertions of emotional labor. Emotional labor has often been associated with gender. Guy, Newman, and Mastracci (2008) argue that emotional labor accounts for some of the differences in pay between women who do more emotion-work-heavy jobs than men. Meier, Mastracci, and Wilson (2013) find that women perform more emotion work in the same jobs, and that this leads to better outcomes in public service, and Newman, Guy, and Mastracci (2009) find that women perform different emotional labor as leaders to create healthier workplaces. When it comes to emotional labor and race, Gooden (2015) argues that emotional management, in particular the management of nervousness surrounding race, affects individual public servants’ ability to produce social equity. Gooden (2015) points out that public servants are less likely to discuss racism if they are nervous to bring up the topic, and overcoming that nervousness requires emotional labor. Additionally, the well-documented Whiteness performances of innocence in the workplace (Brodkin, 1994; Roediger, 1999) as well as the Masculinity performances of unemotional aggression (Cooper, 2000; Morgan, 1996) resemble emotional labor researchers’ descriptions of false face. False face is the work of


43
displaying “an emotion one does not feel” (Guy et al., 2008, p. 109), and is a performance that is closely correlated with burnout.
The psychology of Whiteness and Masculinity gives significant benefits to white men in the form of stereotype promise and intuitive assessments of their performance and appropriate roles. As discussed in chapter one above, taboos and nervousness prevent open discussion of race and gender in public organizations (Gooden, 2015; J. Martin, 1990). When discussions do happen, critical scholars find that common rhetorical devices are used to minimize racism and sexism, or accuse women and people of color of racism and sexism against white men.
Rhetorical Whiteness and Masculinity. In public organizations, nervousness and private life taboos are unable to suppress all discussion of race and gender. For example, Foldy and Buckley (2014) find that welfare agencies that suppress discussion of race develop discriminatory ways of speaking. Similarly, Gooden (2014, p. 72) argues that public servants who do not overcome their nervousness about race use “conversational variability” to “act as a chameleon, expressing shifting messages regarding racial equity in their agency depending upon the setting and the audience.” Several rhetorical performances enable discussion of race and gender without directly confronting nervousness and taboos, including microaggressions, minimization, abstract liberalism, projection, and displacement.
Microaggressions communicate Whiteness and Masculinity while maintaining the power of apparent gender- and color-blindness. A microaggression is a brief or inconspicuous communication of power based on group identity (Sue, 2010). Examples of microaggressions fill scholarly literature and popular media. A woman public servant may be introduced as “Kathy” by her boss even as he introduces all of her men co-workers as “Mr. Smith” (Sue, 2010). Similarly, a white woman in an elevator may clutch her purse and recoil when a black college


44
professor enters (Yancy, 2008). Politically incorrect statements can be microaggressions, as can wall hangings of offensive mascots, or tone-deaf compliments of people who appear to be defying stereotypes associated with their phenotype (Sue, 2010). Taken in conjunction with the work of Foldy and Buckley (2014) and Gooden (2014), microaggressions are the inevitable result of nervousness. If color- and gender-blindness prevents open discussion of race and gender, many of these microaggressions may simply be caused by ignorance on the part of people who are too nervous to ask questions that would help them prevent hurting others. In other words, it could be that microaggressions are like a pop quiz that white men and others too often fail unless they overcome their nervousness to do their homework.
Microaggressive rhetoric is used to downplay discrimination. Sue (2010) identifies the use of the phrase “I believe the best candidate should get the job” regardless of race or gender as a microaggression (p. 13). Bonilla-Silva (2003b) finds the “best candidate” language is used to justify actions that are discriminatory by hiding bigotry behind a veil of abstract neutrality. This neutrality is false given the many advantages that whites and men receive throughout the course of their academic and professional careers (Lipsitz, 2009; Mumby, 2006). However, the claim can appear neutral enough to inoculate the shame and anger that many white men feel when their innocence and noblesse are called into question by accusations of racism and sexism (Andersen, 2016; Bonilla-Silva, 2003b; Matias, 2016; Sue, 2010).
Another strategy for downplaying discrimination is to minimize the experiences of oppressed groups (Bonilla-Silva, 2003b). White men often celebrate symbolic equality gains, distance themselves from oppression by emphasizing the length of time since slavery or primogeniture, or make specious comparisons between the struggles of oppressed groups and the struggles of white men in historic times (D. Bell, 1980; Bonilla-Silva, 2012; Gallagher, 2003).


45
When all else fails to maintain innocence and noblesse, white men sometimes turn to projection where they accuse others of what they themselves are accused (Bonilla-Silva, 2003b). For example, in 2016, 43 percent of whites told Gallup that racism against whites was widespread, and a majority of Republican voters claimed discrimination against whites was more prevalent than against blacks or women (J. M. Jones, 2016).
Displacement is another psychological term that is analogous to microaggressive performance, but displacement works in the intersection between identities. Displacement is a performance from one institution that reinforces or distracts from a different institutional performance. For example, Choudhury (2007) and others (Ali et al., 2015) document how hijab and niqab are targeted by white feminists as oppressive in ways that reinforce white innocence compared to the morality of Muslims. Similarly, feminist scholars argue that the struggle to racially integrate military units and leadership is tainted by arguments that reinforce sexism (Scales, 2013). G. M. Frederickson (1988) argues that neo-Marxists’ focus on class downplays the centrality of race, and thus obscures “the cultural and psychological dimensions of race relations and underplays the historical significance of racism” (p. 4).
In popular culture, displacement is evident in arguments that defend violence perpetrated by poor white men as symptoms of their poverty (Murray, 2012; Vance, 2016), without discussing statistics that show how poor white women do not commit violence at anywhere near the same rate (Gilligan, 1996; Katz, 2013), nor discussing the many ways that this behavior serves the interests of non-poor whites. Allen (2008) points out that non-poor whites argue poor whites are the reason that racism continues to exist, linking racism to poor whites’ failed backwardness and savage masculinity. Yet data show that poor whites are no more racist than non-poor whites, Allen (2008) explains, because it is the wealthy, not the poor, who receive the


46
many benefits of Whiteness. Backing up this claim, experimental research finds that non-poor whites are more likely than poor whites to vote against a candidate that they feel antagonizes people of color, not because they reject racism, but because they fear upending the social structure from which they benefit (Jun, Lowery, & Guillory, 2017).
Individual-level Propositions. Individual performances like those documented in the literature will be the color- and gender-blind normal for most public servants, supporting propositions that can be tested at the individual level.
Individual Proposition #1- Public servants ’ daily tasks include performances of Whiteness and Masculinity.
Individual Proposition #2- Public servants make sense of race and gender through performances of Whiteness and Masculinity.
Organizational-level Whiteness and Masculinity
Masculinity and Whiteness in organizational culture. Organizational culture is the process through which people within an organization place value on work processes (Schein, 2010; Weick, 1995) (see Table 2.3). Gender- and color-blindness are highly successful performances in organizational culture because they are dominant at the system-level, required by law for public organizations, and reinforced by individual nervousness. As discussed above, modern Whiteness and Masculinity rely on the power to suppress discussion and keep white supremacy and patriarchy normalized (Bachrach & Baratz, 1962; Mills, 1997). Public organizations themselves benefit from suppressing talk of race and gender. Heckler (2017) argues that color-blindness in public administration hurts Americans of color and reinforces Whiteness, while increasing the political legitimacy of public organizations by conforming to


Table 2.3:
Organizational-level performances of Whiteness and Masculinity.
Dress, Speech, and Mannerisms Statements, artifacts, and workplace decisions depend on assumptions that white middle-class Midwest suburban masculinity is the norm.
Exclusive Touchstones Using ceremonies, extracurricular activities, and celebrations that are associated with white men.
Innocence The claim that whites or white culture is more morally justifiable.
Noblesse The claim that men are obligated to protect innocents by asserting power over them and POC. Statements, artifacts, and workplace decisions that assume
Private Life Taboo employees have support networks to do their domestic labor, including elder and childcare.
Technocracy Statements and artifacts supporting doing things the most efficient way to justify benefits to white men.
Adhocracy Organization management take on different roles depending on organizational needs, while frontline employees do not.
Expert/Non-Expert Dichotomy A group of experts who are more often white men, separate from a group of non-experts who are more often women and people of color.
Power Shifting The allocation of decision-making and resources to units dominated by white men.
Strong Vertical Allocating more decision-making and resources to executives Bureaucracy in the organizational apex dominated by white men.
Examples
A manager scolds a new employee for having a shirt that is untucked and using too much slang in a staff meeting.
A unit gathers together for an annual super bowl watching party.
A TSA employee thinks nothing of a white man carrying a briefcase because he does not "look like a terrorist."
A TSA employee targeting a man who looks Arabic because of concern
over terrorism.
An agency promotes a white man partially because he was able to work all night at the last minute while his parents cared for his children.
Using racial profiling to target POC on the argument that iteffectively discovers undocumented immigrants.
In response to budget cuts, management starts contracting to analyze the inefficiencies among frontline workers.
Whites and men more common among agency attorneys, and women and POC more common among support staff.
More resources and authority are found in the hands of a division dominated by white men than a similar division dominated by women
and POC.
Upper management in an agency is dominated by white men.
-P*.


48
Constitutional legal fictions prohibiting the consideration of race. Heckler (In Press) also argues that nonprofit law places pressure on boards and staff to conform to Whiteness processes borrowed from the private sector. Similarly, Scales (2006) argues that gender-blind justifications hurt women and maintain patriarchic values (Geduldig v. Aiello, 1974; see also Scales, 1992). Yet under current law, many public organizations are not legally permitted to acknowledge gender and race (Regents of the Univ. of Calif v. Bakke., 1978; Clark v. Jeter, 1988).
Organizational culture extends beyond color- and gender-blindness. Organizations actively repeat Whiteness and Masculinity cultural performances. Gooden (2014) finds that leadership is an important factor in undermining nervousness and creating a culture that embraces discussions of race. This finding suggests that leadership help create cultures of nervousness in the first place. Leadership teams dedicated to not seeing race make organizational decisions like promotions, assignments, hiring, and termination using Whiteness performances (Gooden, 2014; Greenwald, Banaji, & Nosek, 2015a). For example, Carbado and Gulati (2013) describe the experiences of lawyers in trial working through the night and watching the sunrise in the morning. They argue that successful white lawyers describe the sunrise from their office, and leave their offices strewn in papers and takeout boxes as artifacts of their dedication. Successful black lawyers keep their offices tidy and do not emphasize their marathon work sessions because they fear it conforms to black stereotypes of sloppiness and bad planning. The fact that these stories are from the perspective of successful lawyers indicates that it is not the individual attorney who makes these decisions, but law firms that reward people of color who can negotiate the organizational-culture minefield (Carbado & Gulati, 2013).
Similar to findings that whites take advantage of expectations that benefit them, Kanter (1977) finds that men and women have disparate expectations that advantage men. Men are


49
expected to aggressively pursue promotions, playing the weaknesses of other men when necessary to get ahead (Kanter, 1977). Women, by contrast, are expected to be loyal. Women must give credit for their success to men in the hopes that those men will remain loyal to them and help them climb the bureaucracy (Kanter, 1977). However, loyalty behavior in men is not enforced, leaving women in a cultural minefield of their own. In the 21st century, these expectations remain the norm (Cooper, 2000).
Organizational cultures work intersectionally. White men take advantage of organizational cultures that encourage ambition primarily in men (Kanter, 1977), while accepting favorable assumptions that offices occupied by whites are messy as the result of honest hard work, not sloppiness and disorganization (Carbado & Gulati, 2013). White men also take advantage of their intergenerational wealth by securing eldercare and childcare in nursing homes and daycare centers, even as people of color that have had their wealth systematically drained by alienation of their labor and discriminatory housing policies (Oliver & Shapiro, 2006) must negotiate masculine taboos that childcare and eldercare is domestic labor better left at home (J. Martin, 1990). White men enter organizations with cultures that have often been created by white men. This path dependency is evident in cultural touchstones that exclude women and people of color, such as in golfing with the bosses, company softball, watching a sporting event, or chatting about the latest Quentin Terantino film over the water cooler (Cabrera, 2016; Carbado & Gulati, 2013; Matias, 2016; Sue, 2010). Often, these exclusive touchstones make the difference in the careers of white men who more effortlessly engage in the organizational culture.
Masculinity and Whiteness in organizational structure. Masculinity and Whiteness affect organizational structure by power shifting so that more decision-making authority and resources are allocated to whatever part of the organization has more white men. Mintzberg


50
(1983) finds that organizations change their structures to reflect power in those organizations.
For example, a professional bureaucracy, like a law firm, is likely to spend more resources and put more decision-making authority near the bottom of the organization where the lawyers work (Mintzberg, 1983). Organizational theorists have argued that both technocracy and bureaucracy put more power in the hands of white men (Burris, 1996; Ferguson, 1984).
Bureaucracy reinforces performances of Masculinity and Whiteness. Bureaucracies consist of clear hierarchies, documented decision-making guidelines, and careful cultivation of culture as effective means of creating unified organizational control even when organizations are decentralized and separated by geographical and cultural distance (Kaufman, 2006; Weber,
1972). Gulick (1937) finds that bureaucratic coordination requires four steps that are impacted by race and gender. First, managers define the job, a task burdened with intuitive assessments about how jobs should be grouped together. Managers set accountability mechanisms, which reflect trust set by Whiteness and Masculinity. Then managers decide how many people are needed for each task and establish a power structure, both tasks dependent on the same Whiteness and Masculinity performances that impact other organizational tasks.
Ferguson (1984) argues that public administration is a “cult of rationality” (p. 52) because of the emphasis on control over workers to submit to the will of a masculine bureaucracy of “managerial skills and beliefs” above all else. Just as Follett (1924) outlines a model for organization that empowers front-line workers, Ferguson (1984) offers alternative forms of organization, but these alternatives still exist within an institutional framework that is laden with Masculinity and Whiteness. For example, Chen (2009) studies the annual Burning Man event to find that lack of centralized organization ends up reinforcing hegemonic power dynamics that benefit white men.


51
The most common alternative to the strong vertical bureaucracy, technocracy reinforces Masculinity and Whiteness by providing justifications for racist and sexist workforce divisions. Stivers (2000) traces the origin of public administration from a conflict of approaches taken by the New York Bureau of Municipal research dominated by men, and the settlement houses that were mostly the purview of women. Epitomized by Simon (1959), bureau men focused on scientific accountability and efficiency. Contrarily, Jane Addams (1911) pioneered the settlement house movement dedicated to practical solutions and research from within communities in need. Stivers (2002) argues that the efficiency and scientific accountability perspective reflect a masculine focus that has come to dominate modern public administration research and practice. Burris (1993) calls this focus technocracy. Technocracy’s emphasis on efficiency and scientific accountability shifts the focus of public administration from what is being done to how things are done (Burris, 1993). In other words, technocracy encourages public servants to ask if they are “doing things the right way,” instead of whether they are “doing the right things...” (Adams & Balfour, 2004, p. 41).
The technocratic emphasis on expertise divides workers in two groups often by race and gender, and treats each group disparately. Burris (1993) explains that technocracy is not possible without an expert class to provide scientific answers to questions about efficiency. As a result, organizations become divided into expert and non-expert sectors, and while egalitarianism takes over the expert sector (Mintzberg, 1979), the non-expert sectors find their work becomes increasingly dictatorial (Burris, 1993). Burris (1996) finds the expert sector occupations are those that are dominated by men, thus, a system that seems to be about efficiency becomes “a system of male dominance in which some men (and women who adopt patriarchal values and practices) dominate other men and most women...” (p.64). The argument extends to people of color


52
through an emphasis on degrees as proof of expertise combined with the challenges that people of color face in university settings (Fryar & Hawes, 2012; Steele, 2010). Even as the expert class becomes flatter in the technocratic meritocracy, the non-expert jobs assigned to women and people of color pay less, and become ever more tightly controlled (Burris, 1996). A focus on efficiency serves these technocratic organizations because it enables organizations to survive in an environment that is itself focused on efficiency over equality, while at the same time giving more value to microaggressive “best candidate” and most efficient process rhetoric. Recent technocratic movements in public administration include New Public Management (Page, 2005), the Reinvention of Government (Ingraham, Thompson, & Sanders, 1998), and e-governance (Dunleavy, Margetts, Bastow, & Tinkler, 2006). As public organizations are buffeted by these tides of demands for more scientific accountability and efficiency, the gulf between expert and non-expert widens.
Public organizations are pushed by Whiteness and Masculinity from systems-level institutional performances, even as they are pulled from within by individuals performing Whiteness and Masculinity. This squeezing shapes organizational structures and cultures, and is normalized by the fact that they are set against a society that is mostly white and masculine. However, normalization does not erase problems. Whiteness and Masculinity in public organizations are both well-documented, and almost entirely unnoticed because of gender- and color-blindness among researchers in many fields. This theory isolates Whiteness and Masculinity and subjects the intersection between them to the same scrutiny that has been placed on other institutions such as capitalism (Weber, 2004), consumerism (Campbell, 1987), Blackness (Fanon, 1963), and Femininity (Bartky, 1990).
Organizational-level propositions. Organizations form and persevere in societies that


53
are dominated by whites and men, and are occupied by individuals who must negotiate Whiteness and Masculinity. As a result of these pressures from both sides, four propositions can be explored in organizations.
Organizational Proposition #1- Public organizational structures allocate power and resources to whites and men.
Organizational Proposition #2- Public organizational cultures reflect Whiteness and Masculinity performances.
Organizational Proposition #3- Public organizations with relatively fewer performances of Whiteness and Masculinity will be more welcoming to people of color and women.
Organizational Proposition #4- Public managers implement Whiteness and Masculinity performances as operational processes and systems.
Whiteness and Masculinity in Public Service Motivation and Job Satisfaction
Internal organizational race and gender are not as well studied in public organizations as in organizations more generally (Gooden, 2014). The representative bureaucracy literature shows gender and race of public servants affect a variety of outcomes from public organizations (Riccucci & Van Ryzin, 2017; Selden, 1997). At the same time, organizational theorists find that Masculinity and Whiteness impact the internal workings of organizations generally (D. L. Collinson & Hearn, 1996; Macalpine & Marsh, 2005). Even reconceptualizing much of the representative bureaucracy literature as implicitly analyzing Whiteness and Masculinity, there is little to no research on Masculinity and Whiteness in the internal workings of public organizations. One throughput that is unique to public organizations is public service motivation (PSM) (Perry & Wise, 1990).
Masculinity and Whiteness affect PSM of white men and others in public organizations.


54
Perry and Wise (1990) find that employees of public organizations are motivated differently than employees of private sector organizations. They argue that public employees are driven in part by the rational motives often theorized in the business management literature, along with norm-based motives to serve the public interest, and affective motives to serve the missions of public organizations (Perry & Wise, 1990). Perry (1996) defines four dimensions of PSM that have been widely tested and found to be related to various outcomes including job satisfaction, lower job turnover, and willingness to choose a public service job (Perry, Hondeghem, & Wise, 2010, p. 683). People exhibiting these four dimensions are attracted to public policy making, desire to work in the public interest, are driven by compassion, and have a willingness to self-sacrifice (Perry, 1996, p. 15). In the 25 years since Perry and Wise (1990), dozens of researchers have operationalized one or more of these dimensions (Perry et al., 2010).
In multiple tests, PSM is found to differ based on racial and gender performance. Most of the time, these differences are reported in a study’s controls. While Liu and Perry (2014, p. 13) find no difference between PSM in men in the Chinese government, many other studies have found differences in American public workers. Men have been found to have less attraction to policy making in one study (Moynihan & Pandey, 2007, p. 46), and be less likely to volunteer in a model predicting PSM (Perry, Brudney, Coursey, & Littlepage, 2008, p. 451). Dehart-Davis, Marlowe, and Pandey (2006, p. 880) find that women are more attracted to policy making and more driven by compassion before exhorting researchers to “focus on a better and more nuanced understanding of the role that gender plays” in developing PSM. While researchers have stipulated that formative experiences like race and gender performance do create and maintain PSM, little research has looked at mechanisms (Weaver, 2015, p. 445).
PSM can be changed by organizational factors. Organizations manage PSM in their


55
employees through “attraction-selection-attrition” (Perry et al., 2010, p. 683). Organizational factors like red tape, hierarchical authority (Moynihan & Pandey, 2007, p. 46), and strong organizational identity (Liu & Perry, 2014) and mission (Lee & Wilkins, 2011) change PSM over time. While individuals are generally assumed less malleable than organizations in the development of PSM, at least two studies find that organizational characteristics can orient individuals toward higher levels of PSM in the same individuals over time (Liu & Perry, 2014; Ward, 2014). Between attraction, selection, and attrition within an organization, and development of PSM through public management, the research shows that PSM should change over time within public organizations generally, evolving with the institutions around them and the individuals within them.
Job satisfaction relates to public service motivation, but is not unique to public organizations. Herzberg’s (1974) hygiene hypothesis makes sense of the difference between motivation and satisfaction. Herzberg (1974) argues that motivation and satisfaction are driven by different kinds of factors. While PSM is a theorization of motivation unique to public organizations, hygiene factors like supervision, interpersonal relationships, working conditions, salary, status, and job security are attributes of all organizations with employees (Herzberg,
1974, p. 20). Herzberg (1974) argues that motivation factors in private companies will not impact retention unless hygiene factors are adequately addressed, and that simply acknowledging a problem and resolving it is only the beginning step in motivating employees to more productivity (Herzberg, 1974, p. 25). It is unclear from existing literature if racial and gender performance impact motivation or hygiene, or whether PSM is impacted differently.
Some research on satisfaction and turnover indicates that gender and race performance may impact job satisfaction in public organizations. Weaver (2015) finds that men and whites are


56
less likely than people of color and women to leave a federal job for another federal agency, but not less likely to leave for another sector. Pitts (2009) finds that whites are less satisfied than non-whites in their job after diversity training. While empirical studies on the ways that racial and gender performance affect job satisfaction are scant, this literature suggests that Whiteness and Masculinity performances are likely to impact job satisfaction.
Organizational-level Hypotheses. The literature cited here supports a theory that organizational Masculinity and Whiteness impacts the PSM and job satisfaction of whites, men, and white men in public organizations. These public servants embody all of the individual, organizational, and systems-level performances in Tables 1-3. Organizational Whiteness and Masculinity will be different in each organization, as each organization negotiates and embeds their institutional makeup. The literature indicates that organizations with the most embedded Masculinity and Whiteness will give substantial benefits to whites, men, and white men, and those whites, men, and white men are the most likely to take on supervisorial roles, often over women, people of color, and women-of-color (Carbado & Gulati, 2013; Kanter, 1977). Using supervisorial status as a proxy for benefit to whites, men, and white men that indicate Whiteness and Masculinity, the following hypotheses can test internal organizational race and gender by measuring PSM and job satisfaction.
Hla: When whites are over-represented in supervisorial teams, whites in those organizations will exhibit more PSM.
Hlb: When men are over-represented in supervisorial teams, men in those organizations will exhibit more PSM.
Hlc: When white men are over-represented in supervisorial teams, white men in those organizations will exhibit more PSM.


57
H2a: When whites are over-represented in supervisorial teams, whites in those organizations will be more satisfied.
H2b: When men are over-represented in supervisorial teams, men in those organizations will be more satisfied.
H2c: When white men are over-represented in supervisorial teams, white men in those organizations will be more satisfied.
Conclusion
What this chapter describes is a syndemic of Whiteness and Masculinity. A syndemic is a complex of outcomes, mechanisms, and causes that are all intimately related, each cause contributing to an outcome that is also a mechanism that contributes to the original cause. Singer (2000) coined the term syndemic in his study documenting how the AIDS crisis was more than just a series of quantifiable risk factors, but an intertwined social pattern where violence, drug use, and HIV infection were causally related in multidirectional relationships. In a more recent study, Horton (2016) records the intricate relationship between kidney disease, heat-related disease, poverty, diabetes, food inspection laws, immigration policy, and undocumented status among farming communities in California. The propositions and hypotheses established in this chapter help outline a systematic process for exploring a syndemic of individual behaviors and assumptions; organizational structures, cultures, and processes; and societal forces and influences that combine to reinforce white supremacy and patriarchy in public organizations. The hypotheses will be tested using publicly available from the federal government. The study will explore the propositions by interviewing public servants in the federal government, following their social media postings, and tracking their work processes. The next chapter describes this methodology.


58
CHAPTER THREE
OBSERVING WHITENESS AND MASCULINITY IN SOCIETY, INDIVIDUALS, AND
PUBLIC ORGANIZATIONS
Focusing on the federal workforce, this study investigated the hypotheses and propositions presented in Chapter Two. In the first stage, quantitative methods tested hypotheses that race and gender impact public service motivation and job satisfaction. Guided by the findings of the first stage, the second stage conducted an institutional ethnography on agency offices analyzed in stage one. This chapter describes the two stages of research.
Figure 3.1 visually depicts how this dissertation sought to combine stage one quantitative methods with stage two qualitative methods. The findings and summary statistics of stage one identified organizations as typical examples of federal government offices for further qualitative analysis in stage two. By placing the qualitative research in a context developed through quantitative methods, the qualitative research supports a theory that is representative of the scores of offices operating within the civilian federal government (Bourgois, 1999). The stage-two findings formed an empirical basis for theorizing the mechanisms behind the PSM and job satisfaction differences discovered in stage one. This triangulation is suggested by Riccucci (2010) as a way to combine “the benefits of qualitative and quantitative methods” (p. 108).
Stage One: Race and Gender in the Federal Employee Viewpoint Survey
The hypotheses laid out in Chapter Two were tested using quantitative methods on a survey of the federal workforce. This section reviews the hypotheses and describes the dataset, before explaining how the independent, dependent, and control variables were compiled. The analysis subsection describes how multiple estimation tools were used to overcome various
limitations in the data.


59
behind the stage one estimates.
Figure 3.1. Model of the relationship between stage one and stage two research.


60
Hypotheses: Race and Gender Affect Job Satisfaction and Public Service Motivation
Six hypotheses guide this research. Table 3.1 lays out the hypotheses along with the variables needed to test them. The rest of this subsection discusses the dataset, and how this research operationalizes those variables.
Data: the FEVS
Stage one employed the Federal Employee Viewpoint Survey (FEVS) (OPM, 2016) to test these hypotheses. The FEVS (OPM, 2016) is an annual survey distributed to federal employees including military and civilian personnel. The survey includes 84 questions on morale, workplace norms, job satisfaction, and basic demographics. The dataset includes more than 1.5 million observations. The data are longitudinal with a cross-section from six years, 2010-2015. The observations are also nested in 20 agencies and 136 federal offices within those agencies. The FEVS is generally considered a representative dataset (Weaver, 2015a) with 45.8 percent response rate government-wide. Table 3.2 displays some of the similarities between the sample of federal workers who filled out the FEVS and the cumulative population of federal workers in the same five years as described by the OPM. The proportion of men and whites in the sample are very similar to the population. When it comes to paygrade, the FEVS sample is 31.8 percent GS13 or higher, while the total population is 20.3 percent GS13 or higher, indicating some difference in the education and experience of the sample. Of the control variables, paygrade had the lowest impact on initial estimates of satisfaction and PSM, indicating that the difference between the sample and the population was not a fatal limitation. FEVS data included variables for measuring independent, dependent, and control variables for testing the study’s hypotheses.


61
Table 3.1
Stage one hypotheses.
Hypothesis Independent Variable (Office Level) Dependent Variable (Individual Level)
HI a: When whites are over-represented in supervisorial teams, whites in those organizations will exhibit more PSM. The proportion of white supervisors minus the proportion of white employees. PSM Index
Hlb: When men are over-represented in supervisorial teams, men in those organizations will exhibit more PSM. The proportion of man supervisors minus the proportion of man employees. PSM Index
Hlc: When white men are over-represented in supervisorial teams, white men in those organizations will exhibit more PSM. The proportion of white man supervisors minus the proportion of white man employees. PSM Index
H2a: When whites are over-represented in supervisorial teams, whites in those organizations will be more satisfied. The proportion of white supervisors minus the proportion of white employees. Satisfaction Index
H2b: When men are over-represented in supervisorial teams, men in those organizations will be more satisfied. The proportion of man supervisors minus the proportion of man employees. Satisfaction Index
H2c: When white men are over-represented in supervisorial teams, white men in those organizations will be more satisfied. The proportion of white man supervisors minus the proportion of white man employees. Satisfaction Index
Table 3.2
Similarities between the FEVS and the Federal Workforce generally.
Demographic % of FEVS % of Federal Workforce
Men 51.2% 56.4%
Whites 65.3% 65.9%
Pay grade > GS12 31.8% 20.3%


62
Independent Variables- Racial and Gender Congruence
The stage-one independent variables were race and gender congruence of supervisors to employees in federal offices. The FEVS (2018) asks respondents if they are supervisors, their race, Hispanic or non-Hispanic ethnicity, and whether they are male. To guard respondents’ anonymity, the FEVS lists all persons who are not white non-Hispanic as “minority” employees. For this study, a person who identifies as a male is counted as a man, and a person who selects white non-Hispanic is counted as white. Federal offices range widely from 7.2 percent to 89 percent white employees, 22.1 percent to 82 percent men employees, and 4.5 percent to 63.6 percent white men employees (see Table 3.3).
Table 3.3. shows that whites and men are over-represented in supervisorial roles compared to their employees. To operationalize this over-representation, the proportion of white/man/white man supervisors in each office is subtracted from the proportion of white/man/white man employees in each agency office. Take the example of an office with 60 percent white supervisors, and 60 percent white employees. This office would be 0 percent overrepresentative because 60 minus 60 is 0. However, the hypothetical office that has 5.5 percent too many white supervisors could mean that the supervisors would be 65.5 percent white, while the employees would be only 60 percent white. Table 3.3 shows that employees are 65.3 percent white, 51.2 percent men, and 35.9 percent white men. However, these employees are managed by supervisorial teams with 5.5 percent too many whites, 9.9 percent too many men, and 7.7 percent too many white men in supervisorial roles. Congruence as independent variables were modeled as predictors for PSM and job satisfaction as dependent variables.


63
Table 3.3
Independent variables summary statistics
Independent Variables N Mean Std. Dev. Min Max
Percent white 1,676,564 65.3% 10% 7.2% 89%
Percent men 1,750,885 51.2% 14% 22.1% 82%
Percent white men 1,676,564 35.9% 12% 4.5% 63.6%
Percent too many white sups 1,750,885 5.5% 4.4% 18.9% 47.4%
Percent too many men sups 1,750,885 9.9% 4.4% -2.9% 33.4%
Percent too many white men sups 1,676,564 7.7% 3.6% -3.1% 36.4%


64
Dependent Variables- PSM and Job Satisfaction.
The dependent variables for the stage one analysis were PSM and job satisfaction.
Several existing studies have used the FEVS to capture public service motivation and job satisfaction (e.g. Alonso & Lewis, 2001; Harrington & Lee, 2015; Pitts, 2009; Weaver, 2015). The five facets of PSM are attraction to policy making, commitment to public interest, civic duty, compassion, and self-sacrifice (Perry, 1996). This study measures three of these facets; commitment to public interest, civic duty, and self-sacrifice. Using the FEVS, Weaver (2015, p. 446) captures commitment to public interest with interaction terms derived from responses to the questions “The work I do is important,” multiplied with “I know how my work relates to the agency’s goals and priorities.” Using the list of questions that capture PSM provided by Perry (1996), this dissertation develops a more nuanced measure of commitment to public interest by combining the responses to “I like the kind of work I do,” and “Work gives me personal accomplishment” with “I know how my work relates to the agency’s goals and priorities.” An index variable was constructed from the three terms with a Cronbach’s alpha of .7957.
Civic duty was measured using an index of the responses to the questions “I am constantly looking for better ways to do my job,” “My agency is successful at accomplishing its mission,” and “Considering everything, how satisfied are you with your organization?” (OPM, 2016). These three responses were each multiplied by “I know how my work relates to the agency’s goals and priorities” to represent civic duty (a = .9414). A single question, “When needed I am willing to put in extra effort” measures self-sacrifice. These operationalizations were similar to, or directly quoting Perry’s (1996) list of questions capturing PSM. The other two facets of PSM, compassion and policy making were not readily captured using responses to the FEVS. As displayed in Table 3.4, the overall PSM index had a skewed mean of 3.73 out of five


65
possible points (the mean with no skew would be exactly three), indicating that federal employees exhibit high levels of PSM. Whites, men, and white men have lower levels of PSM compared to the entire sample, revealing racial differences in PSM.
Job satisfaction was measured with an index constructed from the responses to seven satisfaction questions including “How satisfied are you with pay?,” . .management?”, “.. .your involvement in decisions that affect your work?”, “..recognition on the job?”, “.. .promotion opportunity?”, “.. .training?”, and “.. .overall?’, with an alpha of .884. “Overall” is given equal weight to all six of the other satisfaction variables combined in the index. As with PSM, the mean satisfaction index for federal workers is skewed high. Table 3.4 shows the mean federal worker indicating a 3.48 out of five satisfaction score. Unlike with PSM, whites, men, and white men did not show statistically significant differences in job satisfaction compared to the entire sample. To verify that the PSM and job satisfaction of these employees is affected by the congruence of race and gender of their supervisors, the study must eliminate alternative explanations by using control variables.
To account for common source bias, factor loading revealed that the dependent variables did not hang well with the explanatory independent variables. Meier and O’Toole (2013) explain that common source bias is a threat to validity when the dependent variable and the independent variable both originate from the same source, in this case the same respondent. To test for common source bias, factor analysis was run on all variables included in the model to determine if the dependent variables hang with the explanatory independent variables. The highest factor score for any dependent variable was <.08, revealing very low risk of common source bias. Control Variables
Common control variables are found in the literature on PSM and job satisfaction (Pitts, 2009; Weaver, 2015a). These control variables include military/veteran status, paygrade,


66
Table 3.4
Dependent variables summary statistics
Dependent Variables N Mean (out of 5) Std. Dev.
PSM 1,621,339 3.73 0.791
if white 1,012,456 3.71 0.784
if man 834,454 3.69 0.820
if white man 562,163 3.67 0.812
Job Satisfaction 1,701,402 3.48 0.928
if white 1,056,954 3.49 0.920
if man 875,630 3.48 0.936
if white man 587,578 3.48 0.929
Table 3.5
Control variables summary statistics.
Control Variables N Mean Std. Dev. Min Max
GS 13 1,245,158 31.8%
>5 Years’ Experience 1,750,885 73.6%
Under 40 1,723,327 24.8%
Active Duty or Veteran 586,169 28.1%
Unit Resources 1,750,885 3.32 0.167 2.790 3.950
Observations perorg/Year 1,750,885 10,531 12317 177 46687


67
experience/tenure in the government, age, and organizational resources, and organization size. Veteran/Military status, pay grade, age, and tenure are self-reported on the FEVS. The job resources variable is borrowed directly from Pitts (2009, p. 333), and consists of an index of six measures including responses to “my work unit is able to recruit people with the rights skills,” “I have sufficient resources to get my job done,” “my workload is reasonable,” “physical conditions allow employees to perform their jobs well,” “I receive the training I need to do my job,” and “considering everything, how satisfied are you with your pay?” The unit resources were measured at the agency office level to capture the group perception of resources, thereby mitigating personal bias. Table 3.5 lists these variables along with some summary statistics. With the variables compiled, the next step was to analyze the control variables’ and independent variables’ impact on PSM and job satisfaction.
Analysis
In this subsection, several estimation tools for testing the hypotheses are described, along with how each tool addresses different limitations. The data are limited by the characteristics of the data including skewness, autocorrelation, and multicollinearity. Estimation tools that can address these limitations include Bayesian, Tobit, binary logit, ordered logit, mixed-effects regression, and propensity score matching. Table 3.6 outlines some of the characteristics and limitations in the data, along with suggested methods of addressing them. Having laid out the multiple analysis tools in Table 3.6, each analysis tool is described below starting with Bayesian estimates.
Bayesian estimation. Bayesian statistics calculate propensity for the observations in a dataset given a certain condition. This differs from most statistical methods used in social science, which calculate the frequency that a value is found among all observations. This means


68
Table 3.6
Analysis strategies for overcoming limitations in the FEVS dataset.
Characteristics of FEVS Appropriate Analysis Tool Limitations Overcoming Limitations
Large-n (>1.5 million observations) creating risk of false positive findings Bayesian analysis to reduce risk of false positive No clear “statistical significance.” Bayesian tools less well developed. Use non-Bayesian (i.e. standard) methods initially
Non-normally distributed, skewed toward higher numbers Tobit regression Not appropriate for variables based on numerical choice (i.e. Likert surveys) Create a dummy variable for logit or probit analysis
Non-normally distributed, skewed toward higher numbers Logit regression Creates arbitrary cutoff that may not be empirically meaningful. Analyze as continuous variable, or Likert variable in Ordered Logit.
Ordered choice dependent variable (Likert scale) Ordered Logit Not appropriate for skewed data, nested data, risks of large-n data Analyze using Bayesian, logit, Tobit, and multi-level methods
Nested at three levels; organization, agency, and year. Mixed-effects (Hierarchical linear Modelling) Does not account for skewed data, assumes linearity, multicollinearity. Models do not “converge” Different optimizers, standardized variables, analyze with propensity score matching, Bayesian, logit, Tobit.
Multicollinearity of important I.V.s Propensity score matching followed by multiple regressions Large-N dataset and nested data Run estimations on separated datasets, then compare using t-tests


69
that Bayesian estimates have different strengths and weaknesses than more standard statistical methods, and can complement other methods. For example, one limitation of standard statistical methods is an increase of false-positive findings in datasets of more than 10,000 observations. Social science methodologists and statisticians agree that Bayesian regression methods are an effective means of mitigating this increased risk of false-positives (Cowles, 2013; Gill & Meier, 2000; Kruschke, 2012; Marchenko, 2015; Masson, 2011; Rice, 2014). The 10 years of data being used in this analysis include over 2 million observations, and so the risk of false-positive findings (type I error) is a limitation that should be addressed.
There are many tools for running Bayesian regressions in R including the BLME package (Dorie, 2015) for estimating Bayesian generalized linear and non-linear modelling. Despite their advantages, Bayesian methods are often critiqued for being less clear than standard statistical methods (Gill & Meier, 2000). Bayesian can be combined with other estimation tools to account for other limitations, such as Tobit and logit methods that account for non-normal distribution.
Tobit and binary logit Tobit and binary logit methods account for the fact that the data are not normally distributed. Most methods assume that data are normally distributed around the mean, but the PSM and job satisfaction indices created for this study are skewed to higher numbers, with too few responses in the lower range. Ultimately, Chapter Four reports binary logit with odds ratios to test the hypotheses. This method accounts for non-normal distribution by picking an arbitrary cut off, such as the median PSM or satisfaction score, and creating a dummy variable where 1 indicates above the median responses, and 0 indicates below the median responses. The limitation in creating this dummy was that the arbitrary cut-off may not have empirical validity. Another way to account for this non-normality is to run Tobit estimations that are designed to account for this skew (Wooldridge, 2013). However, Tobit


70
regressions are not by themselves capable of limiting the threats to validity caused by large numbers of data and nesting. Also, the Tobit assumption of linearity is violated by the fact that the PSM and job satisfaction are not, in fact, linear variables, but ordinal.
Ordered logit The default regression method for surveys in which respondents answer on a 1-5, or Likert scale is ordered logit. Ordered logit assumes that variables are ordinal, like the Likert scales in the FEVS, and accounts for some of the non-normal distribution in the dataset (Wooldridge, 2013). Ordered logit cannot, by itself, account for the skewed data nor for the risk of false-positives created by large datasets. Ordered logit also must be combined with methods like mixed effects to account for the limitations of nested variables.
Multilevel modelling. The data is nested in six different years, 27 different agencies, and >130 offices within those agencies, and so mixed effects, also known as hierarchical linear modelling, is appropriate. Mixed effects regression uses fixed effects to estimate independent variables of interest, along with random effects to estimate the nesting variables, in this case offices, agencies, and years. Mixed effects modelling is considered the gold standard when accounting for nested data (Wooldridge, 2013). Initial mixed-effects regressions reveal that the data are not conducive to this kind of analysis. All initial mixed effects regressions run for this study have drawn convergence errors. Two methods for fixing convergence errors are standardization of variables and experimenting with alternative optimizer algorithms. Both methods have been attempted with no success. While multilevel and Bayesian modelling analyzed data that was aggregated to the unit-level, they were not capable of analyzing individual-level data, and so are not reported in Chapter Four.


71
Stage One Limitations
This study has several limitations and they are heightened by the fact that race and gender is subject to the epistemological blind spots discussed in Chapter One. Therefore, it is important to carefully consider limitations of the data, as well as of understanding, both of which apply to this research.
Some limitations with the data are discussed above in the analysis subsection (Table 3.6). A related limitation is caused by the fact that the data are longitudinal over six years from 2010 to 2015, and so autocorrelation is a threat to validity. Speaking to that threat, OPM (2015) collects an inclusion index based on questions in the FEVS that capture fairness, openness, cooperation, support, and empowerment. OPM (2016) reported that the federal government “inclusion index” increased by 57 percent, and “global satisfaction” increased by 60 percent from 2010-2015. This relationship may indicate autocorrelation by showing that both the inclusion index and global satisfaction are correlated over time, indicating that the time period of an observation may be distorting results. Methods that help delineate nested effects and causal effects will be used, especially mixed-effects regression and propensity score matching (Wooldridge, 2013).
A second type of limitation is in how the FEVS captures Masculinity and Whiteness. If Whiteness and Masculinity are performances as hypothesized here, then anyone can perform them regardless of their identified race and gender. While selecting “white, non-Hispanic” on a survey may indicate a propensity for performing more consistently with Whiteness, it does not necessarily indicate that the person will perform Whiteness such that it changes the person’s PSM or job satisfaction. Even more troubling is the FEVS use of the term “Male” to capture man. When males are openly performing as transgender women, the term male could capture the


72
exact opposite of Masculinity. The masculine act of selecting “Male” on a survey is probably correlated to masculine performances that will affect job satisfaction and PSM, but the overlap is smaller than many assume. This set of limitations is conservative (increases type II error), making it less likely that the study will find that race and gender affect PSM and satisfaction.
The third set of limitations deals with external validity, and applies to all quantitative efforts to capture social interactions like PSM, satisfaction, race, and gender. Quantitative methods are theoretically generalizable to a population, in this case public organizations. However, as a study becomes more generalizable, the finding itself tends to become less meaningful (Riccucci, 2010). It is important that studies such as this link to context developed through qualitative methods.
The large N of the dataset (>2,000,000 observations) creates another challenge to credibility that is complicated by the subject matter. Greenwald, Banaji, and Nosek (2015) argue that their research on implicit bias related to race and gender have wide effects on society, despite “Statistically Small Effects” in quantitative tests. They argue that the universal nature of race and gender on all individuals in the United States means that small explanatory power can have important impacts on anyone who has a race or a gender. This argument dovetails with the critical race and feminist perspectives that race and gender impact every activity in social life, so even aggressions that are “micro” have powerful society-wide effects (Sue, 2010). These insights unmask a fundamental challenge to quantitative analysis of race and gender phenomena. Every independent variable and every dependent variable capturing some aspect of social life (e.g. managerial race and job satisfaction) are always already effected by race and gender. Therefore, race and gender are a common cause in every model, and every statistical test is at risk of type II error (Gerring, 2012). In this reality of ever-present endogeneity, large-N tests such as this create


73
an opportunity to discover these “small effects” that “have societally large effects” (Greenwald et al., 2015).
In the context of race and gender, literature on large-N datasets take on a different light. Kline (2004) convincingly discusses the limitations to methods like those used in this study, and argues that replication and Bayesian analyses are more valid for understanding social reality, reflecting the assertions of Gill and Meier (2000) that public administration research is overly reliant on null hypothesis significance tests. Part of the problem is that large-N datasets are oversensitive, creating an increased risk of type I error. Like the limitation stated by Greenwald et al. (2015), this limitation is unavoidable in the current study. It should be noted, however, that the type I error of this study’s large-N analyses is offset by the type II error derived from the common cause inevitable in models of race and gender mechanisms. All of these limitations are such that the analysis will be strengthened by the profound understanding that is possible through qualitative methods.
Using Stage One Findings to Situate Stage Two Findings
Three statistics obtained in the stage one analysis were used to guide sampling for the stage two research. All agency offices analyzed in stage two were within 1.96 standard deviations from the 1) mean proportion of white men employees to white men supervisors, and 1.96 standard deviations from the mean estimated differences between predicted and actual organizational 2) PSM and 3) job satisfaction. Because we know that the mean proportion of white men employees to white men supervisors in the FEVS is 7.7 percent with a 3.6 percent standard deviation, we can calculate that an acceptable organization for stage two analysis were between 14.75 percent and .64 percent too many white men supervisors. Similarly, the qualitative sample consisted of organizations that were within 1.96 standard deviations of the


74
PSM and job satisfaction that was predicted for a similar organization in the stage one findings. This created a more generalizable base for the qualitative stage two findings.
Stage Two: Institutional Ethnography of Federal Offices Stage two will explore eight propositions using institutional ethnography to discover and analyze “institutional relations of power” (Campbel & Gregor, 2002, p. 15; D. E. Smith, 2005). Table 3.7 states the propositions, along with the types of performances needed to analyze them.
The primary assumptions of institutional ethnography are...
(1) .. .people organize around social interactions that can be mapped and analyzed (D. E. Smith, 2006),
(2) social interactions communicate power over people (D. E. Smith, 2005),
(3) the everyday world contains problematics that can be explicated (D. E. Smith, 1987),
(4) tracking work processes can illuminate institutional realities (Campbel & Gregor,
2002).
Whiteness and Masculinity performances, like social interactions, can be mapped and analyzed using institutional ethnography. Whiteness and Masculinity performances communicate and reinforce power for whites and men in public organizations. To track how social interactions influence individual behaviors, institutional ethnographers focus on routine practices, methods, and procedures (Campbel & Gregor, 2002, p. 78). Based on the literature, this study was predicated on the assumption that race and gender performances are social interactions that can be identified, mapped, and analyzed using institutional ethnography (Smith, 2005; S. Turner, 2006).
The assumption that white supremacy and patriarchy are powerful forces in public organizations operated as the problematic for this study (D. E. Smith, 1987). In institutional


75
Table 3.7
Stage two propositions
Proposition Performances
Individual Proposition #1- Public servants ’ daily tasks include performances of Whiteness and Masculinity. All Individual-level Performances
Individual Proposition #2- Public servants make sense of race and gender through performances of Whiteness and Masculinity. Discriminatory framing, blindness, displacement, minimization, projection, microaggressions, innocence, noblesse
Organizational Proposition #1- Public organizational structures allocate power and resources to whites and men. Technocracy, adhocracy, expert/non-expert dichotomy, power shifting, strong vertical bureaucracy
Organizational Proposition #2- Public managers implement Whiteness and Masculinity performances as operational processes and systems. Dress, speech, mannerisms, private life taboos, touchstones, blindness, innocence, noblesse, discriminatory framing
Organizational Proposition #3- Public organizational cultures reflect Whiteness and Masculinity performances. Touchstones, dress, speech, mannerisms, capitalism, consumerism
Organizational Proposition #4- Public organizations with relatively fewer performances of Whiteness and Masculinity will be more welcoming to people of color and women. All performances, job satisfaction of women and people of color
Societal Proposition #1- Public organizations and public servants refer to society-level Whiteness and Masculinity to resolve conflict. All societal-level performances, blindness, minimization
Societal Proposition #2- Public organizations and public servants refer to society-level Whiteness and Masculinity to maintain their status with other organizations, public servants, and individuals in society. All societal-level performances, blindness, minimization


76
ethnography, a problematic is a “latent puzzle” that drives the research (Campbel & Gregor, 2002, p. 47). This focus on making aspects of everyday life puzzles to be solved is a means of “subverting institutionalization” (Campbel & Gregor, 2002, p. 124) by making institutions apparent. In the current study, the institutions of Whiteness and Masculinity, normally veiled by miscognition (Mills, 2007) were subverted by focusing on their performances as social interactions that were used to dictate power in typical organizations in the federal government. Case Selection: ‘Typical’ Organizations in the Federal Government
After the estimations from the quantitative results are assessed, case selection for the qualitative study will begin. The stage two sample was eight organizations, studied through interviews with 14 federal workers in Denver, Colorado and Washington, D.C. Half of the informants were white men in seven different agency offices. These seven informants worked in offices that results from stage one indicated were “typical cases” (Henry, 1990) as described in more detail above. Gathering typical cases facilitated more effective triangulation by identifying similar cases to achieve saturation more efficiently (Ulin, Robinson, Tolley, & McNeill, 2002). The informants were worked in non-military agencies, were permanent employees, and were not be engaged in abnormal activism, like association with either antiracist or white supremacist groups.
The seven white men informants were with seven people who are either women, people of color or both. These seven confirmatory informants were selected based on their ability to comment on the same organization as the white men informants and provide a position for triangulation (Esterberg, 2002). Race and gender research indicates that choosing informants who are not white men increases the likelihood of observing Whiteness and Masculinity more clearly (Hooks, 1984; Thandeka, 1999).


77
The researcher found informants by leveraging personal connections among federal workers in Denver and Washington, D.C., and by using snowball techniques to expand to other organizations and individuals. Snowballing occurred during correspondence before and after the interview when informants were asked to list other individuals within the organization and in other agencies that were willing to participate. After completing at least seven interviews in Denver, the researcher organized a one-week trip to Washington, D.C. and establishing seven interviews there and in rural Virginia.
Data Collection: Interviews and Social Media Observation
Interview guide. The interview guide was designed to capture the performances underlying the eight propositions, without leading informants to any answers. In line with institutional ethnography (Campbel & Gregor, 2002), the first part of the interview guide was focused on the daily logistics of the informant’s work tasks, looking at concrete actions that the worker takes, why each task is done, and on whose authority. For example, organizational proposition #4 (Table 3.7) states that managers use Whiteness and Masculinity operational processes, indicating performances like technocracy and bureaucracy. This proposition can be captured by asking the informant to talk about tasks and goal setting.
The second section of the interview guide focused on the organization-level propositions by asking the informant about their experiences with race and gender in the workplace. The last section of the interview focused on the individual, with questions about the informant’s feelings on discrimination, affirmative action, and diversity management. This last section was designed around the interview questions used by Matias and Mackey (2016) to capture Whiteness in students taking a teacher training course. The section included questions like “Should your office hire more white people or men?” to explore individual proposition #2 (Table. 3.7) stipulating that


78
Masculinity and Whiteness are used to make sense of race and gender in the workplace. This example question encourages informants to discuss their feelings about race and gender, allowing them to express blindness, but also encouraging them to process intersectional Whiteness and Masculinity. The chances that the researcher will observe institutional performances increased when the setting in which the interviews are collected is comfortable for the informant.
Collecting data. Each interview was conducted in a location of the informant’s choosing, including the informant’s office, quiet coffee shops, or reserved temporary spaces. The interviews were recorded using a small, unobtrusive digital recorder. The digital recording was transcribed soon after the interview using speech recognition software designed for court reporting (Dragon Professional, 2017) with the researcher listening to insure accuracy and add notes on context including surroundings, and mannerisms. After each interview, the researcher added to a digital journal, and followed up to collect supplementary data.
Supplementary data. When the interview was complete, the researcher tracked down documents and social media posts to contextualize the interview responses. Forms, computer applications, regulations, and other documents used by the informant in their daily job were collected using email requests, agency websites, and public records requests when needed. These texts provided context and triangulated interview data.
The researcher requested permission to view the informants’ publicly shared social media posts including Twitter, Facebook, MySpace, Reddit, Vine, Instagram, and other sites where the informant communicates. As noted by Frisoli (in Marhsall & Rossman, 2011), social media is now “a central feature of contemporary social life...” (p. 25). Most public administration research on social media has used the data quantitatively by coding postings made by public organizations


79
(M. D. Jones, Shanahan, & McBeth, 2014), or analyzing how public organizations interact with the public through social media (e.g. Pelizza & Hoppe, 2015; Rethemeyer, 2006; Yetano &
Royo, 2017). A search of several public administration journals including Public Administration Review, Journal of Public Administration Research and Theory, Administration and Society, and Administrative Theory and Praxis revealed no studies that used internet sources as data for institutional ethnography. Relevant to this research, Bryer and Zavattaro (2011) introduce a special issue of Administrative Theory and Praxis by documenting how social media impacts organizational culture and administrative culture. This study looked to other fields that have used social media as ethnographic data for analyzing social interactions and power relations (Beneito-Montagut, 2011; Garcia, Standlee, Bechkoff, & Cui, 2009; Hallett & Barber, 2014).
Given the proportion of social interactions that now take place on the internet, observing the internet footprint of organizations and individuals is a necessary part of modem ethnographic research. In the words of Hallett and Barber (2014), online “spaces have become important for the creation and reproduction of relationships, identities, and social locations.” Hallett and Barber (2014), Garcia et al. (Garcia et al., 2009), Beneito-Montagut (2011), and Marshall and Rossman (2011) all argue that context from digital social interactions is necessary for understanding face-to-face social interactions in the real world. Of these researchers, Beneito-Montagut (2011) most deliberately brings together internet and face-to-face interactions, finding that the combination creates a more complete understanding of the worldview of individuals. By comparing internet derived data to in-person interviews, this study sought to provide an understanding of the value of internet data in thick descriptive studies.


80
Stage Two Analysis: Coding, Hermeneutics, Mapping, and Ethnomethodology
When the data were collected, the researcher used Tables 2.1-2.3 to code by Masculinity and Whiteness performances. The analysis of qualitative data for this study used three related tools; hermeneutics, institutional ethnography, and ethnomethodology.
For coding, Tables 2.1-2.3 provided concrete performances of Whiteness and Masculinity in public organizations, operating like a decoder ring of systemic patriarchy and white supremacy. The researcher coded the data into the performance categories listed in Tables 2.1-2.3. As informants reported performances, the researcher analyzed how these performances appeared across public organizations in the study. At the end of the coding process, the researcher confirmed some of the performances listed in the literature review, provided deeper understanding of those performances (Chapter Five), and analyzed each of the eight propositions to make suggestions for public managers looking to manage Whiteness and Masculinity in their organizations (Chapter Six).
As discussed above, interviews were the primary sources of data for the qualitative stage of this study, while important texts such as forms and regulations, as well as internet ethnography provided further means of triangulating, challenging, and providing context for the interview data. Interviews represented informants’ organization rather than informants themselves. As the organization is the unit of analysis for institutional ethnography, texts, interviews, and internet data were coded as representing their associated organization and analyzed together.
To decrease bias, and deepen the analysis, the research engaged in hermeneutic practice. When a conflict arises between the perspective of researcher and informant, the hermeneutic researcher identifies the assumptions that, if changed, would resolve the conflict between the


81
perspectives (Debesay, Naden, & Sletteb0, 2008). This practice of understanding a conflicting perspective enables the researcher of racial and gender phenomena to understand how both researchers and informants make sense of race and gender (Matias, 2016). Simply identifying the conflict, however, does not resolve conflicting perspectives.
The standard suggested by many hermeneutic philosophers for resolving conflicts between informants, researchers, and research recipients is to defer to informants as experts in their own worldview (Debesay et al., 2008; Gadamer, 1976). However, critical hermeneutic researchers analyze conflicts in understanding created by prejudices and discrimination that limit informants’ ability to understand the world around them (Mills, 1997; Zuberi & Bonilla-Silva, 2008). Critical scholars argue that the normative values determining understanding in most studies are those that support Whiteness and Masculinity (Pini & Pease, 2013; Zuberi & Bonilla-Silva, 2008). In other words, those studies that do not deliberately select race and gender objectives, instead select white supremacy and patriarchy by default.
Stage two resolved conflicts in understanding by deference to transparent assumptions that white supremacy and patriarchy inflate the value of whites and men (L. A. Bell, 2013;
Harro, 2013). After each interview, and at least weekly during the analysis process, the researcher recorded a journal identifying conflicts of understanding with informants, findings, and the literature, and attempting to resolve those conflicts based on this assumption of equality. This journal served as a tool for identifying coding bias, and develop increased reliability.
After the interviews and supplemental texts were coded, the study developed an institutional map of Masculinity and Whiteness performances. To better understand how institutions embed in public organizations, Smith (2006) suggests using S. Turner’s (2006) method of mapping social relations. These maps document work processes by tracking the


82
multiple sources of meaning and knowledge moving through the organization. This study focused on the “work” of reinforcing and maintaining Masculinity and Whiteness. In Chapter Six, the institutional map is presented and analyzed for places where the work of Whiteness and Masculinity can be disrupted or modified.
To understand how informants created meaning around race and gender, the research adopted an ethnomethodological perspective. Ethnomethodologists analyze how people make sense of institutions by inserting a “breach” of the institutional environment (Liberman, 2013). In GarfinkeTs (1963) early studies, the institutional breach could be something like answering “Yes” to the question, “Can you hand me that paperclip?”, but not complying with the request. These breaches were meant to force informants to process their everyday practices (Arminen, 2008; Pollner, 2012). As the informant reconstructs their institutional reality, the ethnomethodologist collects data on the process through which reality is reconstructed and reasserted (Garfinkel, 1963; Pollner, 2012). The method used by Matias and Mackey (2016) and Bonilla-Silva (2003) is understood in this study as an example of ethnomethodology. In these studies, an outwardly racial or gendered question in an environment of race- and gender-blindness creates a “breach” of institutional norms (Mills, 1997), and the informant must then reconstruct their institutional reality to create the least possible dissonance. This study used ethnomethodology to interpret the “work” that informants perform when they recreated race and gender after being asked questions that challenged race- and gender-blind beliefs. Thus, ethnomethodology was used to bring out performances of Whiteness and Masculinity for coding.
Once the data are coded into race and gender performances, the resulting coded data and organization maps were used to analyze each of the eight propositions from Chapter Two. While this proposal outlined 38 Masculinity and Whiteness performances, the analysis process revealed


83
relevant performances that were not initially theorized. These performances were initially coded separately, coded emergently, and used to analyze the eight propositions accordingly.
Eventually, enough data was collected to give evidence supporting, partially supporting, or not supporting each of the propositions, creating a deeper understanding of the mechanisms that develop and reinforce Masculinity and Whiteness in the federal government.
Ethical Considerations
Like all research with human informants, this study has some minor risks to participants that should be acknowledged. Words like racism, sexism, white supremacy, and patriarchy are technical terms in critical whiteness and multiple masculinities research, but are loaded terms to most people. It is this same loaded terminology that could produced a slightly larger risk for participants in this study.
Miscognition surrounding Masculinity and Whiteness creates an additional ethical dilemma, which comes from the fact that informants responded with hostility to terms like Whiteness and Masculinity, thereby making the findings less beneficial. This study attempted to get informants to speak about race and gender in the workplace without triggering their defensive performances. To achieve this delicate balance, the researcher presented the research to informants as focusing on “race and gender in the federal sector,” rather than Whiteness and Masculinity. This technically correct statement provided the necessary information to informants without risking undermining the results that have the potential to benefit all people, including white men in public organizations.
All documents were encrypted using the encryption software, Boxcryptor (“Boxcryptor.com,” 2017), and only the researcher has access to the decryption key. Texts and social media printouts that included identification of individuals were modified so that faces,


84
names, and other identifying characteristics were blurred or encrypted. The encrypted data were stored on a secured CU Denver cloud drive, which is accessible only to the researcher. Nobody apart from the researcher had access to the raw data or the encryption key. All results were presented anonymously.
Stage Two Limitations
The qualitative portion of the research is limited in three main ways. First, “typical cases” case selection is often criticized because of the heterogeneity of populations. In the words of Henry (1990, p. 21), “typical cases” case selection “invites close inspection of the cases selected, and often the credibility of the findings is tarnished when a case appears atypical.” In this case, typical cases case selection was aided by the quantitative data confirming that the organizations selected fit the profile of a typical office in the federal government.
A second limitation derives from the assumption that a person who is not a white man will act as a reliable check on how white men understand organizational race and gender performances. The literature indicates that women and people of color must adopt performances of Whiteness and Masculinity to survive in modern society (Carbado & Gulati, 2013; Ferguson, 1984). While using a second informant verified the experiences of the white men informants, there is no guarantee that these individual women and people of color performed Whiteness and Masculinity differently than their white men colleagues, and therefore no guarantee that data from women and people of color informants provided meaningful triangulation of data.
Another limitation that needs mentioning is the threat to validity caused by President Trump. Between the data collection of the FEVS (OPM, 2016) in 2004-2015, and the qualitative data collection in 2017-18, President Trump’s election changed how federal workers think about their job and its race and gender performances. While it is too soon for academic research to


85
have been published, evidence of a possible shift can be found in Trump’s actions and in the news media. Trump appointed more white men to his cabinet than any president in modern history (Lee, 2017). These appointments include Steven Miller, who attacks the evils of “foreign labor” and bemoans “political correctness” as coddling women and people of color, both performances of Whiteness documented in the literature (Thrush & Steinhauer, 2017). Two more of Trump’s early appointments included Stebastian Gorka and Steve Bannon both of whom left Breitbart News to work for the white house (and have since returned to Breitbart). Breitbart News is a website that is described by Bannon as a “platform for the alt-right” (Fortin, 2017).
The term alt-right was coined by white supremacist Richard Spencer as a way to join a “movement of consciousness and identity for European people in the 21st century” (Florido, 2016). The alt-right is also vocal about the proper position of women as housewives and objects for men to lust after, arguing that the reproductive organs of white women should be under the control of white men (Darby, 2017, p. 26). Another one of Trump’s appointments, Attorney General Jeff Sessions was rejected by the 1986 Senate for a judgeship in part because of his racist opposition to protections to black voting rights (McCann, 2017). Now able to secure his appointment, Sessions pursued a Justice Department investigation of universities’ affirmative action practices under the racial projection claim that whites are being harmed by admission policies (Savage, 2017) despite the continued over-representation of whites in colleges across the country (Ashkenas, Park, & Pearce, 2017). And signaling a masculine hurdle to women who want to work in the Trump administration, Vice President Mike Pence will not have a meal with a woman unless his wife is also present (Parker, 2017). To account for the fact that these examples now represent the leadership in the federal workforce, this study included a question about what has changed in the office since Trump was elected.


86
Conclusion
This study has implications for public administration research on race, gender, diversity management, representative bureaucracy, and social equity. The Critical Whiteness and Multiple Masculinities literature on which the assumptions of this study rely is often either theoretical or experimental. Very little of the extant research is situated in a public organization, and very little uses institutional ethnographic methods to understand race and gender as the consequences and causes of social interactions in organizations. Methodologically, this study adopts these methods, not because they are new and underutilized, but because methodologists in the social sciences are convincing that these methods are the best suited for understanding the impact of institutions embedded in organizations. In countering the institutional miscognition surrounding Whiteness and Masculinity, it is often necessary to use non-standard methods because these same institutions were involved in the development of standard methods (Pini & Pease, 2013; Zuberi & Bonilla-Silva, 2008). The methods described here were selected to expose the institutional effects of Whiteness and Masculinity on public organizations and provide a foundation for making recommendations to manage those effects.


87
CHAPTER FOUR
MASCULINITY AND WHITENESS IN FEDERAL WORK UNITS
The quantitative findings reveal an outline of Whiteness and Masculinity that can be unexpected and nuanced. For example, clustered logistic regressions reveal that Whiteness in management has a more profound impact on satisfaction than Masculinity, but that both institutions impact public service motivation (PSM). Additionally, machine learning analysis of work units reveals central relationships between Whiteness and alternative work arrangements, and Masculinity and merit systems that are deemphasized in the literature.
This chapter begins with an exploration of the Federal Employee Viewpoint Survey using factor analysis and comparative regression analysis between people who perform Whiteness and/or Masculinity by indicating that they are whites and/or men, and those who do not. The next section looks at the impact of overrepresentation of whites and men in management on satisfaction and public service motivation (PSM) among federal workers, directly addressing the hypotheses stated in Chapter Three. After that, a machine learning tool reveals correlations that other methods sometimes miss. The decision trees discover interesting relationships between alternative work arrangements and race, and abstract objectivity and gender that will be explored further using qualitative methods in Chapter Five.
Summary Statistics: Disentangling Masculinity and Whiteness
Factor analysis revealed that the job satisfaction variable initially created in Chapter Three, and which is used in some academic articles using the FEVS (e.g. Weaver, 2015), was more defensible as two different variables. Initially, the job satisfaction variable captured overall employee satisfaction by indexing seven questions. This index had an acceptable alpha of .884. Factor analysis after varimax rotation revealed that job satisfaction fit better as two indices. The


88
first variable is a job satisfaction index including seven questions getting at employees’ satisfaction with training, promotional opportunities, recognition for doing a good job, involvement in decision making, that their talents are well-used, likelihood of recommending the agency to others seeking work, and overall satisfaction. This “job satisfaction index” has a higher alpha of .915, with an Eigenvalue of 4.189.
The second factor discovered in this process included seven questions about employees’ perceptions of satisfaction with supervisors support of work/life balance, commitment to representative bureaucracy, respect for employees, opportunities to demonstrate leadership, supervisors’ listening abilities, supervisors’ respect for employees, employee trust in supervisors, and immediate supervisors’ quality of work. This second variable captures satisfaction with management with an alpha of .953 and an Eigenvalue of 5.278.
Both dependent variables were of interest for this study. As described in Chapter Two, studies indicate that the race and gender of management likely impacts the job satisfaction of employees, but little is known about whether these employees are satisfied over their working conditions or the quality of their management.
Job and Management Satisfaction
White men are on average less satisfied with their jobs than white women and men of color, but not women of color (Table 4.1). Fixed effects were used to predict the means of each dependent variable categorically by the intersectional race and gender of the respondents. The models included control variables for respondents over 40 years old, military status, and whether they have been in government for more than 10 years. The four categories were white men, white women, men of color, and women of color, with clustered standard errors to account for unit-level effects, and fixed effects to account for time. This analysis indicates that the mean


Full Text

PAGE 1

MASCULINITY AND WHIT ENESS IN PUBLIC ORGA NIZATIONS: MANAGING INSTITUTION AL RACE AND GENDER I N FEDERAL AGENCIES by Nuri Heckler B.A., City College of New York, 2005 J.D., University of Denver, 2008 A Dissertation submitted to the Faculty of the Gr aduate School of the University of Colorado in partial fulfillment of the requirements for the degree of Doctor of Philosophy Public Affairs 2019

PAGE 2

ii © 2019 NURI HECKLER ALL RIGHTS RESERVED

PAGE 3

iii This dissertation for the Doct or of Philoso phy d egree by Nuri Heckler has been approved for the Public Affairs Program by Mary Ellen Guy, Chair John Ronquillo Lonnie Schaible Camilla Stivers Date: May 18, 2019

PAGE 4

iv Heckler, Nuri (Ph.D., Public Affairs Program) Whiteness and Masculinity in Publ ic Organizations: Managing Institutional Race and Gender in Federal Agencies Dissertation directed by Professor Mary Ellen Guy ABSTRACT This study examines the institutions of Whiteness and Masculinity and their influence on public organizations. Whiteness and Masculinity are institutions, collections of rules, norms, and culture that benefit whites and men. These institutions are under researched because they are epistemologically veiled behind color and gender blindness. Nevertheless, they disproportiona tely influence public organizations including the federal government that has been dominated by white men in all three branches, and historically studied by mostly white men scholars. This study unveils Masculinity and Whiteness by identifying performances of the se institutions that are likely to appear in public organizations and using those performances to explore two research questions. 1) What effects do Whiteness and Masculinity have on public c managers, and public administrators perform Whiteness and Masculinity. The research was conducted in two stages. In the first stage, publicly available data was analyzed using clustered fixed effects on white/men/white men job satisfaction, management sa tisfaction, and public service motivation when a larger proportion of their managers are also whites/men/white men. The quantitative data are then analyzed using machine learning tools to explore relationships between myriad independent variables , and sati sfaction and public service motivation for whites/men/white men compared to their people of color /women colleagues. In the next stage, the quantitative data identify work units that are typical of the federal government for in depth interviews to conduct

PAGE 5

v i nstitutional ethnography. The findings indicate that many of the performances of Masculinity and Whiteness documented in the literature are present in federal work units, and that new performances have evolved that are meaningfully impacting federal servic e. Examples include the myth of a generation gap that is justifying inaction especially among younger federal workers, and a complex relationship between color blindness, alternative work arrangements , and camaraderie. From these data, a map of institution al Whiteness and Masculinity is developed that is leveraged to make suggestions for better managing the race and gender institutions dominant in most public organizations. The form and content of this abstract are approved. I recommend its publication. App roved: Mary Ellen Guy

PAGE 6

vi This dissertation is dedicated to Micky, Luca, and Pippin w ho serve as my examples of meaningful and authentic engagement with justice and equality . May this work share som e of our love with the universe.

PAGE 7

ACKNOWLEDGEMENTS W ithout Dr. Mary E. Guy, public administration research and practice would be far less interesting and just. Her contributions to this dissertation literally changed my life and made every part of this research possible. Her model of courage and uncompromis ing good humor in the face of serious issues is reflected in diminished form in this dissertation . I also need to thank Dr. Cheryl Matias whose leadership, mentorship, and radical mothering challenged me to better understand myself as a white man. Many of my law professors influenced this dissertation, including Robert Chang, Catherine Smith, Rashmi Goel, Andy Reid, and the late Ann Scales who I believe is still having fun and raising hell. Of the many mentors that I have had at CU Denver, Dr. John Ronquill o, Dr. Lonnie Schaible, Dr. Seba Bishu, Dr. Tanya Heikkila, Dr. Todd Ely, and Dr. Jessica Sowa left thei r thumbprints all o ver this research. Each of these experts imparted wisdom and knowledge, and I would never have been introduced to any of them without the investment that the School of Public Affairs made in my training. Thank you all. Thank you also to the anonymous informants. Your willingness to participate and your openness was both unexpected and refreshing. I wish I could get to know all of you be tter. I also need to thank Dr. Camilla Stivers and Dr. Brandi Blessett who modeled careers and ways of thinking about public administration that made me feel I could bring my true self to my research and pursue it conscientiously a nd without apology. Along with the Critical Perspectives CIG at ARNOVA, th ey convinced me that academia was a life that could be lived authentically driven by hard work, intelligence, and mercy . Last but not least, thank you M om. I am who I am because of you.

PAGE 8

TABLE OF CONTENTS CH APTER I . . 1 . . 2 . . Whiteness and Masculinity Research in Public Administratio . 11 18 II .. 20 The Institutional Fr 21 Public 2 5 .. . 2 7 .. 2 8 Society .. . . 30 Individual level Whiteness an d Masculinity.. .. . .3 7 Organizational level Whiteness and Masculinity .. .4 6 5 3 . .. 5 7 III RESEARCH . 5 8 5 8 Hypotheses: Race and Gender Affect Job Satisfaction and Public Service Motivation 60 Data : the FEVS . 60 Independent Variables: Race and G 2

PAGE 9

ix Dependent Variables: PSM and Job . 6 4 . ... 6 5 . .6 7 Stage One Li mitatio ... . 1 Using Stage One Findings to Situa .. . 3 ..7 4 6 Data Collection: Interviews and So 7 Stage Two . 8 0 ...8 3 S tage Two . . 8 4 Conc 8 6 IV. QUANTITATIVE FINDING S 7 8 7 Job and Management Satisf 8 Public Service Motivat 90 Additional Independent Var 91 Individual level Logit Ana 4 Decision Trees: Machine Learning about Whi teness and Mas . .. 101 C 3 V. QUALITATIVE FINDINGS 4 Diversity Management, Color blindness, and Altern 6 Masculinity, (Whiteness), and Abstrac 8

PAGE 10

x Other Perfor mances of Whiteness and Masculi . ... 13 4 . 1 5 1 4 VI. CONCLUSION 5 Managing Whiteness and Masculinity in Pu 5 ... 15 6 6 0 Managing Linkages . 6 1 Managing Perfor man 4 7 0 Conclusion: Discovering and Pulling Social Justice Levers in Public Organizatio 7 1 REFERENCES ... . 17 3 APPENDIX 195

PAGE 11

xi LIST OF TABLES TABLE 2.1: Society 2.2: Individual level performances of Whiteness and Masculi 8 2.3: Organizational ..4 7 . 61 . 6 1 3.3: Inde 6 3 6 ....6 6 3.6: Analysis strategies for overcoming limitations in 8 3. 7 5 4.1: Unit clustered categorical predicted means after controlling for time, age, military status, 9 4.2: Unit clustered categorical predicted means after controlling for time, age, military status, and tenure in federal governme 9 4.3: Individual level logit regressions on J 6 4.4: Individual level logit regressions on manage 8 4.5: Individual level logit regressions on ... . 1 00 4.6: Table of confirmation or rejection of hypotheses 10 2 . 1 5 2

PAGE 12

xii LIST OF FIGURES FIGURE 2.1: Model of institutional effects of Whiteness and Masculinity on public organizations . 2 2 4 3.1: Model of the relationship between stage one 9 5 6.1: Map of Whiteness and Masculinity Performances 7 A1 : Unit level ANOVA regression tree on Comparative White J ob Satisfaction 19 5 A2 : Unit level ANOVA regression tree on Comparative White Management Satisfaction 19 6 A3 : Unit level ANOVA regression tree on Comparative White 19 7 A4 : Unit level ANOVA regression tree on Comparative Sati sfaction 19 8 A5 : Unit level ANOVA regression tree on Comparative Management Satisfaction 19 9 A6 : Unit level ANOVA regression tree on Comparative 200 A7 : Unit level ANOVA regression tree on Comparative White Sati sfaction 201 A8 : Unit level ANOVA regression tree on Comparative White Management Satisfaction . . 2 02 A9 : Unit level ANOVA regression tree on Comparative White 2 03

PAGE 13

1 CHAPTER 1 THE PERSONAL IS PUBL IC A DMINISTRATION (Hanisch, 2006) , and critical race . Following these cues , this dissertation argues that the personal is public administration . B ehaviors of race and gender learned in childhood homes, primary and secondary education, and adult relationships influence how public administrators do their jobs. For an outsized number of public administrators, Whiteness and Masculinity are their prim ary race and gender influences. This dissertation focuses on a significant gap surrounding Whiteness and Masculinity in public administration res earch. The gap is highlighted by considering two findings side by side. (1) Representative bureaucracy studies organizational outcomes (Selden, 1997) . (2) The white racial identity and the masculine gender identity are the most represented in many if not most American public organizations (Guy & Newman, 2005) . overarching research question fits in this gap. H ow do Whiteness and Masculinity affe ct public organization s? For public administration scholars, Whiteness and Masculinity are particularly salient and problematic. Representative bureaucracy s cholars find that teacher gender affects how well girls perform on math tests in public schools (Keiser, Wilkins, Meier, & Holland, 2002) , police officer gender affects willingness of victims to report domestic violence (Theobald & Haider Markel, 2009) , an d spokesperson gender affects willingness of citizens to cooperate with municipal recycling programs (Riccucci, Van Ryzin, & Li, 2015) . Teacher race affects whether children of color are referred to gifted and talented programs (Elhoweris, Mutua, Alsheikh, & Holloway,

PAGE 14

2 2005) (Pigott & Cowen, 2000) . W elfare officer race affects interactions with recipients (Watkins Hayes, 2011) , and physician race affects the belief that patients are telling the truth about reported levels of pain (Meghani, Byun, & Gallagher, 2012) . If public administration is a reflection of institutions (Wamsley et al., 1990) influenced by white men like the founders of the nation (Hamilton, Madison, & Jay, 1894; Rohr, 1990) , the white men who originally set out to self consciously study American public administration (Hoffman, 2002; W. Wilson, 1887) , and 44 of 45 American presidents, then influence is necessary for understanding public organizatio ns. This chapter describes the blind spot around Whiteness and Masculinity in public (1997) concept of the epistemological racial contract, and nervousness as a mechanism of that epistemol ogy of ignorance (Gooden, 2014; Mills, 2007) . The next section begins illuminating that blind spot by elucidating Whiteness and Masculinity. The last section outlines the implications of Masculinity and Whiteness research for research on social equity, rep resentative bureaucracy, diversity Whiteness and Masculinity. Epi stemological and Empirical Gaps There is a gap in thinking, an epistemological gap around Whiteness and Masculinity , which in turn causes an empirical gap, a gap in knowledge and understanding. Take some examples from the representative bureaucracy litera ture. Keiser et al . (2002) find that girls improve math test scores when in contact with women who are successful in math. If the study

PAGE 15

3 was focused on Masculinity, the findings could just as easily indicate that something about the way men teach math is hu rting the potential math scores of their girl students. Similarly, Selden (1997) finds that bureaucrats of color are more likely than white bureaucrats to provide subsidized loans to farmers of color. Focusing on Whiteness, the finding could just as easily have been that white bureaucrats make loan determinations that discriminate against people of color ( people of color ). Representative bureaucracy scholarship has made substantial contributions in empirical findings about how people of color and women impa ct the public organizations in which they work. However, because of epistemological gaps around Masculinity and Whiteness , little is known about the impact of the race and gender that dominate many if not most public organizations. E pistemological gaps ar e features of Whiteness and Masculinity rather than academic oversight. Mills (1997) argues that race is a political, moral, and epistemological contract. The t society (Mills, (Gooden, 2014) (W. Ayers, 2004) (Foldy & Buckley, 2014) (Sue, 2004) . Of all the racial identities, Whiteness is the most susceptible to misinterpretation (Mills, 1997; Yancy, 2008) . Whiteness is commonly conceived only as non Otherness ( Fanon, 1963; Yancy, 2008) , a neutral blank slate against which to perceive other races as inferior (Mills, 2007) . The epistemological contract makes discussion of Whiteness taboo (Carbado & Gulati, 2013) , and contributes to the gap surrounding Whiteness as an academic study (Zuberi & Bonilla Silva, 2008) . The epistemological gap surrounding Masculinity is similar to the gap surrounding

PAGE 16

4 Whiteness. Collinson and Hearn (1996) Masculinity and the workpla vii). Ferguson (1984) maintains the patriarchic bureaucracy, while Stivers (1993) (p. 5) of masculine public administration maintains its legitimacy. Like Mills (1997), these scholars find that misinterpretation is a key feature of Masculin ity, indicating a gender epistemological contract working alongside the racial contract. If there are racial and gender epistemological contracts, then social scientists are signatories to these agreements. Zuberi and Bonilla Silva (2008) edit a volume ded icated to understanding how the assumptions of most social science methods are based on a logic of white racial superiority. In their chapter, Zuberi and Bonilla Silva (2008, pp. 330 331) point out that most analyses of race and gender focus on women, blac ks, and workers as problems injected into a n experimental laboratory that is assumed raceless and genderless . Zuberi (2000) details the history of sociology from the perspective of a quantitative sociologist of color, finding that many of the basic statist ical regression methods used today were deemed reliable on field tests proving the basic tenet of eugenic science that white people are superior . Similarly, Pini and Pease (2013) edit a volume critiquing the masculinity of modern methods and masculine scie ntific theorizing. Pini and Pease (2013) their findings. Relatedly , Stivers (2000) argues that researchers have mostly abandoned questions and methods pioneered by women to the detriment of both women and public administration research. Meanwhile, Gooden (2014) nervousness, and Alexander and Stivers (2010)

PAGE 17

5 administration reflects assumptions of profes sional neutrality and innocence, and a lack of gendering of social science research. Just as a student of biology must stain a slide with iodine before examining a microscopic cell, Whiteness and Masculinity researchers must deliberately color their world view. Du Bois (1980 [1898] ) laments that sociology focuses on how the problems of black America impact the critical whiteness studies (CWS) and multiple masculinities (MM) scholars shift perspectives by explicitly problematizing Whiteness and Masculinity (D. L. Collinson & Hearn, 1996; Harris, 1993) . Like staining a m icroscopic slide, this shift in perspective is a deliberate attempt to mark the unmarked Masculinity and Whiteness that pervade daily life (Frankenberg, 2001; D. E. Smith, 1987) . This marking of phenomena is an important step toward filling the gaps surrou nding Whiteness and Masculinity. Marking Masculinity and Whiteness To mark Whiteness and Masculinity as affirmative notions rather than the absence of otherness means starting with clear definitions (Bar On, 1993; Frankenberg, 2001; Goertz, 2006) . To that end, Whiteness is a collection of performances that reinforce white supremacy. Masculinity is a collection of performances that reinforce patriarchy. These definitions are simple, of course, and the rest of this section will deconstruct what is meant by co llection of performances and by white supremacy and patriarchy. Whiteness and Masculinity are collections of performances (De Beauvoir, 2009; Du Bois, 1935) (Frankenberg, 2001) and Mas culinity (Carrigan, Connell, & Lee, 1985) , and that is also an accepta ble way of

PAGE 18

6 thinking about them. However, t concepts of gender and race as constant and ongoing collections of performances (F rankenberg, 2001) , rather than a construction that once created remains relatively static. The other reason to think of Masculinity and Whiteness as collections of performances is because no individual performance alone can maintain patriarchy and white su premacy, but many performances combined become self reinforcing (Sherman, 1996) . Thus power recreates itself by associating seemingly unrelated performances like supervisorial authority with lighter skin tone, or expertise wi th golfing and larger shoulders . White men, then, are not born white and men, but made into w hite men through years of trained and reinforced race and gender performances (De Beauvoir, 2009; Foucault, 1977; Pascoe, 2007; Thandeka, 1999) . The Whiteness and Masculinity collection s of perf ormances are individual, Whiteness and Masculinity starts at birth. In the first six months, caregivers spend less time making eye contact with babies wearing blue t han with babies wearing pink , and are more likely to let babies in masculine colors babies wearing feminine colors (Eliot, 2009) . These findings are particularly significant in a society where the first question commo nly asked of a parent is (Eliot, 2009) . Males cry slightly more than females in infancy partially due to the fact that they are, on average, smaller and less fully formed at birth, but by five years old males have learned to be boys and are significantly less likely than girls to cry in experimental conditions (Eliot, 2009) . By high school many boys showily express callousness and willingness to act violently for fear that they will be shamed as homosexual or inadequately masculine (Pascoe, 2007) . Due to ongoing racial segregation in housing and schooling (Edsal & Edsal, 1991; Orfield, Frankenberg, Ee, & Kuscera, 2014) ,

PAGE 19

7 phenotypically white babies are unlikely to encounter a significant number of people of color for the first decade of their lives (Helms, 1990) . When white children encounter people of color , Thandeka (1999) argues that their curiosity is often rebuffed by the adults they trust, leading them to avoid speaking about race and feel nervous when the topic comes up (Matias, 2016) . A s adults, white men are more likely to have implicit biases that benefit whites and men (Banaji & Greenwald, 2013) , while speaking and acting in ways that implicitly and/or explicitly prioritize white and masculine experiences (Bonilla Silva, 2012; Cooper, 2000) . By adulthood, men are less likely to express emotion of any kind, except anger (Bancroft, 2002; Katz, 2013) . Whites too are less likely to show emotion, except for happiness, especially in the workplace (Roediger, 1999) . Being a white American mean s having an upper middle class Midwest suburban demeanor, accent, and appearance including dress and lighter skin tone (small w whiteness) (Carbado & Gulati, 2013) . Men similarly are expected to be less emotional, more rational, more focused on work than f amily, and have physical attributes including wearing pants and suits rather than skirts and dresses (small m masculinity) (Burris, 1996; Glauber, 2008; Pascoe, 2007) . These prescriptions of dress, behavior, demeanor, and psychology are part of collections of performances that ultimately sum up to Masculinity and Whiteness in individuals. These individual performances happen within organizations that are also performing Whiteness and Masculinity. T he level of analysis that is of primary concern to this stud y is the organization. The literature on Masculinity in organizations exposes many aspects of management and bureaucracy as innately masculine and even patriarchic (D. L. Collinson & Hearn, 1996) . Multiple Masculinity (MM) organizational theorists cite Kan ter (1977) as their intellectual foremother, who finds that men maintain power without having to act individually, but through organizational culture and structures. As a recent example, Henderikse et al. (2017)

PAGE 20

8 find that organizational culture is more imp ortant for achieving gender parity in organizational staffing than even human resources policies. Organizational cultures reinforce Masculinity when creating social opportunities focused around masculine recreational activities (Kanter, 1977) like golf (Ca brera, 2016) (Weiss, 2014) . Kanter (1977) points out that secretaries, who are mostly women, are rewarded for their fealty to their bosses who are mostly men and discouraged from maneuvering to protect themse lves or seek advancement. This division continues in the present day with a new polarization of gendered occupations (Burris, 1996) , to the extent that men are unwilling to train for jobs considered feminine even in an economy where they are more plentiful than traditionally masculine jobs (Miller, 2017) . Interestingly, men of color are far more likely than white men to take on those traditionally pink collar jobs (Miller, 2017) . Like Masculinity, Whiteness is a workplace norm in many if not most organizati ons. Whiteness traits such as American mid west, suburban speech, behavior, dress, and mannerisms are norms in most workplaces (Foldy & Buckley, 2014; Lipsitz, 2009) . Any transgression from these established workplace norms by people of color coworkers is viewed as deviant. Carbado and Gulati (2013) give the example of co workers sharing stories about watching the sunrise after a long night in the office. White employees are assumed to be dedicated and hardworking, while people of color employees must defen d themselves against accusations of laziness, bad planning, or even throwing a party in the office (Carbado & Gulati, 2013) . Another example of (2015) finding that Muslim American women feel less job satisfaction when the y wear hijab than when they do not, even after accounting for religiosity. Directly addressing the organization as a moderator of the impact of race, Kochan et al. (2003) find that most of the effects of racial diversity found in the business literature ar e moderated by

PAGE 21

9 organizational factors such as culture, communication, and job satisfaction. Individuals perform Whiteness and Masculinity as soloists, and simultaneously for and with organizations as big bands. Society also performs constantly evolving Mas culinity and Whiteness . Sociological literatu re refers to these collections of performances as institutions (W. R. Scott, 2005) . These performance collections are similar to critical concepts of the hegemonic discourse (Foucault, 1977) (1997) r acial contract. F or all members of American society assumptions that Whiteness and Masculinity performances are collected in certain configurations guide individual and organizational actions . Much of the CWS and MM literature is from this perspective. CWS scholars find that Whiteness slowly evolved from the 1600s through a series of legal, economic, and social movements (Roediger, 1999) (1944) , known as the GI Bill of Rights, provided benefits to put veterans of World War II back to work, giving ethnic whites, like Jews and Italians who were substantially more likely to receive honorable discharges than blacks , the leg up that they needed to move from a pseudo Whiteness to full participation in the dominant ra cial category (Brodkin, 1994) presidency and color blindness combined to reproduce Whiteness creating optimal circumstances for a backlash against people of color (Bonilla Silva & Dietrich, 2011) . Meanwhile, American Masculinity evo lves to protect the power of men by polarizing the business leaders extol fluid organizational structures for technical experts of software engineers who are most ly men (Cooper, 2000), the mostly women clerical and frontline staff are subject to Panopticon seem unobtrusive (Burris, 1996). This new technocratic Masculinity (Burris, 1996)

PAGE 22

10 strength, and rational behavior to establish power, all traits associated with men. White supremacy and patriarchy are the allocation of power to people who most convincingly perform Whiteness and Masculinity respectively. Power refers to the directing of resources by controlling the behavior of others (Foucault, 1977; Marx, 2004) . The power of Whiteness and Masculinity is often a conservative power to restrict und erstanding and discussion of race and gender oppression structures, and therefore stop progress before it starts (Bachrach & Baratz, 1962; Lindstead, 2000; Mills, 2007) . This conservative power is how segregation became the norm in the South over the cours e of many decades of so called "separate and equal" jurisprudence (Du Bois, 1980) . More than simply maintaining the status quo, this conservative power slowly erodes dissent as dominant players find ways to recreate the status quo to benefit themselves (D. Bell, 1980) . When conservative power fails, some turn to aggressive power to maintain the position of the powerful (Foucault, 1977) . This happened when the Brown v. Board of Education decision brought a serious critique of the separate but equal doctrine to the center of American politics (D. Bell, 1980) . No longer able to maintain the status quo, many whites turned to an aggressive response including church bombings, murders, and brutal policing of blacks in cities across the country (D. Bell, 1980) . Thes e two faces of power (Bachrach & Baratz, 1962) keep people from deviating from Whiteness and Masculinity through conservative discipline, while providing an aggressive punishment when deviance does occur (Foucault, 1977) . These forces help maintain advanta ges for men and whites. Some may object to the terms white supremacy and patriarchy because they are linked to anachronistic bigotry. This study chooses these terms for two reasons. The first is that historical

PAGE 23

11 bigotry did not disappear to be replaced by t he modern allocation of power to benefit whites and men. Modern white supremacy and patriarchy is a continuation of an American history of oppression (Bonilla Silva, 2012; Stivers, 1993) . The second reason is that the terms accurately depict an ongoing emp irical reality where men and whites continue to dominate most social, political, and religious life in America (Lipsitz, 2009; Status of women in the States, 2016) . Whiteness and Masculinity work in tandem to allocate power to white men. Crenshaw (1991) co ins the term intersectionality to describe how race and gender intersect to reproduce racism and sexism against women of color who are victims of battery and rape. Referring to a similar concept, hooks (1984) argues that feminist researchers have struggled not understand fully the inter relatedness of sex, race, and class oppression or refuse to take this inter These studies show that Whiteness and Masculinity interact in complicated ways that may contribute to one or both concepts. To date, little is known about how Whiteness and Masculinity interact and conflict. Whiteness and Masculinity Research in Public Administration Most of the research around race and gender in public administration focuses attention on those who are harmed by Whiteness and Masculinity, rather than the phenomena doing the harm. Social equity (Gooden, 2014) , representative bureaucracy (Selden, 1997) , and diversity management (Riccucci, 2002) studies mostly highlight the problems that ar ise when women and people of color interact with public administration either as bureaucrats in or clients of the system. These insights are invaluable for casting light on race and gender in public sculinity can complement these programs by exposing the mechanisms of power that create discriminatory outcomes. This study also positions itself within institutional research on the power of society and organizations to

PAGE 24

12 control the choices of individuals (W. R. Scott, 2005) , and makes contributions to that literature by detailing specific performances that reproduce power. Social equity . Research on Masculinity and Whiteness potentially influences a great deal of social equity research. Social equity is de fined as a more flexible measure than equality, (Guy & Mccandless, 2012, p. S5) , and understanding equality in a context of previous injustice. Gooden (2015) argues the current era is the fifth epoch research significantly informs four out of five of these epoch s. The first two epoch (Gooden, 2015, p. 211) . If, as is argued in this study, Whiteness and Masculinity influences every aspect of life and government in America, both concepts are an important part of the context for equality. Masculinity and Whiteness research can contribute to understandin g the c ontext of social equali ty as masculine and white . While Gooden (2015, p. 211) argues that (1977) and Ferguson (1984) find that the role of masculinity in organizations is just as deeply entrenched and both formal and informal. Harris (1993) finds that Whiteness is enforced partially through a set of formally recognized property rights that undermine race equality, while Roediger (1999) , Brodkin (1994) , and Ignatiev (1995) d ocument the American history of Whiteness through the evolving definition of who is white. Similarly, understanding how Whiteness and Masculinity reproduce power by separating we . In particul ar, Gooden (2015) argues that public administrators themselves, and those who research public administration, seek to better understand who are. This task requires understanding the

PAGE 25

13 racial and gender identities of whites and men who make up an over re presentative portion of public administration managers and researchers. Whiteness and Masculinity research has some catching up to do when it comes to understanding the context and identity of social equity scholarship on the cutting edge of (2015 ) final two epoch (2015) fourth epoch seeks to understand why inequity persists. Of all the big questions of social equity, this is the one that Masculinity and Whiteness studies address most directly. As defined above, Whiteness and Masculinity are collections of performances that reproduce and maintain white supremacy and patriarchy (D. Collinson & Hearn, 2003; Ignatiev, 1997) . Inequity persists because we have yet to dismantle social institutions like Whiteness and Masculinity. This leads dire (2015) contemporary epoch which looks at how accountability for social equity can be achieved. This study addresses this by identifying the individual performances that maintain and reproduce white supremacy and patriarchy, thereby teeing up future research and experimentation on developing accountability mechanisms for these performances. A rich understanding of Whiteness and Masculinity in public organizations contributes to the social equity research agenda by overcoming nervousness to reveal how the personal lives of administrators in their homes, their childhood, and their relationships contributes to the (2014, p. 4) kely to occur in a serious way if organizations are nervousness can be understood as a mechanism of power that maintains the racial epistemological contract (Bachrac h & Baratz, 1962; Mills, 1997) . Martin (2000) argues that the contract extends to gender in the form of taboos that reinforce patriarchy in organizations.

PAGE 26

14 Understanding nervousness as a conservative power mechanism indicates that nervousness will be more l imiting around discussions of Whiteness and Masculinity, because to maintain power these collections of performances must continue to appear as natural and inevitable consequences of small w whiteness and small m masculinity (Burris, 1996; Marx, 2004; Mill s, 1997) . Public administrators learn nervousness in their private lives from their closest friends and family, even when they do not condone prejudice (Bonilla Silva, 2015) . Therefore, changing management, leadership, and the organization means focusing o n the private life as a factor impacting public (Gooden, 2014, p. 4) , taking nervousness head on by identifying what Whiteness and Masculinity performances reproduce inequali ty, and how to observe those performances in public organizations. Representative bureaucracy and diversity management . Like this dissertation, much of representative bureaucracy and diversity public management research assumes that the personal lives of p ublic administrators, namely their race, gender, and other identities, influence how services are provided. Representative bureaucracy assumes that government operates differently when public administrators demographically reflect the people they serve (Se lden, 1997) . One thing that representative bureaucracy gains from research of Whiteness and Masculinity is a path to pragmatic solutions short of replacing white and man bureaucrats with people of color and women. This is an intractable problem for represe ntative bureaucracy scholars ever since the original texts lamented that public administrators could never fully represent the poor and unemployed because they, by definition, have jobs and income (Kingsley, 1944) . While the modern era has witnessed a dram atic increase in the diversity of public administration, representativeness is still a long way off (Guy & Newman, 2005) . Representative bureaucracy findings provide few tools for public managers looking for solutions short of

PAGE 27

15 replacing large portions of t heir white men employees with women and people of color. This is particularly troubling in light of a recent finding that human resources policies had no impact on creating gender equality after controlling for factors like organizational culture, diversit y knowledge, and leadership (Henderikse et al., 2017) . Research on Whiteness and Masculinity can bridge the gap between representative bureaucracy and diversity management by providing solutions related to the performances that create the discriminatory ou tcomes discovered by representative bureaucracy researchers. Whiteness and Masculinity research illuminates how seemingly neutral behaviors in organizations have unintended discriminatory outcomes. Researchers find that work hours can reproduce white supre macy (Carbado & Gulati, 2013) and patriarchy (Cooper, 2000) , staffing choices are polarized by race and gender (Burris, 1996) , and fluid organizational structures help maintain power in upper level managers still dominated by white men (Burris, 1996; Coope r, 2000) . The causes of discrimination in many organizations are less often bigotry and more often unconscious bias (Banaji & Greenwald, 2013) or unintended consequences (Yancy, 2008) . While researchers focus on managing diversity in the workplace (Riccucc i, 2002) , these performances are found to be more common in workforces without any diversity whatsoever (Carter, Helms, & Juby, 2004) . For diversity management and representative bureaucracy, Whiteness and Masculinity are levers for understanding and chang ing what is happening in the organizations that the increasingly diverse workforce is entering (Riccucci, 1997) , thereby addressing the question of how to recruit diversity into organizations that are not yet diverse. Another contribution to representative bureaucracy is Whiteness a help making sense of passive, active, and symbolic representation. Representative bureaucracy scholars find that simply hiring people of color and women is not necessarily sufficient to create

PAGE 28

16 representati ve outcomes (Riccucci & Van Ryzin, 2017) . The act of including people of color and women in the bureaucracy is called passive representation, while outcomes that are less discriminatory signal that there is some kind of active representation (Riccucci & Va n Ryzin, 2017) . If simply seeing a people of color or a woman in a role is enough to help a beneficiary trust the bureaucracy, that is called symbolic bureaucracy (Van Ryzin, Riccucci, & Li, 2016) . Conceptualizing race and gender as performances, as do Du Bois (1903) and De Beauvoir (2009) simplifies some of the differences between passive, active, and symbolic representation. If organizations are normally performing in ways that conform to Whiteness and Masculinity, when people of color or women enter such an organization their performances will change the way the organization operates and simultaneously conform to the organization. However, this passive representation will not translate as well to active representation if the new people of color and women organizational members are trained to perform the Whiteness and Masculinity dominant in most organizations. The conceptualization of race and gender as collections of performances is a challenge to the idea that representation can ever be truly passive. A basic assumption of this study is that race and gender are not firm demographic identities, but performances. Based on this assumption, scholars find that people of color and women perform Whiteness and Masculinity as a strategic attempt to get ahead, and/ or as a result of oppression (Hanisch, 2006; Neville, Coleman, Falconer, & Holmes, 2005) . When people of color and women enter organizations performing Whiteness and Masculinity, they perform race and gender, but not necessarily the race and gender researc he r s assume they will perform. For example, Carbado and Gulati (2013) discuss the pressure on people of color to conform to displays of Whiteness in the workplace, while Martin (1990) and Cooper (2000) find that women struggle to conform to Masculinity.

PAGE 29

17 Wh at has been called passive representation, then, is the active representation of Whiteness and Masculinity. But this is not necessarily bad news for achieving a representative bureaucracy . Just as people of color and women can perform Whiteness and Masculi nity, whites and men can potentially dismantle Whiteness and Masculinity when performances are identified and changed. Findings from contemporary sources indicate that many whites and men wish to help end discrimination ( NBC News/Wall Street Journal Survey , 2016; Stepler, 2016) . CWS and MM research provide tools for acting on that desire. Institutional public o rganization t heory. By theorizing Whiteness and Masculinity as institutions, this study will contribute to institutional theory by exploring the exte nt to which organizational environments consist of distinct overlapping institutions. Unpacking that sentence, organizational institutional theory originated from attempts to understand how the world around an organization, the environment in which it sits influences the organization itself (Lawrence & Lorsch, 1967; W. R. Scott, 2001) . This resulted in a number of studies finding that organizations are influenced by their environments in myriad ways (e.g. Bozeman, 1987; DiMaggio & Powell, 1983; Meyer & Rowa n, 1977; Ostrom, 2005; Roy, 1997; Stivers, 2000; Uzzi, 1999) . The foci of those studies were on how changes in rules impact i nstitutions in ways that change many organizations at once (Roy, 1997; Uzzi, 1999) , or on how the entire collection of institutions , the institutional arrangement taken as a whole operated on a small subset of organizations (Ostrom, 2005) . Looking at Whiteness and Masculinity as distinct but related makes it possible to understand performances as parts of embedded institutions. This u nderstanding enables researchers to identify embedded Whiteness and Masculinity distinctly from one another, and determine better how power is acting on individuals in organizations as compared to other similar organizations.

PAGE 30

18 This study also contributes by incorporating insights from CWS and MM research into institutional theory. If Whiteness and Masculinity can be conceptualized as institutions (Sherman, 1996) , then CWS, MM, and related researchers have been analyzing these institutions for more than a cen tury. One example of potential contribution is Derrick that marked the beginning of critical race theory. In that article, D. Bell (1980) describes the decline in school integration from the time of Brown v. Board of Education (1954) through the maintenance of conservative power. This leads D. Bell (1980) to argue that racism is permanent and evolving, reflecting the arguments of institutional scholars who find that institutions maintain power through evolutions that can be traced usin g historical data (Dobbin, 2005) . The insights of CWS and MM researchers into the embeddedness and power of race and gender in organizations and individuals represents a contribution to institutional thinking that stands to inform analyses of other institu tions such as capitalism (Weber, 1972) and consumerism (Campbell, 1987) . In other words, Whiteness and Masculinity are just two sources of performances that limit individual behavior, and focusing on how these two embed in organizations will help scholars understand how other institutions influence public organizations and other action arenas. Conclusion: Research on Whiteness and Masculinity in an Institutional Framework bring with them Whiteness and Masculinity from the most personal sectors of their lives including unconscious biases learned in the cradle, the classroom, and intimate relationships. This argument has the potential to impact representative bureaucracy rese arch, social equity theory, diversity management, and institutional organizational studies. The rest of this study fills in this argument, starting in chapter two with the development of a theory of Whiteness and

PAGE 31

19 Mascu linity in public organizations. The th ird chapter outline s a method for gathering and analyzing data on Whiteness and Masculinity performances in the federal government , and tracking their impact in organizational structure and culture, in the behavior of managers, and in the actions of street level bureaucrats. Chapter Four reports the findings of a quantitative study using the Federal Employee Viewpoint Survey (2016) n. In Chapter Five, the qualitative findings elucidates how white men make sense of their race and gender while performing public service. In the concluding chapter, an institutional map reveals strategies for public servants to manage Masculinity and Whi teness in public organizations .

PAGE 32

20 CHAPTER TWO INSTITUTIONAL WHITEN ESS AND MASCULINITY: A THEORY OF INTERSECTIONALITY Chapter O ne describes race and gender as personal behaviors initially learned in infancy pole by arguing that gender and race are global geological forces that originate in the cretaceous period when dinosaurs roamed the shores of North America. Some of those shores were covered in plankton that flo ated in the coast off what is now the southern United States (McClain, 2012) . These plankton left deep rich soils in a crescent from the coas t of Virginia, through Atlanta to Bolivar County on the banks of the Mississippi. 65 million years later, people fr om Africa were kidnapped and forced to cultivate cotton grown in the rich soils the plankton left behind. In 2016, crescent of counties along that ancient coastline where black vot ers who descended from those enslaved people voted for the democratic candidate (CNN, 2016) . This example illustrates how the personal is impacted by large forces (Roberts, 2014) , and also how factors like race are fecund variables for explaining large sca le human behavior. view it is impossible to begin from either psychology or economics and explain everything in a one way causation. Instead, the relations of society m (Sherman, 1996) . Said another way, race and gender emanate simultan eously from microscopy of a single individual, and from the panorama of society . Performances at all levels influence and are influenced by rules, norms, and culture s that dictate meaning for social life, what sociologists call institutions (W. R. Scott, 2001) . These institutions create a framework that affects every

PAGE 33

21 aspect of public organizational life, and individuals in public organizations. Whiteness and Masculini ty are two such institutions. To analyze these institutions, t his study will investigate two related secondary research questions. (1) What effects do Whiteness and Masculinity (2) How do public organizatio ns, public managers, and public administrators perform Whiteness and Masculinity? To answer these research questions, this chapter develops a theory of Masculinity and Whiteness in public organizations (see F igure 2. 1 ) . The first section develops the insti tutional framework of which Masculinity and Whiteness are two institutions. The next section substantiates a number of performances that are influenced by Whiteness and Masculinity at the society , individual, and organizational levels, laying out propositi ons for study in the process. At the end of that section, the literature on public service motivation and job satisfaction establishes dependent variables for quantitative testing of the impact of Masculinity and Whiteness in public organizations. The last section of the chapter is a short conclusion. White Men in the Institutional Framework The Institutional F ramework Whiteness and Masculinity are two institutions in the institutional framework surrounding public organizations. Institutions are collection s of rules, norms, and cultures that dictate meaning for social life (W. R. Scott, 2001) . Rules refer to formal rules that are passed through deliberative bodies or carry the force of law (Ostrom, 2005; W. R. Scott, 2001) . Norms are shared understanding ab out the best or most correct way of doing things (Meyer & Rowan, 1977; Ostrom, 2000; Rousseau, 1896) . Norms include ethical and moral beliefs that are shared by a society, such as the protestant ethic valuing hard work (Weber, 2004) or the romantic ethic v aluing beauty and aesthetics (Campbell, 1987) . Norms also include what Ostrom (1990, p. 72)

PAGE 34

22 Figure 2.1 : Model of institutional effects of Whiteness and Masculinity on public organizations

PAGE 35

23 in T ogether, norms and rules make claims about how individuals s hould act when they are sharing r esources in a society. Culture is a lens for making sense of the world in a particular time and place (LeCompte & Schensul, 2010) . Weick (1995) calls culture a synonym for paradigm (p. 120), an interpretation of reality that help s diverse people share their understanding of the world. Thus, shared artifacts . Just as the Rosetta stone includes three translations of a unique text, cultur e is the expression of unique values through multiple stories, symbols, statements of belief, holidays, media, and other experiences where things, actions, and words (performances) take on emotional meaning (Schein, 2010) . When these performances take on e motional meaning, they have value so organizations work to guard them (Weick, 1995) . Culture, rules, and norms make up the institutions that hold organizations and societies together (W. R. Scott, 2001) . Whiteness and Masculinity are two institutions that operate in an institutional framework surrounding public organizations (see Figure 2 .2 ) (Sherman, 1996) . Other institutions in the framework includ e capitalism, bureaucracy (Weber, 1972) , consumerism (Campbell, 1987) , technocracy (Burris, 1993) , Blackness (Fanon, 1963) , Femininity (Bartky, 1990) , and many others. An institution embeds in an organization when a performance from that institution occurs in the organization (Granovetter, 1985) . For example, the institution of consumerism includes a norm for aes thetic beauty that may be performed by constructing a sleek open floorplan in a computer store, or an eclectic treasure hunt in an antique shop (Campbell, 1987) . These aesthetics are performances of consumerism that are embedded as artifacts in the compute r store and the antique shop. At first, these performances appear disparate , but when taken in their

PAGE 36

24 Figure 2.2. Masculinity and Whiteness in an institutional framework.

PAGE 37

25 institutional context they can both be understood as bound together by the institution of consumerism (Campbell, 1987) . Institutions and organizations change. Similar to Darwinian adaptation, organizations inherit institutional traits or performances (Blyth, Hodgson, Lewis, & Steinmo, 2011; Fürstenberg, 2016) . For example, managers working in corner offices on higher floors is a performance of the insti tution of capitalism (Schein, 2010) . However, adhocracy is replacing bureaucracy as an overlapping institution in many modern capitalist organizations (Burris, 1993; Mintzberg, 1983) . Adhocracy is defined by symbols of equality (Mintzberg, 1979) , at least among expert classes (Burris, 1996) . This emphasis on equality has led to the design of capitalist offices that no longer adhere to previous spatial norms (Cooper, 2000; Schein, 2010) . Assigning corner offices to managers is still a performance of capitali sm, but so too is a newer open floorplan (De Croon, Sluiter, Kuijer, & Frings Dresen, 2005) . If, over time, the new open floorplans become more common than cellular floorplans with reserved corner offices, that would be an example of the institution of cap italism evolving (Fürstenberg, 2016) . Like capitalism, Whiteness and Masculinity evolve (Brodkin, 1994; Ignatiev, 1995; Roediger, 1999) . Importantly, institutional evolution is often circular and regressive (Davis, Diekmann, & Tinsley, 1994; DiMaggio & Pow ell, 1983; Fürstenberg, 2016; Sherman, 1996) . Many scholars have argued that current institutional trends in public organization are both circular (e.g. Goodsell, and regressive (E ikenberry, 2009; Ingraham, 1993; B. Sandberg, 2016; Stivers, 2008) . Public Organizations Public organizations range a wide s pectrum from socially responsible business (Bartley, Koos, Samel, Setrini, & Summers, 2015) and social enterprise (Kerlin, 2006) , th rough

PAGE 38

26 community nonprofit organizations (Never & Westberg, 2016) and social welfare nonprofits (Romzek & Johnston, 2005) , to traditional government agencies (Weber, 1972) . What unites these organizations as being public is that they ha ve public inputs and/ or public outputs. The input that delineates public from private organizations is political authority (Bozeman, 1987) , while Moulton calls t he public organizational output realized public value (Moulton, 2009) . While scholars point out that both political authority and realized public value can be outputs as well as inputs, thinking of them this way helps clarify the definition of publicness. Bozeman (1987) separates organizational inputs into economic and political authority that are used to create goods a nd services. Economic authority is the ability to barter and trade for resources, while p olitical authority is legitimacy brought to an organization by popular or democratic means (Lindblom, 1977) . One recent example of the use of political authority is so buy their products (Bartley et al., 2015) . A company like Patagonia uses their ethically sourced manufacturing model to encourage customers to choose their products over oth er products that may be less expensive (Bartley et al., 2015) . Therefore, a socially responsible business is able to convert a political input into added outdoor clothing they manufacture and thei r profit, and the realized public ness is the high paying jobs and reduced environmental impact. Moulton (2009) argues that the realized publicness output is as important as political authority in changing the character of an organization along the spectrum from most private to most public. What is less known is how these public inputs and outputs impact race and gender performances within individuals associated with public organizations along the spe ctrum. Public organizations are particularly important con tributors to Whiteness and Masculinity

PAGE 39

27 even as they are constrained by Whiteness and Masculinity. As described in further detail below, public organizations created modern Whiteness, and reinforce both Whiteness and Masculinity through laws, regulations, a nd the distribution of services based on the social construction of individual citizens and beneficiaries (Heckler, 2017; Mackinnon, 1989; Roediger, 1999; Schneider & Ingram, 1993) . Meanwhile, as described above, white men have outsized influence on almost every aspect of public administration from the founding of the country up to today (Witt, 2006) . As public organizations stretch further into the private sphere, the influence of Whiteness and Masculinity grows because the private sector has a larger prop ortion of white men, and white men dominate the executive apexes of private corporations, especially in Fortune 500 companies that are more likely to be run by white men today than in recent decades (Berman, 2015; Guy & Newman, 2005; Lindzon, 2016) . Perhap s most importantly, public organizations are uniquely positioned to affect social equity (Gooden, 2014) , and Whiteness and Masculin ity constrain efforts of equity minded public servants. Emotionally B ounded I ndividuals The more emotional investment culture can create in the performances valued by an institution, the more power the institution has over individuals within it (Weick, 1995) . Because values are created by emotional investment, the bounded emotionality model of the individual proposed by Mumby an d Putnam (1992) provides a useful concept for making sense of how individuals interact with institutions. Emotionally bounded individuals are driven initially by in (Mumby & Putnam, 1992, p. 474) . This (Mumby & Putnam, 1992, p. 474) ,

PAGE 40

28 which can be understood as the pressures placed on the individual by relationships including institutions, organizations, friendships, and familial ties. The boundedly emotional individual (Mum by & Putnam, 1992, p. 475) to create shared experiences and share their values with others. Thus, the balance between individual values and group values is not solely a self serving calculus, but a question of giving what they are able while at the same ti me investing in something bigger than themselves (Mumby & Putnam, 1992) . Bounded emotionality recognizes as problematic the basic psychological assumptions of many of the bounded rationality theories of public administration. Before the boundedly rational actor can select a method of achieving goals and objectives based on the best available information (B. D. Jones, 2003) , the goals and objectives themselves have already been set by emotionality in an institutional framework (Mumby & Putnam, 1992) . These v alues are determined by boundedly emotional individuals influenced by institutions like Whiteness and Masculinity. Whiteness and Masculinity Whiteness and Masculinity are the institutions that maintain and reinforce white supremacy and patriarchy. Not all white men are the same, but Whiteness and Masculinity have institutional effects that can be observed through the regularity of white and masculine performances in which they engage. Multiple Masculinities (MM) scholars point out that men are unique and st ress that scholars should avoid essentializing (Boyle, 2002) , and Critical Whiteness Studies scholars argue the same is true of whites (Allen, 2008) . Thus, white men are not homogenous, but Whiteness and Masculinity are hegemonic institutions that influenc e white men and others (Connell & Messerschmidt, 2005) . These ideals are hegemonic because they are

PAGE 41

29 both unattainable and aggressively enforced (Pascoe, 2007; Thandeka, 1999; Wise, 2011) . All men and white people perform their Masculinity and Whiteness to different degrees along a spectrum, facing ostracism, and threats of violence for those men and whites who do not adequately perform (Allen, 2008; Ignatiev & Garvey, 1996; Majors & Billson, 1992; Pascoe, 2007) . Whiteness and Masculinity are almost always impossible to attain, and literature documents how feelings of inadequacy related to this catch 22 become rage, shame, violence, and tears (e.g. Andersen, 2016; Bancroft, 2002; Bonilla Silva, 2003b; Cooper, 2000; Katz, 2013; Matias, 2016; Thandeka, 1999) . (2017) finding that self induced mortality is increasing among white men due to drugs, alcohol, and suicide. While economists like Case and Deaton (2015, 2017) cite Murray (2012) to argue that economic inequality i s the primary driver of white self harm, their own research acknowledges that all other racial (except Asian) and gender groups earn less and have lower workforce participation, but yet are not seeing the same increases in self induced death rates. The exp lanation preferred by critical race and gender theorists is that the ideals of Whiteness and Masculinity are so unattainable and aggressively policed that white men who fail to approximate them suffer from their allegiance to their race and gender (Allen, 2008; Majors & Billson, 1992) . Where Herrnstein and Murray (1994) argue financial hardship is the result of cultural inferiority (see also Murray, 2012; Vance, 2016) , critical race and gender scholars argue tha t the concepts of superiority and inferiority themselves require redefinition (Butler, 1990; Fanon, 1963) . Given the power of this hegemony over white men, understanding behaviors in individuals requires understa nding system level performances .

PAGE 42

30 Society l evel Whiteness and Masculinity Society level performances of Whiteness and Masculinity fall into three main categories; economic, cultural, and political (see Table 2.1 ). Starting with the political, Du Bois (1903) argues that Congress granted the franchis e to black men in an effort to maintain moral authority substantially greater racial equality. In the latter half of the 20 th Century, the Republican party became t he home for whites who resisted black and brown demands for equality (Edsal & Edsal, 1991) . As a result, Edsal and Edsal (1991) find that whites vote consistently as a block in a way that indicates an aversion to black and brown people. Continuing this tre nd, white men and women voted against voters of color (and Jews) in every Presidential election so far this century (CNN, 2004, 2008, 2012, 2016; Kuhn, 2009) . Masculinity too performs in the electorate at the society level. Men make up over 75 percent of s tate legislatures, and higher proportions of U.S. Congress, Senate, governors and mayors (Center for American Women and Politics, 2017) . Over 94 percent of state legislators are white, men, or both. Just as black suffrage was enacted as a concession to Whi teness (Du Bois, r effort (Goldstein, 2001) . Research finds (Silverman, 1987, p. 190) in how they wish to achieve foreign policy objectives. Today, the gap between how men and women vote is larger than ever, but not as large as the raci al voting gap (CNN, 2016) . Both gaps are important for public organizations because politicians selected by these racialized and gendered voting processes set the tone for public service, and create laws that are artifacts of Whiteness and Masculinity.

PAGE 43

31 Table 2.1. Society level performances of Masculinity and Whiteness.

PAGE 44

32 The legal structure of the United States was originally based on assumptions that white men were full citizens, women existed in relation to them, blacks were their property, and all other races were their adversaries (Harris, 1993; Scales, 2006) . Le gal scholars argue that (1989) scathing critique of how rape laws contribute to a culture of sexual abuse against women. Similarly, Scales (2013) argues that gender b lind Masculinity prevents open debate of militarism, and contributes to the ever increasing power of the President . Whiteness was originally a legal construction with the introduction of laws classifying the children of black women slaves as chattel, there fore creating the original multigenerational race based slavery in the Virginia Colony (Roediger, 1999) . In modern law, Harris (1993) documents how Whiteness itself is treated as property through legal color blindness. Even in the case of affirmativ e actio n in university settings American law forbids race consciousness when used to right historical wrongs, but permits it when used for diversity that benefits white students (Harris, 1993; Regents of the Univ. of Calif. v. Bakke. , 1978) . Similarly, Lipsitz (2 009) documents the process through which white homeowners benefit as a result of color blindness in f ederal policymaking in the 20 th century. Even today, color blindness in a housing policy promulgated by the Obama administration is likely to benefit mostl y white homeowners in still segregated housing communities (Heckler, 2017) . Like elected officials, laws dictate the daily life of public administrators (Bertelli & Lynn, Jr., 2006) , and these critiques should make public administration scholars cautious a bout using law to provide foundations for the field that are neither racist nor sexist (Rohr, 1990; Rosenbloom & Piotrowski, 2005) . Even as political and legal performances continue, society creates and propagates a culture of Whiteness and Masculinity. Re call that culture is the process of creating meaning and

PAGE 45

33 value (Weick, 1995) . Media and art create meaning around Whiteness and Masculinity (Katz, 2013; Vera & Gordon, 2003) , such as common stereotypes that have been well documented in the psychological li terature, and that will be described further in the section on white and masculine psychology below. In addition to creating and reinforcing stereotypes, media communicate images of men as having a duty to unilateral action and strength by force (Gilligan, 1996, 2003) . Gilligan (2003, p. 1167) argues that the desire for masculin e noblesse oblige , and shame at not achieving it, is the fundamental driving force behind this call to violence. Katz (2013) and Scales (2009, 2013) similarly find that media and law create rhetorical connections between Masculinity, militarism, and nobles se . These m edia combine masculine noblesse with methods of organization common in the military, such as technocratic pragmatism (Burris, 1996; Ferguson, 1984) and top down bureaucracy (Katz, 2013; Scales, 2013) . Stivers (1993) argues these emphases on top down that of women, and finds this places women public servants in dilemmas for whi ch there are no easy solutions. If Masculinity values noblesse , Whiteness cult ure inflates the value of innocence. To Mills (1997) , the racial moral contract originates with the connection between lowercase w whiteness and innocence through the labeling of geography with moral value. Thus, Africa is (Mills, 1997, p. 50) . This labelling continues through racialized geographical titles like the foreignness of the barrio , the desperation of the ghetto, and innocence of the segregated suburbs

PAGE 46

34 (Le onardo & Hunter, 2009) . Baldwin (1962) refers to this innocence when he laments that it is innocence contributes to Whiteness in public service. The image of the well meaning white teacher naively but resourcefully rescuing classrooms full of black and brown students is a common cultural construct (Matias, 2013; Solorzano, 1997) . This storyline has profound impact on the jobs of teachers who come to be trusted by their students, putting them in a position to reinforce links between innocence and Whiteness for their students (Leonardo & Boas, 2013) . The outsized value of innocence and noblesse for white men combine with laws, rules, and media to create Whiteness an d Masculinity as valuable property creating wealth, income, health, and other outcomes for whites and men (Harris, 1993; Mackinnon, 1989) . For example white people are more likely to go to a school that is less than 25 percent black or brown than at any ti me since 1968 (Orfield et al., 2014) , and housing continues to be racially segregated (Lipsitz, 2009) (1991) finding that, like voting, white housing and schooling choices are partially motivated by racial aversion. Men contin ue to earn greater income than women, and whites earn more than all other racial groups except Asian Americans (B.L.S., 2017) . Meanwhile, the white wealth gap continues unabated as whites have more average wealth than all other groups including Asian Ameri cans (Oliver & Shapiro, 2006) . Whites have longer lifespans than brown and black people (Case & Deaton, 2017) , while studies show that shorter lifespans are positively correlated with masculine behavior even among females (Gilligan, 2003; Lippa, Martin, & Friedman, 2000) . One contributor to lower white mortality is that whites are significantly less likely to live in areas that have environmental hazards, and race is a stronger predictor of (Lipsitz, 2009) . Taken toget her, Masculinity and Whiteness have profound economic impacts.

PAGE 47

35 Very little has been written about the intersection between Whiteness and Masculinity at the society level, but some guidance can be gleaned from analysis of other intersections. For example, c ritical race scholars document that white women teachers resort to crying in efforts to gain sympathy and claim they are the victims of conversations exposing their racism, thereby using feminine vulnerability to enact their Whiteness (Matias, 2016) . Meanw hile, Majors and Billson (1992) find that some black men sacrifice heavily to uphold images of masculine noblesse (2003) understanding that noblesse is a defensive mechanism against fee lings of shame, the sacrifice black men make for their noblesse can be understood as an attempt to maintain Masculinity despite not having access to Whiteness. For white men, this is likely to mean that feelings of deficit, or shame at failing to attain id eal Whiteness could lead to more aggressive efforts to approximate Masculinity , while shame at l acking Masculinity could lead white men to place more emphasis on racial divisions . Contemporary commentaries including Vance (2016) and Andersen (2016) documen t violent aggression among mostly men in white communities resulting from perceived deficits, indicating that the same phenomenon at play among some black men is happening among white men turning to masculine noblesse to compensate for feelings of deficit in white innocence. Less documented is whether white men who have feelings of deficit in masculine noblesse turn to white innocence to compensate. Could this lead white men in public service to position themselves as martyrs in the same way that white wome n teachers play the victim to silence critiques from people of color (Leonardo & Boas, 2013; Matias, 2016) ? Could it create men who use their white innocence to inoculate controlling behaviors toward those over whom they have power in the same way as spous al abusers (Bancroft, 2002) ? There is not enough empirical

PAGE 48

36 literature on the intersection of Whiteness and Masculinity to venture a guess. The literature indicates that white men need both adversaries and objects of value, and the intersection of Whiteness and Masculinity provide both. White ness perceives the white race as the absence of color, a race that is identified by not being the other (Fanon, 1963) . Therefore, Whiteness cannot exist without black, brown, yellow, and red bodies (Fanon, 1963) . Masculi ne noblesse also relies on others. Noblesse requires an adversary to conquer and a feminine prize to win (Katz, 2013; Mackinnon, 1989; Pascoe, 2007) . By valuing white people as innocent (Mills, 1997) , Whiteness helps white men justify the targeting of peop le of color , whose imagined (Mills, 1997; Schneider & Ingram, 1993) . Similarly, this situation serves Whiteness by providing objects for white men to protect so they can justify the use of force without disrupting their white innocence. The two institutions then, like so many other overlapping institutions in the framework, are interdependent and cannot be understood without simultaneously considering the effects of both. Society level prop ositions . These institutional performances create pressures for public organizations to conform to Whiteness and Masculinity (DiMaggio & Powell, 1983) , and support Mas culinity and Whiteness are the societal norm, and they will subtly and overtly make reference to those institutions in their communications and actions to justify their work and guide their behavior. Societal Proposition #1 Public organizations and public servants refer to society level Whiteness and Masculinity to resolve conflict. Another proposition derives from DiMaggio and Powells (1982) concept of isomorphism that organizations in a society must conform to the institutional pressures of the organizat ions

PAGE 49

37 around them to obtain resources. Societal Proposition #2 Public organizations and public servants refer to society level Whiteness and Masculinity to maintain their status with other organizations, public servants, and individuals in society. Individ ual level Whitenes s and Masculinity White men train in Whiteness and Masculinity performances from the moment they are born and continue to enact those performances until they are buried under a headstone reading band and Father . In public org anizations individuals face Masculinity valuing tough guise (Katz, 2013) , a rejection of so called feminine emotionality in favor of rationality, calculation, and aggression (see Ta ble 2.2 ) (1992) cool pose communicate power. Researchers find that this aggressiveness is causally linked to the propensity to act violently and criminally (Bancroft, 2002; Gilligan, 2003; Katz, Heisterkamp, & Fleming, 2011; Klinesmith, Kasser, & McAndrew, 2006) . In professional white men, the tough guise mixes with white innocence to become the image of the noble ma n who hides his personal connections and emotions behind a veil of competence (Kendall, 2000) (Kendall, 2000) adopts toughness, not through displays of physical aggression, but through all (Cooper, 2000, p. 385) . These men spending long nights away their families dismiss private life concerns to avoid workplace taboos, or leverage those concerns to prove their willi ngness to sacrifice family for work (Cooper, 2000; J. Martin, 1990) .

PAGE 50

38 Table 2.2: Individual level performa nces of Whiteness and Masculinity.

PAGE 51

39 The tough guise combines with individual level Whiteness performances to give white men two layers of benefits. White racial performances include suburban mid west vernacular, dress, and mannerisms (Carbado & Gulati, 2013) . The clothing that is considered normal in the workplace more closely resembles clothing and hairstyles that are easily associated with white people than people of color , meaning that professional attire and hair goe s unnoticed for white people (Carbado & Gulati, 2013) , in much the same way that the appearance of men is less comment worthy than the appearance of women generally (Dellinger & Williams, 1997; Weitz, 2001) . Similarly, professionally recommended mannerism conversations in the manner of men, and pantsuits that on men are simply called suits (Chamberlain, 2017; S. Sandberg, 2013) . Masculinity and Whiteness both value hard work, but while Masculinity values aggress ive boastful martyrism to express noblesse, Whiteness values agreeable polite gentility to express innocence. Modern American Whiteness evolved out of the move from the traditional European agrarian societies to industrial factories (Roediger, 1999) . 19 th Century European Americans had advantages over blacks even when they first set foot in the United States (Guglielmo, 2003) , and each wave of immigrants achieved full Whiteness by participating in the economics of the newly forming spirit of capitalism and the industrial workforce (Brodkin, 1994; Ignatiev, 1995; Roediger, 1999; Weber, 2005) . W hite business owners engaged in the industrial revolution valued punctuality and agreeableness over trade skills and self regulation important in old world commerce (Ro ediger, 1999) . European immigrants adopted these polite performances as a cloak against anti immigrant sentiment, a means of proving that new i mmigrants were as white as the established U.S. power structure (Ignatiev, 1995; Roediger,

PAGE 52

40 1999) . Whiteness poli tesse and agreeableness, then, are defensive mechanisms for the newly white, reminding Guglielmo (2003, p. 41) of an Italian coal worker covered in soot who ran from a 19th century race riot s (1994) argues that whit es continued to perform that capitalist version of gentility even as the Jews entered Whiteness in the m iddle of the 20 th century 1 , and Bonilla Silva (2003a) argues that many Latinos are attempting to make the same pivot to white agreeableness and insoucia nce in the 21 st . There is not much known about how dutiful masculine noblesse and innocent white gentility intersect in white men public servants. Both images focus on exerting great amounts of effort on behalf of employers to serve the interests of organi zations, but many potential conflicts are suggested by the combination of aggressiveness and gentility. It is possible that white men managers use Whiteness performances to smooth over workplace conflicts, while bringing out Masculine performances to make authoritative decisions , thereby effectively embedding both institutions in ways that improve organizational effectiveness and efficiency, while hurting equity . Psychology of Whiteness and Masculinity . Like all people, white men have imperfect perceptions of reality. Kahneman and Tversky (1984) explain that thinking happens at two levels simultaneously. The intuitive system reacts quickly and automatically to associate ideas with other ideas and feelings (Kahneman, 2003) . This part of the brain is quick to make conclusions, slow to learn, and emotional (Kahneman, 2003) . This intuitive system is almost effortless. It takes substantially more effort (and calories) to use the deliberative system, as is indicated by 1 This move by Jews bewilders Baldwin (1998, pp. 179 180) who asserts that American Jews latent suburban hos Jews consistently vote with people of color in major elections (CNN, 2016) , and Jews are still targeted with vandalism and violence at their homes, businesses, and places of worship.

PAGE 53

41 increased heart rate, pupil dilation, and incr eased respirations (Kahneman, 2011) . However, the deliberative system is more flexible and controllable, enabling individuals to reject initial assessments that are often biased (Kahneman, 2011) . To reject a biased initial assessment is like swimming up ri ver by using a system that takes a substantial amount of effort to reject the conclusions of a system that takes almost none (Kahneman, 2011) . When a human being process es reality, if the intuitive system gets it wrong, it is likely to stay wrong unless th at person has the appropriate information and puts in the effort to overcome that incorrect perception. When it comes to race and gender, the intuitive system often gets it wrong. Another pair of psychologists, Banaji and Greenwald (2013) have spent their career documenting intuitive assessments based on stereotypes. Their findings are both stunning and exactly what scholars of race and gender expect. It takes a fraction of a second less to group flowers and sunshine with pictures of white children than pic tures of black children. It takes a fraction of a second more to associate nurses with men and doctors with women (Banaji & Greenwald, 2013) . Greenwald and Banaji (1995) find that American s intuitive systems create presumptions for fallacious stereotypes benefiting whites and men, and other researchers find that people act upon these presumptions even after admitting they are false (Nelson, Acker, & Manis, 1996) . Intuitive assessments impact how public servants think about others, and how they think about themselves. Stereotype threat research indicates that when a person is aware of a stereotype that is salient to their identity, they unconsciously conform to that stereotype (Steele, 2010) . White men told their athletic ability is being compared to black m en perform worse in mini golf than white men who are not told they are being compared to black men (Steele, 2010) . Similarly, white men perform worse in calculus classes when they share a room with large numbers of Asian men (J. L. Smith & White, 2002) . Me n perform worse on tests of reading body

PAGE 54

42 (Koenig & Eagly, 2005) . One small (n = 33) experiment finds some (p < 0.15) relationship between hi gher gender identification in men and improved math performance (Schmader, 2002) . Also, Danso and Esses (2001) find that whites score higher on intellectual tests when the experimenter is black. By isolating the relationship of proctor to test taker, this (Danso & Esses, 2001, p. 159) . These studies suggest that white men benefit from a stereotype promise based in an intuitive b elief in their own superiority. The work required to maintain perceptions of superiority suggests that Whiteness and Masculinity performances require exertions of emotional labor. Emotional labor has often been associated with gender. Guy, Newman, and Mast racci (2008) argue that emotional labor accounts for some of the differences in pay between women who do more emotion work heavy jobs than men. Meier, Mastracci, and Wilson (2013) find that women perform more emotion work in the same jobs, and that this le ads to better outcomes in public service, and Newman, Guy, and Mastracci (2009) find that women perform different emotional labor as leaders to create healthier workplaces. When it comes to emotional labor and race, Gooden (2015) argues that emotional mana gement, in particular the management of nervousness surrounding race, (2015) points out that public servants are less likely to discuss racism if they are nervous to bring up the topic, and overcoming that nervousness requires emotional labor. Additionally, the well documented Whiteness performances of innocence in the workplace (Brodkin, 1994; Roediger, 1999) as well as the Masculinity performances of unemotional aggression (Coope r, 2000; Morgan, 1996)

PAGE 55

43 (Guy et al., 2008, p. 109) , and is a performance that is closely correlated with burnout. The psych ology of Whiteness and Masculinity gives significant benefits to white men in the form of stereotype promise and intuitive assessments of their performance and appropriate roles. As discussed in chapter one above, taboos and nervousness prevent open discus sion of race and gender in public organizations (Gooden, 2015; J. Martin, 1990) . When discussions do happen, critical scholars find that common rhetorical devices are used to minimize racism and sexism, or accuse women and people of color of racism and sex ism against white men. Rhetorical Whiteness and Masculinity . In public organizations, nervousness and private life taboos are unable to suppress all discussion of race and gender. For example, Foldy and Buckl e y (2014) find that welfare agencies that suppre ss discussion of race develop discriminatory ways of speaking. Similarly, Gooden (2014, p. 72) argues that public servants chameleon, expressing shifting message s regarding racial equity in their agency depending upon gender without directly confronting nervousness and taboos, including microaggressions, minimization, abst ract liberalism, projection, and displacement. Microaggressions communicate Whiteness and Masculinity while maintaining the power of apparent gender and color blindness. A microaggression is a brief or inconspicuous communication of power based on group i dentity (Sue, 2010) . Examples of microaggressions fill by her boss even as he introduces all of her men co (Sue, 2010) . Similarly, a white wo man in an elevator may clutch her purse and recoil when a black college

PAGE 56

44 professor enters (Yancy, 2008) . Politically incorrect statements can be microaggressions, as can wall hangings of offensive mascots, or tone deaf compliments of people who appear to be defying stereotypes associated with their phenotype (Sue, 2010) . Taken in conjunction with the work of Foldy and Buckley (2014) and Gooden (2014) , microaggressions are the inevitable result of nervousness. If color and gender blindness prevents open disc ussion of race and gender, many of these microaggressions may simply be caused by ignorance on the part of people who are too nervous to ask questions that would help them prevent hurting others. In other words, it could be that microaggressions are like a pop quiz that white men and others too often fail unless they overcome their nervousness to do their homework. Microaggressive rhetoric is used to downplay discrimination. Sue (2010) identifies the best candidate should ge a microaggression (p. 13). Bonilla Silva (2003b) justify actions that are discriminatory by hiding bigotry behind a veil of abstract neutrality. This neutrality is fa lse given the many advantages that whites and men receive throughout the course of their academic and professional careers (Lipsitz, 2009; Mumby, 2006) . However, the claim can appear neutral enough to inoculate the shame and anger that many white men feel when their innocence and noblesse are called into question by accusations of racism and sexism (Andersen, 2016; Bonilla Silva, 2003b; Matias, 2016; Sue, 2010) . Another strategy for downplaying discrimination is to minimize the experiences of oppressed gro ups (Bonilla Silva, 2003b) . White men often celebrate symbolic equality gains, distance themselves from oppression by emphasizing the length of time since slavery or primo geniture, or make specious comparisons between the struggles of oppressed groups and the struggles of white men in historic times (D. Bell, 1980; Bonilla Silva, 2012; Gallagher, 2003) .

PAGE 57

45 When all else fails to maintain innocence and noblesse, white men sometimes turn to projection where they accuse others of what they themselves are accused (Bonilla Silva, 2003b) . For example, in 2016 , 43 percent of whites told Gallup that racism against whites was widespread, and a majority of Republican voters claimed discrimination against whites was more prevalent than against blacks or women (J. M. Jones , 2016) . Displacement is a nother psychological term that is analogous to microaggressive performance, but displacement works in the intersection between identities. Displacement is a performance from one institution that reinforces or distracts from a diff erent institutional performance. For example, Choudhury (2007) and others (Ali et al., 2015) document how hijab and niqab are targeted by white feminists as oppressive in ways that reinforce white innocence compared to the morality of Muslims. Similarly, f eminist scholars argue that the struggle to racially integrate military units and leadership is tainted by arguments that reinforce sexism (Scales, 2013) . G. M. Frederickson (1988) argues that neo Marxists focus on class downplays the centrality of race, In popular culture, displacement is evident in arguments that defend violence perpetrated by poor white men as symp toms of their poverty (Murray, 2012; Vance, 2016) , without discussing statistics that show how poor white women do not commit violence at anywhere near the same rate (Gilligan, 1996; Katz, 2013) , nor discussing the many ways that this behavior serves the i nterests of non poor whites. Allen (2008) points out that non poor whites argue poor backwardness and savage masculinity . Yet data show that poor whites are no more racist than non poor whites, Allen (2008) explains, because it is the wealthy, not the poor, who receive the

PAGE 58

46 many benefits of Whiteness. Backing up this claim, experimental research finds that non poor whites are more likely than poor whites to vote again st a candidate tha t they feel antagonizes people of color , not because they reject racism, but because they fear upending the social structure from which they benefit (Jun, Lowery, & Guillory, 2017) . Individual level Propositions . I ndividual performances l ike those documented in the literature will be the color and gender blind normal for most public servants, supporting propositions that can be tested at the individual level. Individual Proposition #1 Whiteness and Masculinity. Individual Proposition #2 Public servants make sense of race and gender through performances of Whiteness and Masculinity. Organizational level Whiteness and Masculinity Masculinity and Whiteness in organizational culture . O rga nizational culture is the process through which people within an organization place value on work processes (Schein, 2010; Weick, 1995) ( see Table 2.3 ). Gender and color blindness are highly successful performances in organizational culture because they a re dominant at the system level, required by law for public organizations, and reinforced by individual nervousness. As discussed above, modern Whiteness and Masculinity rely on the power to suppress discussion and keep white supremacy and patriarchy norma lized (Bachrach & Baratz, 1962; Mills, 1997) . Public organizations themselves benefit from suppressing talk of race and gender. Heckler (2017) argues that color blindness in public administration hurts Americans of color and reinforces Whiteness, while inc reasing the political legitimacy of public organizations by conforming to

PAGE 59

47 Table 2.3: Organizational level performances of Whiteness and Masculinity.

PAGE 60

48 Constitutional legal fictions prohibiting the consideration of race. Heckler (In Press) also argues that nonprofit law places pressure on boards and staff to conform to White ness processes borrowed from the private sector. Similarly, Scales (2006) argues that gender blind justifications hurt women and maintain patriarchic values ( Geduldig v. Aiello , 1974 ; see also Scales, 1992) . Yet under current law, many public organizations are not legally permitted to acknowledge gender and race ( Regents of the Univ. of Calif. v. Bakke. , 1978; Clark v. Jeter , 1988) . Organizational culture extends beyond color and gender blindness. Organizations actively repeat Whiteness and Masculinity cul tural performances. Gooden (2014) finds that leadership is an important factor in undermining nervousness and creating a culture that embraces discussion s of race. This find ing suggests that leadership help create cultures of nervousness in the first place . Leadership teams dedicated to not seeing race make organizational decisions like promotions, assignments, hiring, and termination using Whiteness performances (Gooden, 2014; Greenwald, Banaji, & Nosek, 2015a) . For example, Carbado and Gulati (2013) descr ibe the experiences of lawyers in trial working through the night and watching the sunrise in the morning. They argue that successful white lawyers describe the sunrise from their office, and leave their offices strewn in papers and takeout boxes as artifa cts of their dedication. Successful black lawyers keep their offices tidy and do not emphasize their marathon work sessions because they fear it conforms to black stereotypes of sloppiness and bad planning. The fact that these stories are from the perspect ive of successful lawyers indicates that it is not the individual attorney who makes these decisions, but law firms that reward people of color who can negotiate the organizational culture minefield (Carbado & Gulati, 2013) . Similar to findings that whites take advantage of expectations that benefit them, Kanter (1977) finds that men and women have disparate expectations that advantage men. Men are

PAGE 61

49 expected to aggressively pursue promotions, playing the weaknesses of other men when necessary to get ahead (K anter, 1977). Women, by contrast, are expected to be loyal. Women must give credit for their success to men in the hopes that those men will remain loyal to them and help them climb the bureaucracy (Kanter, 1977) . However, loyalty behavior in men is not en forced, leaving women in a cultural minefield of their own. In the 21 st century, these expectations remain the norm (Cooper, 2000) . Organizational cultures work intersectionally. White men take advantage of organizational cultures that encourage ambition p rimarily in men (Kanter, 1977) , while accepting favorable assumptions that offices occupied by whites are messy as the result of honest hard work, not sloppiness and disorganization (Carbado & Gulati, 2013) . White men also take advantage of their intergene rationa l wealth by securing eldercare and childcare in nursing hom es and daycare centers, even as people of color tha t have had their wealth systematically drained by alienation of their labor and discriminatory housing policies (Oliver & Shapiro, 2006) mu st negotiate masculine taboos that childcare and eldercare is domestic labor better left at home (J. Martin, 1990) . White men enter organizations with cultures that have often been created by white men. This path dependency is evident in cultural touchston es that exclude women and people of color , such as in golfing with the bosses, company softball, watching a sporting event, or chatting about the latest Quentin Terantino film over the water cooler (Cabrera, 2016; Carbado & Gulati, 2013; Matias, 2016; Sue, 2010) . Often, these exclusive touchstones make the difference in the careers of white men who more effortlessly engage in the organizational culture. Masculinity and Whiteness in organizational structure . Masculinity and Whiteness affect organizational st ructure by power shifting so that more decision making authority and resources are allocated to whatever part of the organization has more white men. Mintzberg

PAGE 62

50 (1983) finds that organizations change their structures to reflect power in those organizations. For example, a professional bureaucracy, like a law firm, is likely to spend more resources and put more decision making authority near the bottom of the organization where the lawyers work (Mintzberg, 1983) . Organizational theorists have argued that both technocra cy and bureaucracy put more power in the hands of white men (Burris, 1996; Ferguson, 1984) . Bureaucracy reinforces performances of Masculinity and Whiteness. Bureaucracies consist of clear hierarchies, documented decision making guidelines, and c areful cultivation of culture as effective means of creating unified organizational control even when organizations are decentralized and separated by geographical and cultural distance (Kaufman, 2006; Weber, 1972) . Gulick (1937) finds that bureaucratic co ordination requires four steps that are impacted by race and gender. First, managers define the job, a task burdened with intuitive assessments about how jobs should be grouped together. Managers set accountability mechanisms, which reflect trust set by Wh iteness and Masculinity. Then managers decide how many people are needed for each task and establish a power structure, both tasks dependent on the same Whiteness and Masculinity performances that impact other organizational tasks. Ferguson (1984) argues t because of the emphasis on control over workers to submit to the will of a masculine (1924) outlines a model for org anization that empowers front line workers, Ferguson (1984) offers alternative forms of organization, but these alternatives still exist within an institutional framework that is laden with Masculinity and Whiteness. For example, Chen (2009) studies the an nual Burning Man event to find that lack of centralized organization ends up reinforcing hegemonic power dynamics that benefit white men.

PAGE 63

51 The most common alternative to the strong vertical bureaucracy, technocracy reinforces Masculinity and Whiteness by pr oviding justifications for racist and sexist workforce divisions. Stivers (2000) traces the origin of public administration from a conflict of approaches taken by the New York Bureau of Municipal research dominated by men, and the settlement houses that we re mostly the purview of women. Epitomized by Simon (1959) , b ureau men focused on scientific accountability and efficiency. Contrarily, Jane Addams (1911) pioneered the settlement house movement dedicated to practical solutions and research from within com munities in need. Stivers (2002) argues that the efficiency and scientific accountability perspective reflect a masculine focus that has come to dominate modern public administration research and practice. Burris (1993) calls this focus technocracy . Techno accountability shifts the focus of public administration from what is being done to how things are done (Burris, 1993) . In other words, technocracy encourages public servants to ask if they are (Adams & Balfour, 2004, p. 41) . The technocratic emphasis on expertise divides workers in two groups often by race and gender, and treats each group disparately. Burris (1993) explains tha t technocracy is not possible without an expert class to provide scientific answers to questions about efficiency. As a result, organizations become divided into expert and non expert sectors, and while egalitarianism takes over the expert sector (Mintzber g, 1979) , the non expert sectors find their work becomes increasingly dictatorial (Burris, 1993) . Burris (1996) finds the expert sector occupations are those of m ale dominance in which some men (and women who adopt patriarchal values and practices) people of color

PAGE 64

52 through an emphasis on degrees as proof of expertise combined with the challenges tha t people of color face in university settings (Fryar & Hawes, 2012; Steele, 2010) . Even as the expert class becomes flatter in the technocratic meritocracy, the non expert jobs assigned to women and people of color pay less, and become ever more tightly co ntrolled (Burris, 1996) . A focus on efficiency serves these technocratic organizations because it enables organizations to survive in an environment that is itself focused on efficiency over equality, while at the same time giving more value to microaggres technocratic movements in public administration include New Public Management (Page, 2005) , the Reinvention of Government (Ingraham, Thompson, & Sanders, 1998) , and e governance (Dunleavy, M argetts, Bastow, & Tinkler, 2006) . As public organizations are buffeted by these tides of demands for more scientific accountability and efficiency, the gulf between expert and non expert widens. Public organizations are pushed by Whiteness and Masculinity from systems level institutional performances, even as they are pulled from within by individuals performing Whiteness and Masculinity. This squeezing shapes organizational structures and cultures, and is normalized by the fact that they are set against a society that is mostly white and masculine. However, normalization does not erase problems. Whiteness and Masculinity in public organizations are both well documented, and almost entirely unnoticed because of gender and color blindness among researchers in many fields. This theory isolates Whiteness and Masculinity and subjects the intersection between them to the same scrutiny that has been placed on other institutions such as capitalism (Weber, 2004) , consumerism (Campbell, 1987) , Blackness (Fanon, 1963 ) , and Femininity (Bartky, 1990) . Organizational level p ropositions . Organizations form and persevere in societies that

PAGE 65

53 are dominated by whites and men, and are occupied by individuals who must negotiate Whiteness and Masculinity. As a result of these pres sures from both sides, four propositions can be explored in organizations. Organizational Proposition #1 Public organizational structures allocate power and resources to whites and men. Organizational Proposition #2 Public organizational cultures reflect Whiteness and Masculinity performances. Organizational Proposition #3 Public organizations with relatively fewer performances of Whiteness and Masculinity will be more welcoming to people of color and women. Organizational Proposition #4 Public managers implement Whiteness and Masculinity performances as operational processes and systems. Whiteness and Masculinity in Public Service Motivation and Job Satisfaction I nternal organizational race and gender are not as well studied in public organizations as i n organizations more generally (Gooden, 2014) . The representative bureaucracy literature shows gender and race of public servants affect a variety of outcomes from public organizations (Riccucci & Van Ryzin, 2017; Selden, 1997) . At the same time, organizat ional theorists find that Masculinity and Whiteness impact the internal workings of organizations generally (D. L. Collinson & Hearn, 1996; Macalpine & Marsh, 2005) . Even reconceptualizing much of the representative bureaucracy literature as implicitly ana lyzing Whiteness and Masculinity, there is little to no research on Masculinity and Whiteness in the internal workings of public organizations. One throughput that is unique to public organizations is public service motivation (PSM) (Perry & Wise, 1990) . Masculinity and Whiteness affect PSM of white men and others in public organizations.

PAGE 66

54 Perry and Wise (1990) find that employees of public organizations are motivated differently than employees of private sector organizations. They argue that public employe es are driven in part by the rational motives often theorized in the business management literature, along with norm based motives to serve the public interest, and affective motives to serve the missions of public organizations (Perry & Wise, 1990). Perry (1996) defines four dimensions of PSM that have been widely tested and found to be related to various outcomes including job satisfaction, lower job turnover, and willingness to choose a public service job (Perry, Hondeghem, & Wise, 2010, p. 683). People exhibiting these four dimensions are attracted to public policy making, desire to work in the public interest, are driven by compassion, and have a willingness to self sacrifice (Perry, 1996, p. 15). In the 25 years since Perry and Wise (1990), dozens of r esearchers have operationalized one or more of these dimensions (Perry et al., 2010). In multiple tests, PSM is found to differ based on racial and gender performance. Most of the time, these differences are reported While Liu and P erry (2014, p. 13) find no difference between PSM in men in the Chinese government, many other studies have found differences in American public workers. Men have been found to have less attraction to policy making in one study (Moynihan & Pandey, 2007, p. 46), and be less likely to volunteer in a model predicting PSM (Perry, Brudney, Coursey, & Littlepage, 2008, p. 451). Dehart Davis, Marlowe, and Pandey (2006, p. 880) find that women are more attracted to policy making and more driven by compassion before exhorting researchers have stipulated that formative experiences like race and gender performance do create and maintain PSM, little r esearch has looked at mechanisms (Weaver, 2015, p. 445). PSM can be changed by organizational factors. Organizations manage PSM in their

PAGE 67

55 selection (Perry et al., 2010, p. 683). Organizational factors like red tape, hierarchical authority (Moynihan & Pandey, 2007, p. 46), and strong organizational identity (Liu & Perry, 2014) and mission (Lee & Wilkins, 2011 ) change PSM over time. While individuals are generally assumed less malleable than organizations in the develo pment of PSM, at least two studies find that organizational characteristics can orient individuals toward higher levels of PSM in the same individuals over time (Liu & Perry, 2014; Ward, 2014). Between attraction, selection, and attrition within an organiz ation, and development of PSM through public management, the research shows that PSM should change over time within public organizations gen erally, evolving with the institutions around them and the individuals within them. Job satisfaction relate s to publ ic service motivation, but is not unique to public organizations. (1974) hygiene hypothesis makes sense of the difference between motivation and satisfaction. Herzberg (1974 ) argues that motivation and satisfaction are driven by different kinds of factors. While PSM is a theorization of motivation unique to public organizations , hygiene factors like supervision, interpersonal relationships, working conditions, salary, status, and job security are attributes of all organizations with employees (He rzberg, 1974, p. 20). He rzberg (1974) argues that motivation factors in private companies will not impact retention unless h ygiene factors are adequately addressed, and that simply acknowledging a problem and resolving it is only the beginning step in moti vating employees to more productivity (Herzberg, 1974, p. 25). It is unclear from existing literature if racial and gende r performance impact motivation or hygiene, or whet her PSM is impacted differently. Some research on satisfaction and turnover indicate s that gender and race performance may impact job satisfaction in public organizations . Weaver (2015) finds that men and whites are

PAGE 68

56 less likely than people of color and women to leave a federal job for another federal agency, but not less likely to leave f or another sector. Pitts (2009) finds that whites are less satisfied than non whites in their job after diversity training. While empirical studies on the ways that racial and gender performance aff ect job satisfaction are scant, this literature suggests t hat Whiteness and Masculinity perform ances are likely to impact job satisfaction . Organizational level Hypotheses . The literature cited here supports a theory that organizational Masculinity and Whiteness impacts the PSM and job satisfaction of whites, men , and white men in public organizations. These public servants embody all of the individual, organizational, and systems level performances in Tables 1 3. Organizational Whiteness and Masculinity will be different in each organization, as each organization negotiates and embeds their institutional makeup. The literature indicates that organizations with the most embedded Masculinity and Whiteness will give substantial benefits to whites, men, and white men, and those whites, men, and white men are the most likely to take on supervisorial roles, often over women, people of color , and women of color (Carbado & Gulati, 2013; Kanter, 1977) . Using supervisorial status as a proxy for benefit to whites, men, and white men that indicate Whiteness and Masculinity, th e following hypotheses can test internal organizational race and gender by measuring PSM and job satisfaction. H1a: When whites are over represented in supervisorial teams, whites in those organizations will exhibit more PSM . H1b: When men are over represe nted in supervisorial teams, men in those organizations will exhibit more PSM . H1c: When white men are over represented in supervisorial teams, white men in those organizations will exhibit more PSM .

PAGE 69

57 H2a: When whites are over represented in supervisorial t eams, whites in those organizations will be more satisfied . H2b: When men are over represented in supervisorial teams, men in those organizations will be more satisfied . H2c: When white men are over represented in supervisorial teams, white men in those or ganizations will be more satisfied . Conclusion What this chapter describes is a syndemic of Whiteness and Masculinity. A syndemic is a complex of outcomes, mechanisms, and causes that are all intimately related, each cause contributing to an outcome that is also a mechanism that contributes to the original cause. Singer (2000) coined the term syndemic in his study documenting how the AIDS crisis was more than just a series of quantifiable risk factors, but an intertwined social pattern where violence, drug use, and HIV infection were causally related in multidirectional relationships. In a more recent study, Horton (2016) records the intricate relationship between kidney disease, heat related disease, poverty, diabetes, food inspection laws, immigration pol icy, and undocumented status among farming communities in California. The propositions and hypotheses established in this chapter help outline a systematic process for exploring a syndemic of individual behaviors and assumptions; organizational structures, cultures, and processes; and societal forces and influence s that combine to reinforce white supremacy and patriarchy in public organizations. The hypotheses will be tested using publicly available from the federal government . The study will explore the pr opositions by interviewing public servants in the federal government, following their social media postings, and tracking their work processes. The next chapter describes this methodology.

PAGE 70

58 CHAPTER THREE OBSERVING WHITENESS AND MASCULINITY IN S OCIETY, INDI VIDUALS, AND PUBLIC ORGANIZATIONS Focusing on the federal workforce, this study investigated the hypotheses and propositions presented in Chapter Two . In the first stage, quantitative methods test ed hypotheses that race and gender impact public service mot ivation and job satisfaction. Guided by the findings of the first stage, the second stage conduct ed an institutional ethnography on agency offices analyzed in stage one. This chapter describes the two stages of research . Figure 3.1 visually depicts how thi s dissertation sought to combine stage one quantitative methods with stage two qualitative methods. T he findings and summary statistics of stage one identified organizations as typical examples of federal government office s for further qualitative analysis in stage two. By placing the qualitative research in a context developed through quantitative methods , the qualitative research support s a theory that is representative of the scores of offices operating within the civilian federal government (Bourgois, 1 999) . T he stage two findings form ed an empirical basis for theorizing the mechanisms behind the PSM and job satisfaction dif ferences discovered in stage one. This triangulation is suggested by Riccucci (2010) Stage One : Race and Gender in t he F ederal Employee Viewpoint Survey The hypotheses laid out in Chapter Two were tested using quantitat ive methods on a survey of the f ederal workforce. This section reviews the hypothese s and describes the dataset, before explaining how the independent, dependent, and control variables were compiled. The analysis subsection describe s how multiple estimation tools were used to overcome various limitations in the data .

PAGE 71

59 F igure 3 .1 . Model of the relationship between stage one and stage two research.

PAGE 72

60 Hypotheses: Race and Gender Affect Job Satisfaction and Public Service Motivation Six hypotheses guide this research. Table 3.1 lays out the hypotheses along with the variables need ed to test them. The rest of this sub section discusses the dat aset, and how this research o p erationalizes those variables. Data: the FEVS Stage one employed the Federal Employee Viewpoint Survey (FEVS) (OPM, 2016) to test these hypotheses. The FEVS (OPM, 2016) is an annual survey distributed to federal employees including military and civilian pe rsonnel. The survey includes 84 questions on morale, workplace norms, job satisfaction, and basic demographics. The dataset includes more than 1.5 million observations. The data are longitudinal with a cross section from six years, 2010 2015. The observati ons are also nested in 20 agencies and 136 federal offices within those agencies. The FEVS is generally considered a representative dataset (Weaver, 2015a) with 45.8 percent response rate governmen t wide. Table 3.2 displays some of the similarities between the sample of federal workers who filled out the FEVS and the cumulative population of federal workers in the same five years as described by the OPM. The proportion of men and whites in the sample are very similar to the population. When it comes to payg rad e, the FEVS sample is 31.8 percent GS 13 or higher, while t he total population is 20.3 percent GS 13 or higher, indicating some difference in the education and experience of the sample. Of the control variables, paygrade had the lowest impact on initial e stimates of satisfaction and PSM, indicating that the difference between the sample and the population was not a fatal limitation . FEVS data included variables for measuring independent, dependent, and control variables for testing the

PAGE 73

61 Table 3.1 Stage one hypotheses. Hypothesis Independent Variable ( Office Level) Dependent Variable (Ind ividual Level) H1a: When whites are over represented in supervisorial teams, whites in those organizations will exhibit more PSM. The proportion of w hite supervisors minus the proportion of white employees. PSM Index H1b: When men are over represented in supervisorial teams, men in those organizations will exhibit more PSM. The proportion of man supervisors minus the proportion of man employees. P SM Index H1c: When white men are over represented in supervisorial teams, white men in those organizations will exhibit more PSM. The proportion of white man supervisors minus the proportion of white man employees. PSM Index H2a: When whites ar e over represented in supervisorial teams, whites in those organizations will be more satisfied. The proportion of white supervisors minus the proportion of white employees. Satisfaction Index H2b: When men are over represented in supervisorial teams, men in those organizations will be more satisfied. The proportion of man supervisors minus the proportion of man employees. Satisfaction Index H2c: When white men are over represented in supervisorial teams, white men in those organizations will be mor e satisfied. The proportion of white man supervisors minus the proportion of white man employees. Satisfaction Index Table 3.2 Similarities between the FEVS and the Federal Workforce generally. Demographic % of FEVS % of Federal Workforce Men 51.2% 56.4% Whites 65.3% 65.9% Paygrade > GS12 31.8% 20.3%

PAGE 74

62 Independent V ariable s Racial and Gender Congruence The stage one independent variables were race and gender congruence of supervisors to employees in federal offices . The FEVS (201 8 ) asks res pondents if they are supervisors , their race, Hispanic or non Hispanic ethnicity, and whether they are male. anonymity, the FEVS lists all persons who are not white non . For this study, a person who id entifies as a male is counted as a man , and a person who selects white non Hispanic is counted as white. Federal offices range widely from 7.2 percent to 89 percent white employees , 22.1 percent to 82 percent men employees , and 4.5 percent to 63.6 percent white men employees ( see Table 3.3 ). Table 3.3. shows that w hites and men are over represented in supervisorial roles compared to their employees. To operationalize this over representation, the proportion of white/man/white man supervisors in each office is subtracted from the proportion of white/man/white man employees in each agency office . Take the example of an office with 60 percent white supervisors, and 60 percent white employees. This office would be 0 percent over representative because 60 minus 6 0 is 0. However, the hypothetical office that has 5 .5 percent to o many white supervisors could mean that the supervisors would be 65 .5 percent wh ite, while the employees would b e only 60 percent white. Table 3.3 shows that employees are 65.3 percent white, 51.2 percent men, and 35.9 percent white men. However, these employees are managed by supervisorial teams with 5.5 pe rcent too many white s , 9.9 percent too many men, and 7.7 percent too many white men in supervisorial roles . Congruence as independent vari ables were modeled as predictors for PSM and job satisfaction as dependent variables.

PAGE 75

63 Table 3.3 Independent variables summary statistics Independent Variables N Mean Std. Dev. Min Max Percent white 1,676,564 65.3% 10% 7.2% 89 % Percent men 1,750,885 51.2% 14% 22.1% 82% Percent white men 1,676,564 35.9% 12% 4.5% 63.6% Percent too many white sups 1,750,885 5.5% 4.4% 18.9% 47.4% Percent too many men sups 1,750,885 9.9% 4.4% 2.9% 33.4% Percent too many white men sups 1,676,5 64 7.7% 3.6% 3.1% 36.4%

PAGE 76

64 D ependent V ariables PSM and J ob S atisfaction . The dependent variables for the stage one analysis were PSM and job satisfaction. Several existing studies have used the FEVS to capture public service motivation and job satisfa ction ( e.g. Alonso & Lewis, 2001; Harrington & Lee, 2015; Pitts, 2009; Weaver, 2015). The five facets of PSM are attraction to policy making, commitment to public interest, civic duty, compassion, and self sacrifice (Perry, 1996) . This study measures three of these facets; commitment to public interest, civic duty, and self sacrifice . Using the FEVS, Weaver (2015, p. 446) captures commitment to public interest with interaction terms derived from responses to th e Using the list of questions that capture PSM provided by Perry (1996), this dissertation develops a more nuanced measure of commitment to public interest by combining the respons ives me personal accomplishment An index variable was constructed from the three terms with alpha of .7957 . C ivic duty wa s 20 sacrifice. These op erationalizations were The other two facets of PSM, compassion and policy making were not readily captured using responses to the FEVS. As displayed in Table 3.4, t he overall P SM index ha d a skewed mean of 3.73 out of fiv e

PAGE 77

65 possible points (the mean with no skew would be exactly three) , indicating that federal employees exhibit h igh levels of PSM. Whites, men, and white men ha ve lower levels of PSM compared to the entire sample , revealing racial differences in PSM. Job s atisfaction was measured with an index constructed from the responses to seven involvement in decisions that affect your wo a alpha of .884. is given equal weight to all six of the other satisfaction variables combined in the index . As with PSM, the mean satisfaction ind ex for federal workers is skewed high. Table 3.4 shows the mean federal worker indicating a 3.48 out of five satisfaction score . Unlike with PSM , whites, men, and white men did not show statistically significant differences in job satisfaction compared to the entire sample . To verify that the PSM and job satisfaction of these employees is affected by the congruence of race and gender of their supervisors, the study must eliminate alternative explanations by using control variables. To account for common source bias, factor loading revealed that the dependent vari ables did not hang well with the explanatory independent variables. Meier and O T oole (2013) explain that common source bias is a threat to validity when the dependent variable and the independent variable bo th originate from the same source, in this case the same respondent. To test for common source bias, factor analysis was run on all variables included in the model to determine if the dependent var iables hang with the explanatory independent variables. T h e highest f actor score for a ny dependent variable was <.0 8 , revealing very l ow risk of common source bias. C ontrol V ariables Common c ontrol variables are found in the literature on PSM and job satisfaction (Pitts, 2009; Weaver, 2015a) . These control variables include military/veteran status, paygrade,

PAGE 78

66 Table 3. 4 Dependent variables summary statistics Table 3.5 Control variables summary statistics. Control Variables N Mean Std. Dev. Min Max GS 13 1,245,158 3 1.8% >5 Years Experience 1,750,885 73.6% Under 40 1,723,327 24.8% Active Duty or Veteran 586,169 28.1% Unit Resources 1,750,885 3.32 0.167 2.790 3.950 Observations per org/Year 1,750,885 10,531 12317 177 46687

PAGE 79

67 experience/tenure in the government, age, and organizational resources , and organiza tion size. Veteran/Military status, paygrade, age, and tenure are self reported on the FEVS . The job resources variable is borrowed directly from Pitts (2009, p. 333), and consists of an index of six The unit resources were measured at the agency office level to capture the group perception of resources, thereby mitigating personal bias. Table 3.5 lists these variables along with some summary statistics . With the variables compiled, t he next step was d independent Analysis In this sub section, several estimation tool s for testing the hypotheses are described, along with how each tool addresses different limitations . The data are limited by the characteristi cs of the data including skew ness , autocorrelation, and multicollinearity . E stimation tools that can address these limitations include Bayesian, Tobit, binary logit, ordered logit, mixed effects regression, and propensity score matching. Table 3.6 outlines some of the characteristics and limitations in the data, alo ng with suggested methods of addressing them . Having laid out the multiple analysis tools in Table 3.6, each analysis tool is described below starting with Bayesian estimates. Bayesian estimation . Bayesian statistics calculate propensity for the observations in a dataset given a certain condition. This differs from most statistical methods used in social science, which calculate the frequency that a value is found among all observations. This mea ns

PAGE 80

68 Table 3.6 Analysis strategies for overcoming limitations in the FEVS dataset. Characteristics of FEVS App ropriate Analysis Tool Limitations Overcoming Limitations Large n (>1.5 million observations) creating risk of false positive findings Bayesian analysis to reduce risk of false positive Bayesian tools less well develop ed. Use non Bayesian (i.e. standard) methods initially Non normally distributed, skewed toward higher numbers Tobit regression Not appropriate for variables base d on numerical choice (i.e. Lik ert surveys) Create a dummy variable for logit or probit analys is Non normally distributed, skewed toward higher numbers Logit regression Creates arbitrary cut off that may not be empirically meaningful. Analyz e as continuous variable, or Li kert variable in Ordered Logi t . Ordered choice dependent variable (Lik ert s cale) Ordered Logit Not appropriate for skewed da ta, nested data, risks of large n data Analyze using Bayesian, logit, T obit, and multi level methods Nested at three levels; organization, agency, and year. Mixed effects (Hierarchical linear Modelling) Doe s not account for skewed data, assumes linearity, multicollinearity. Models do not Different optimizers, standardized variables, analyze with propensity sc ore matching, Bayesian, logit, T obit. Multicollinearity of important I.V.s Propensity sco re matching followed by multiple regressions Large N dataset and nested data Run estimations on separated datasets, then compare using t tests

PAGE 81

69 that Bayesian estimates have different strengths and weaknesses than more standard statistical methods, and can complement other methods. For example, one limitation of standard statistical methods is an increase of false positive findings in datasets o f more than 10,000 observations. S ocial science methodologists and statisticians agree that Bayesian regression methods are an effective means of mitigating this increased risk of false positives (Cowles, 2013; Gill & Meier, 2000; Kruschke, 2012; Marchenko , 2015; Masson, 2011; Rice, 2014) . The 1 0 years of data being used in this analysis include over 2 million observations, and so the risk of false positive f indings (type I error) is a limitation that should be addressed. There are many tools for running Bayesian regressions in R including the BLME package (Dorie, 2015) for estimating Bayesian generalized linear and non linear modelling. Despite their advantages, Bayesian methods are often critiqued for being less clear than standard statistical methods (Gill & Me ier, 2000) . Bayesian can be combined with other estimation tools to account for other limitations, such as Tobit and logit methods that account for non normal distribution. Tobit and binary logit . Tobit and binary logit methods account for the fact that th e data are not normally distributed. Most methods assume that data are normally distributed around the mean, but the PSM and job satisfaction indices created for this study are skewed to higher numbers, with too few responses in the lower range. Ultimately , Chapter Four reports binary logit with odds ratios to test the hypotheses. This method accounts for non normal distribution by pick ing an arbitrary cut off, such as the median PSM or s atisfaction score, and creating a dummy variable where 1 indicates abo ve the median responses, and 0 indicates below the median responses. The limitation in creating this dummy wa s that the arbitrary cut off may not have empirical validity. Another way to account for this non normality is to run Tobit estimations that are de signed to account for this skew (Wooldridge, 2013) . However, Tobit

PAGE 82

70 regressions are not by themselves capable of limiting the threats to validity caused by large numbers of data and nesting. Also, the Tobit assumption of linearity is violated by the fact th at the PSM and job satisfaction are not, in fact, linear variables , but ordinal . Ordered logit . The default regression method for surveys in which res pondents answer on a 1 5, or Li kert scale is ordered logit. Ordered logit assumes that variables are ordin al, like the Likert scales in the FEVS , and accounts for some of the non normal distribution in the dataset (Wooldridge, 2013). Ordered logit cannot, by itself, account for the skewed data nor for the risk of fa lse positives created by large datasets. Orde red logit also must be combined with methods like mixed effects to account for th e limitations of nested variable s. Multilevel modelling . The data is nested in six different year s, 27 different agencies, and >130 offices within those agencies, and so mixed effects, also known as hierarchical linear modelling, is appropriate. Mixed effects regression uses fixed effects to estimate independent variables of interest, along with random effects to estimate the nesting variables , in this case offices , agencies, a nd years . M ixed effects modelling is considered the gold standard when accounting for nested data (Wooldridge, 2013) . Initial mixed effects regressions reveal that the data are not conducive to this kind of analysis . All initial mixed effects regressions r un for this study have drawn convergence errors. Two methods for fixing convergence errors are standardization of variables and experimenting with alternative optimizer algorithms. Both methods have been attempted with no success. While multilevel and Baye sian modelling analyzed data that was aggregated to the unit level, they were not capable of analyzing individual level data, and so are not reported in Chapter Four.

PAGE 83

71 Stage One Limitations T his study has several limitations and they are heightened by the f act that race and gender is subject to the epistemological blind spot s discussed in Chapter One. Therefore , it is important to carefully consider limitations of the data, as well as of understanding, both of which apply to this research. Some l imitations w ith the data are discussed above in the analysis subsection (Table 3.6). A related limitation is cause d by the fact that the data are longitudinal over six years from 2010 to 2015, and so autocorrelation is a threat to validity. Speaking to that threat, OP M (2015) collects an inclusion index based on questions in the FEVS that capture fairness, openness, cooperation, support, and empowerment. OPM (2016) reported that the federal government percent, increased by 60 percent from 2010 2015. This relationship may indicate autocorrelation by showing that both the inclusion index and global satisfaction are correlated over time, indicating that the time period of an observation may be distorting results. Methods that help delineate nested effects and causal effects will be used , especially mixed effects regression and propensity score matching (Wooldridge, 2013) . A second type of limitation is in how the FEVS captures Masculinity and Whiteness . If Whitenes s and Masculinity are performances as hypothesized here, then anyone can perform survey may indicate a propensity for performing more consistently with Whitenes s, it does not man . When males are openly performing as transgender wom en, the term male could capture the

PAGE 84

72 probably correlated to masculine performances that will affect job satisfaction and PSM, but the overlap is smaller than many assume. Th is set of limitations is conservative (increase s t ype II error), making it less lik ely that the study will find that race and gender affect PSM and satisfaction. The third set of limitations deals with external validity, and applies to all quantitative eff orts to capture social interactions like PSM, satisfaction, race, and gender. Quantitative methods are theoretically generalizable to a population, in this case public organizations. However, as a study becomes more generalizable, the finding itself tends to become less meaningful (Riccucci, 2010) . I t is important that studies such as this link to context developed through qualitative methods. The large N of the dataset (>2,000,000 observations) creates another challenge to credibility that is com plicated by the s ubject matter . Greenwald, Banaji, and Nosek (2015) argue that their research on implicit bias related to race and gender have w ide effects on society, race and gender on all individuals in the United States means that small explanatory power can have important impacts on anyone who has a race or a gender. This argumen t dovetails with the critical race and feminist perspectives that race and gender impact every activity in social life, so society wide effects (Sue, 2010) . These insights unmask a fundamental challenge to quantitative analysis of race and gender phenomena. Every independent variable and every dep endent variable capturing some aspect of social life (e.g. managerial race and job satisfaction) are always already effected by race and gender. Therefore, r ace and gender are a common cause in every model , and every statistical te st is at risk of t ype II error (Gerring, 2012) . In thi s reality of ever present endogeneity, large N tests such as this create

PAGE 85

73 an opportunity to discover these small effects t hat have societally large effects (Greenwald et al., 2015) . In th e context of race and gender , l iterature on large N datasets take on a different light. Kline ( 2004) convincingly discusses the limitations to methods like those used in this study , and argues that replication and Bayesian analyses are more valid for understanding social reality, reflecting the assertions of Gill and Meier (2000) that public administration research is overly reliant on null hypothesis significance tests . Part of the problem is that large N datasets are oversensitive, creating an increa sed risk of t ype I error. Like the limitation stated by Greenwald et al. (2015), t his limitation is unavoidable in the current study . It should be noted, however, that t he type I error of this study s large N analyses is offset by the type II error derived from the common cause inevitable in models o f race and gender mechanisms . All of these limitations are such that the an a l ysis will be strengthened by the profound understanding that is possible through qualitative methods . Using Stage One Findings to Situate Stage Tw o Findings Three statistics obtained in the stage one analysis were used to guide sampling for the stage two research . All agency offices analyzed in stage two were within 1.96 standard deviations from the 1) mean proportion of white men employees to white men supervisors, and 1.96 standard deviations from the mean estimated differences between predicted and actual organizational 2) PSM and 3) j ob satisfaction. Because we know that the mean proportion of white men employees to w hite men supervisors in the FEVS is 7.7 percent with a 3. 6 percent standard deviation, we can calculate that an acceptable organization for stage t wo analysis were between 14.75 percent and .64 percent too many white men supervisors. Similarly, the qualita tive sample consisted of organizations that were within 1.96 standard deviations of the

PAGE 86

74 PSM and job satisfaction that was predicted for a similar organization in the stage one findings. This created a more generalizable base for the qualitative stage two f indings. Stage Two : Institutional Ethnography of Federal Offices Stage two will explore eight propositions u sing institutional ethnography to discover and analyze (Campbel & Gregor, 2002, p. 15; D. E. Smith, 2005 ) . Table 3.7 states the propositions, along with the types of pe rformances needed to analyze them . The primary assumptions o (1) people organize around social interactions that can be mapped and analyzed (D. E. Smith, 2006) , (2) so cial interactions communicate power over people (D. E. Smith, 2005) , (3) the everyday world contains problematics that can be explicated (D. E. Smith, 1987) , (4) tracking work processes can illuminate institutional realities (Campbel & Gregor, 2002) . Whiteness and Masculinity performances, like social interactions, can be mapped and analyzed using institutional ethnography. Whiteness and Masculinity performances communicate and reinforce power for whites and men in public organizations. To track how social interacti ons influence individual behaviors, institutional ethnographers focus on routine practices, methods, and procedures (Campbel & Gregor, 2002, p. 78) . Based on the literature, this study was predicated on the assumption that race and gender performances are social interactions that can be identified, mapped, and analyzed using institutional ethnography ( Smith, 2005; S. Turner, 2006) . The assumption that white supremacy and patriarchy are powerful forces in public organizations operate d as the problematic for this study (D. E. Smith, 1987) . In institutional

PAGE 87

75 Table 3.7 Stage two propositions Proposition Performances Individual Proposition #1 include performances of Whiteness and Masculinity . All Individual level Performances Individual Proposition #2 Public servants make sense of race and gender through performances of Whiteness and Masculinity. Discriminatory framing, blindness, displacement, minimization, projection, microaggressions, innocence, noblesse Organizational Proposition #1 Public organizational structures allocat e power and resources to whites and men . Technocracy, adhocracy, expert/non expert dichotomy, power shifting, strong vertical bureaucracy Organizational Proposition #2 Public managers implement Whiteness and Masculinity performances as operational proces ses and system s . Dress, speech, mannerisms, private life taboos, touchstones, blindness, innocence, noblesse, discriminatory framing Organizational Proposition #3 Public organizational cultures reflect Whiteness and Masculinity performances . Touchstones, dress, speech, mannerisms, capitalism, consumerism Organizational Proposition #4 Public organizations with relatively fewer performances of Whiteness and Masculinity will be more welcoming to people of color and women . All performances, job satisfaction of women and people of color Societal Proposition #1 Public organizations and public servants refer to society level Whiteness and Masculinity to resolve conflict . All societal level performances, blindness, minimization Societal Proposition #2 Public organizations and public servants refer to society level Whiteness and Masculinity to maintain their status with other organizations, public servants, and individuals in society . All societal level performances, blindness, minimization

PAGE 88

76 (Campbel & Gregor, 2002, p. 47) . This focus on making aspects of everyday life puzzles to be solved is a means of ionalization (Campbel & Gregor, 2002, p. 124) by making institutions apparent. In the current study, the institutions of Whiteness and Masculinity, normally veiled by miscognition (Mills, 2007) were subverted by focusing on their performances as social in teractions that were used to dictate power in typical organizations in the federal government . Case Selection After the estimations from the quantitative results are assessed , case selection for the qualit ative study will begin. The stage two sample was eight organizations, studied through interviews with 1 4 federal workers in Denver, Colorado and Washington, D.C. Half of the informants were white men in seven different agency offices . These seven informant s work ed in offices that results from stage one indicate d were (Henry, 1990) as described in more detail above . Gathering typical cases facilitated more effective triangulat ion by identifying similar cases to achieve saturation more efficie ntly (Ulin, Robinson, Tolley, & McNeill, 2002) . The informants were work ed in non military agencies, were permanent employees, and were not be engaged in abnormal activis m, like association with either antiracist or white supremacist groups. The seven whit e men informants were with seven people who are either women, people of color or both. T hese seven confirmatory informants were selected based on their ability to comment on the same organization as the white men informants and provide a position for trian gulation (Esterberg, 2002) . Race and gender research indicates that choosing informants who are not white men increases the likelihood of observing Whiteness and Masculinity more clearly (Hooks, 1984; Thandeka, 1999) .

PAGE 89

77 The researcher found informants by lev eraging personal connections among f ederal workers in Denver and Washington, D.C., and by using snowball techniques to expand to other organizations and individuals. Snowballing occurred during correspondence before and after the interview when informant s were asked to list other individua ls within the organization and in other agencies that were willing to participate. After completing at least seven interviews in Denver, the researcher organized a one week trip to Washington, D.C. and establishing seven i nterviews there and in rural Virginia . Data Collection : Interviews and Social Media Observation I nterview guide. The interview guide was designed to captur e the performances underlying the eight propositions , without leading informants to any answers . In l ine with institutional ethnography (Campbel & Gregor, 2002) , the first part of the interview guide was focus ed worker takes, wh y each task is done, and on who s e auth ority. For example, organizational proposition #4 (Table 3.7 ) s tates that managers use Whiteness and Masculinity operational processes , indicating performances like technocracy and bureaucracy . This proposition can be captured by asking the informant to ta lk about tasks and goal setting. The second section of the interview guide focus ed on the organization level propositions by asking the informant about their experiences with race and gender in the workplace. The last section of the interview focus ed on th on discrimination, affirmative action, and diversity management. This last section was designed around the interview questions used by Matias and Mackey (2016) to capture Whiteness in students tak ing a teacher training course. The section include d (Table. 3 .7 ) stipulating that

PAGE 90

78 Masculinity and Whiteness are used to make sense of race and gender in the workplace. This example question encourages informant s to discuss their feelings about race and gender, allowing them to express blindness, but also encouraging them to process intersectional Whiteness and Masculinity . The chances that the researcher will observe institutional performances increased when the setting in which the interviews are collected is comfortable for the informant. Collecting data . Each interview was conducted in a location of the informant , fice, quiet coffee shops, or reserved temporary spaces. The interviews were recorded using a small, unobtrusive digital recorder. The digital recording was trans cribed soon after the interview using speech recognition software designed for court reporting ( Dragon Professional , 2017) with the researcher listening to insure accuracy and add notes on context including surroundings, and mannerisms. After each interview, the researcher add ed to a digital journal, and follow ed up to collect supplementary data. S u pplementary d ata . When the interview was complete, the research er tracked down documents and social media posts to contextualize the interview responses. Forms, computer applications, regulations, and other documents used by the informant in their daily jo b were collected using email requests, agency websites, and public records requests when needed. These texts provide d context and triangulate d interview data. The researcher request ed permission to view publicly shared social media posts in cluding Twitter, Facebook, MySpace, Reddit, Vine, Instagram, and other sites where the informant communicates . As noted by Frisoli (in Marhsall & Rossman , 2011 ) , social media is ministration research on social media has used the data quantitatively by coding postings made by public organizations

PAGE 91

79 (M. D. Jones, Shanahan, & McBeth, 2014) , or analyzing how public organizations interact with the public through social media (e.g. Pelizz a & Hoppe, 2015; Rethemeyer, 2006; Yetano & Royo, 2017) . A search of several public administration journals including Public Administration Review , Journal of Public Administration Research and Theory , Administration and Society , and Administrative Theory and Praxis revealed no studies that use d internet sources as data for institutional ethnography. Relevant to this research , Bryer and Zavattaro (2011) introduce a special issue of Administrative Theory and Praxis by documenting how social media impacts org anizational culture and administrative culture. T his study looked to other fields that have used social media as ethnographic data for analyzing social interactions and power relations (Beneito Montagut, 2011; Garcia, Standlee, Bechkoff, & Cui, 2009; Halle tt & Barber, 2014) . Given the proportion of social interactions that now take place on the internet, observing the internet footprint of organizations and individuals is a necessary part of modern ethnographic research. In the words of Hallett and Barber ( 2014) Barber (2014), Garcia et al. (Garcia et al., 2009) , Beneito Montagut (2011) , and Marshall and Rossman (2011) all argue that context from digital social interactions is necessary for understanding face to face social interactions in the real world. Of these researchers, Beneito Montagut (2011) most deliberately brings together internet and face to face interactions, f inding that the combination creates a more complete understanding of the worldview of individuals. By comparing internet derived d ata to in person interviews, this study sought to provide an understanding of the value of internet dat a in thick descriptive studies.

PAGE 92

80 Stage Two Analysis : Coding, Hermeneutics, Mapping, and Ethnomethodology When the data were collected, the researcher use d Tables 2.1 2.3 to code by Masculinity and Whiteness performances. The analysis of qualitative data for this study use d three related tools; hermeneutics, institutional eth nography, and ethnomethodology. For coding, Tables 2. 1 2. 3 provide d concrete performances of Whiteness and Masculinity in public organizations, operating like a decoder ring of systemic patriarchy and white sup remacy. The researcher code d the data into the performance categories listed in Tables 2.1 2.3 . As informants report ed performances, the researcher analyze d how these performances appear ed across public organizations in the study. At the end of the coding process , the researcher confirmed some of the performances listed in the literature review, provided deeper understanding of those performances (Chapter Five) , and analyze d each of the eight propositions to make suggesti ons for public managers looking to m anage Whiteness and Masculinity in their organizations (Chapter Six) . As discussed above, interviews were the primary sources of data for the qualitative stage of this study, while important texts such as forms and regulations, as well as internet ethnogra phy provide d further means of triangulating, challenging, and providing context for the interview data. Interviews r epresent ed organization rather than informants themselves . As the organization is the unit of analysis for institutional ethnogr aphy, t exts, interviews, and internet data were coded as representing their associated or ganization and analyzed together. To decrease bias, and deepen the analysis, the research engage d in hermeneutic practice . When a conflict arises between the perspecti ve of researcher and informant, the hermeneutic researcher identifies the assumptions that, if changed, would resolve the conflict between the

PAGE 93

81 perspectives (Debesay, Nåden, & Slettebø, 2008) . This practice of understanding a conflicting perspective enables the researcher of racial and gender phenomena to understand how both researchers and informants make sense of race and gender (Matias, 2016) . Simply identifying the conflict, however, does not resolve conflicting perspectives. The standard suggested by ma ny hermeneutic philosophers for resolving conflicts between informants, researchers, and research recipients is to defer to informants as experts in their own worldview (Debesay et al., 2008; Gadamer, 1976) . However, critical hermeneutic researchers analyz e conflicts in understanding created by prejudices and discrimination that limit (Mills, 1997; Zuberi & Bonilla Silva, 2008) . Critical scholars argue that the normative values determining understandin g in most studies are those that support Whiteness and Masculinity (Pini & Pease, 2013; Zuberi & Bonilla Silva, 2008) . In other words, those studies that do not deliberately select race and gender objectives, instead select white supremacy and patriarchy b y default. Stage two resolve d conflicts in understanding by deference to transparent assumptions that white supremacy and patriarchy inflate the value of whites and men (L. A. Bell, 2013; Harro, 2013) . After each interview, and at least weekly during the analysis process, the researcher record ed a journal identifying conflicts of understanding with informants, findings, and the literature, and attempt ing to resolve those conflicts based on this assumption of equality. This journal serve d as a tool for iden tifying coding bias, and develop increased reliability. After the interviews and supplemental texts were coded, the study develop ed an institutional map of Masculinity and Whiteness performances. To better understand how institutions embed in public organi zations, Smith (2006) suggests using S. method of mapping social relations. These maps document work process es by tracking the

PAGE 94

82 multiple sources of meaning and knowledge moving through the organization. This study focused inforcing and maintaining Masculinity and Whiteness. In Chapter Six, the institutional map is presented and analyzed for places where the work of Whiteness and Masculinity can be disrupted or modified. To understand how informants create d meaning around ra ce and gender, the research adopt ed an ethnomethodological perspective. Ethnomethodologists analyze how people make (Liberman, 2013) . In (1963) early studies, the in stitutional breach could be something like answering These breaches were meant to force informants to process their everyday practices (Arminen, 2008; Pollner, 20 12) . As the informant reconstructs their institutional reality, the ethnomethodologist collects data on the pro cess through which reality is re constructed and reasserted (Garfinkel, 1963; Pollner, 2012) . The method used by Matias and Mackey (2016) and Boni lla Silva (2003) is understood in this study as an example of ethnomethodology. In these studies, an outwardly racial or gendered question in an environment of race and gender norms (Mills, 1997) , and the info rmant must then reconstruct their institutional reality to create the least possible dissonance. This study use d perform when they recreate d race and gender after being asked questions that challenge d race and gender blind beliefs . Thus, ethnomethodology was used to bring out performances of Whiteness and Masculinity for coding . Once the data are coded into race and gender performances, the resulting coded data and organization maps were used to anal yze each of the e ight propositions from Chapter T wo. While this proposal outlined 38 Masculin ity and Whiteness performances , t he analysis process reveal ed

PAGE 95

83 relevant performances that were not initially theorized . These performances were initially coded sepa rately , coded emergently, and used to analyze the eight propositions accordingly. Eventually, enough data was collected to give evidence supporting, partially supporting, or not supporting each of the propositions, creating a deeper understanding of the me chanisms that develop and reinforce Mas culinity and Whiteness in the federal government . Ethical Considerations Like all research with human informants, this study has some minor risks to participants that should be acknowledged. Words like racism, sexism, white supremacy, and patriarchy are technical terms in critical whiteness and multiple masculinities research, but are loaded terms to most people. It is this same loaded terminology that could produce d a slightly larger risk for participants in this stud y. Miscognition surrounding Masculinity and Whiteness create s an additional ethical dilemma, which comes from the fact that informants respond ed with hostility to terms like Whiteness and Masculinity, thereby making the findings less beneficial. This study attempt ed to get informants to speak about race and gender in the workplace without triggering their defensive performances. To achieve this delicate balance, the researcher present ed the research to Masculinity. This technically correct statement provide d the necessary information to informants without risking undermining the results that have the potential to benefit all people, including white men in public orga nizations. All documents were encrypted using the encryption software, Boxcryptor , and only the researcher has access to the decryption key. Texts and social media printouts that include d identification of individuals were modified so that faces,

PAGE 96

84 names, and other identifying characteristics were blurred or encrypted. The encrypted data were stored on a secured CU Denver cloud drive, which is accessible only to the researcher. Nobody apart from the researcher had access to the raw da ta or the encryption key . All results were presented anonymously . Stage Two Limitations The qualitative portion of the research is limited in three main case selection is often criticized because of the heterogeneity of populat ions. In the words of Henry (1990, p. 21), typical cases case selection was aided by the quantitative data confirming that the o rganizations selected fit the profile of a typical office in the federal government. A second limitation derives from the assumption that a person who is not a white man will act as a reliable check on ho w white men understand organizational race and gender performances. The literature indicates that women and people of color must adopt performances of Whiteness and Masculinity to survive in modern society (Carbado & Gulati, 2013; Ferguson, 1984) . While us ing a second informant verified the experiences of the white men informants, there is no guarantee that these individual women and people of color perform ed Whiteness and Masculinity differently than their white men colleagues , and therefore no guarantee t hat data from women and people of color informants provide d meaningful triangulation of data . Another limitation that needs mentioning is the threat to validity caused by President Trump. Between the data collection of the FEVS (OPM, 2016) in 2004 2015, an d the qualitative data collection in 2017 their job and its race and gender performances. While it is too soon for academic research to

PAGE 97

85 have been published, evidence of a possible shift news media. Trump appointed more white men to his cabinet than any president in modern history (Lee, 2017) people of color , both performances of Whiteness documented in the literature (Thrush & Steinhauer, 2017) . Two more early appointments include d Stebastian Gorka and Steve Bannon both of whom left Breitbart New s to work for the white house (and have since returned to Breitbart) . Breitbart News is (Fortin, 2017) . The term alt right was coined by white supremacist Richard Spencer as a way to j oin a st (Florido, 2016) . The alt right is also vocal about the proper position of women as housewives and objects for men to lust after, arguing that the reproductive organs of white women should be under the control of white men (Darby, 2017, p. 26) General Jeff Sessions was rejected by the 1986 Senate for a judgeship in part because of his racist opposition to protections to black voting rights (McCann, 2017) . Now able to secure his appointment, Sessions pursue d action practices under the racial projection claim that whites are being harmed by admission policies (Savage , 2017) despite the continued over representation of whites in colleges across the country (Ashkenas, Park, & Pearce, 2017) . And signaling a masculine hurdle to women who want to work in the Trump administration, Vice President Mike Pence will not have a m eal with a woman unless his wife is also present (Parker, 2017) . To account for the fact that these examples now r epresent the leadership in the f ederal workforce, this study included a question about what has changed in the office since Trump was elected.

PAGE 98

86 Conclusion This study has implications for public administration research on race, gender, diversity management, representative bureaucracy, and social equity. The Critical Whiteness and Multiple Masculinities literature on which the assumptions of this study rely is often either theoretical or experimental. Very little of the extant research is situated in a public organization, and very little uses institutional ethnographic methods to understand race and gender as the consequences and causes of social interactions in organizations. Methodologically, this study adopts these methods, not because they are new and underutilized, but because methodologists in the social sciences are convincing that these methods are the best suited for understanding the impa ct of institutions embedded in organizations. In countering the institutional miscognition surrounding Whiteness and Masculinity , it is often necessary to use non standard methods because these same institutions were involved in the development of standard methods (Pini & Pease, 2013; Zuberi & Bonilla Silva, 2008) . The methods described here were selected to expose the institutional effects of Whiteness and Masculinity on public organizations and provide a foundation for making recommendations to manag e tho se effects.

PAGE 99

87 CHAPTER FOUR MASCULINITY AND WHIT ENESS IN FEDERAL WORK UNITS The quantitative findings reveal an outline of Whiteness and Masculinity that can be unexpected and nuanced. For example, clustered logistic regressions reveal that Whiteness in manag ement has a more profound impact on satisfaction than Masculinity, but that both institutions impact public service motivation (PSM). Additionally, machine learning analysis of work units reveals central relationships between Whiteness and alternative work arrangements, and Masculinity and merit systems that are deemphasized in the literature. This chapter begins with an exploration of the Federal Employee Viewpoint Survey using factor analysis and comparative regression analysis between people who perform Whiteness and/or Masculinity by indicating that they are whites and/or men, and those who do not. The next section looks at the impact of overrepresentation of whites and men in management on satisfaction and public service motivation (PSM) among federal w orkers, directly addressing the hypotheses stated in Chapter Three. After that, a machine learning tool reveals correlations that other methods sometimes miss. The decision trees discover interesting relationships between alternative work arrangements and race, and abstract objectivity and gender that will be explored further using qualitative methods in Chapter Five. Summary Statistics: Disentangling Masculinity and Whiteness Factor analysis revealed that the job satisfaction variable initially created in Chapter Th ree , and which is used in some academic articles using the FEVS (e.g. Weaver, 2015) , was more defensible as two different variables. Initially, the job satisfaction variable captured overall employee satisfaction by indexing seven questions. This index had an acceptable alpha of .884. F actor analysis after varimax rotation revealed that job satisfaction fit better as two indices. The

PAGE 100

88 first variable is a job satisfaction index includ ing satisfaction with traini ng, promotional opportunities, recognition for doing a good job, involvement in decision making, that their talents are well used, likelihood of recommending the a h igher alpha of .91 5, with an Eigenvalue of 4.189. The second factor discovered in this process included seven questions about employees perceptions of satisfaction with supervisors support of work/life balance, commitment to representative bureaucracy, re spect for employees, opportunities to demonstrate leadership, supervisors listen ing abilities , supervisor supervisor s , and immediate supervisors quality of work. This second variable capture s satisfaction with management with an alpha of .953 and an Eigenvalue of 5.278. Both dependent variables were of interest for this study . As described in Chapter Two, studies indicate that the race and gender of management likely impacts the job satisfaction of employees, bu t little is known about whether these employees are satisfied over their working conditions or the quality of their management. Job and Management Satisfaction White men are on average less satisfied with their jobs than white women and men of color, but n ot women of color (Table 4.1) . Fixed effects were used to predict the means of each dependent variable categorically by the intersectional race and gender of the respondents . The models included control variables for respondents over 40 years old, military status, and whether they have been in government for more than 10 years. The four categories were white men, white women, men of color, and women of color, with clustered standard errors to account for unit level effects, and fixed effects to account for time. This analysis indicates that t he mean

PAGE 101

89 Table 4.1 Unit clustered categorical predicted means after controlling for time, age, military status, and tenure in federal government work. Centered Variables Job Satisfaction Satisfaction with Management Pub lic Service Motivation Women of Color .085 .731*** 1.677*** Men of Color .423*** .063*** 3.128*** White Women 0.062*** .241 .825*** White Men (Omitted) 0.068 .361 2.563 Numbers represent t he centered predicted response. *=p<.05, **=p<.01, ***=p< .001. White men is the constant. Table 4.2 Unit clustered categorical predicted means after controlling for time, age, military status, and tenure in federal government work. Variables Abstract Objectivity Diversity Management Childcare Alternative Wor k Arrangements Women of Color .004*** .604*** .111*** .272*** Men of Color .243*** .190*** .123*** .204 White Women .056* .153*** .143 .238*** White Men (Omitted) .093 .336 .135 .198 Numbers for abstract objectivity and diversity management r epresent the centered predicted response. Numbers for childcare and alternative work arrangment are coded as 1=not available, 0=available but not used, and 1=used by respondent. *=p<.05, **=p<.01, ***=p<.001. White men is the constant.

PAGE 102

90 predicted j ob satisfaction for white men is statistically significantly lower than for white women or men of color at the p <.001 level. Men of color are more satisfied with their jobs than any other group, with a mean centered satisfaction index of posi tive .423 on a 28 point scale. Women of color experience no statistically significant difference in job satisfaction than white men. While white men indicate less than the mean job satisfaction, the model predicts that their satisfaction with their managers is above the mean . This is particularly interesting given that these two variables were initially combined, and similar measures have been combined in published research. This finding indicates that the racial influence on job satisfaction and management satisfaction partially cancel s each other out, thereby creating risk of t ype II error. Unlike job satisfaction, race is a bigger influence on management satisfaction than gender. Overall, w hite people are happier with their managers than people of color . The average pe rson of color indicates lower than average satisfaction with management, while white people indicate higher than average satisfaction with management. M en of color indicate a mean management satisfaction that is .063 below the mean of the sample, while wom en of color indicate a mean management satisfaction that is even lower at .731. In both cases this difference is significant at the p <.001 level, and the significance does not drop off after accounting for time, unit level clustering, military status, age of the respondent, or their tenure in the government. These findings indicate that satisfaction with management is partially driven by racial relations in federal agencies. Public Service Motivation White men are statistically significantly likely to indi cate lower than the mean public service motivation, with a mean coefficient of 2.563 out of the original five point scale described in Chapter Three . Women and men of color on average indicate above the mean PSM ,

PAGE 103

91 with white women and women of color at .82 5 ( p <.001) and 1.677 ( p <.001) respectively. The group that indicates the highest level of public service motivation in the FEVS is men of color, who are at 3.128 above the mean ( p <.001). This indicates that the impact of race and gender on PSM is not linea r as the institutions intersect. Additional Independent Variables The researcher constructed additional variables to capture additional performances of Whiteness and Masculinity indicated by the literature . The literature overviewed in Chapter Two describe s multiple reasons why childcare and alternative work schedules would support women and people of color more than white men , including social expectations that women are primary caregivers, and wealth inequities that make people of color more responsible f or their parents and children . To look analyze these mechanisms , the FEVS includes question asking the respondent to indicate their access to employer supported childcare and alternative work arrangements including work from home and flexible work hours. T hese variables were coded so that a person who uses the service is identified by a 1, a person who has access to the service but does not use it is coded as 0, and a person who has no access to the service is coded as 1. This coding enables researchers to understand not only if the option was available in the abstract, but the extent to which the option was in fact available based on the working conditions and organizational culture within the unit. Using fixed effects to account for unit level and time cl ustering and the same control variables as above , categorical fixed effects regressions predict ed the mean likelihood that white men, white women, men of color, and women of color were using these services. W hite people are statistically significantly less likely to use employer provided childcare than people of color at the p <.001 level, and men are significantly less likely to use alternative work arrangements than women at p <.001 (Table 4. 2 ).

PAGE 104

92 The findings are somewhat surprising given literature indicati ng that childcare is primarily a gendered activity. In fact, in the federal government, this model indicates that using childcare benefits is more closely related to race than gender, which may be linked to wealth as a sociological performance of Whiteness . Generally, w hite people have more wealth than people of color , and this remains true even after accounting for income (Oliver & Shapiro, 2006) . Wealth is often intergenerational, which mean s that white parents are more often able to use their retired par ents to support them in childcare, or to send their children to a private childcare center. While the model indicates that childcare was mostly differentiated by race, alternative work arrangements were more closely associated with gender. This finding is in line with literature indicating that women undertake more responsibility for childcare than men even when they have fulltime jobs , and literature showing that alternative work arrangements are often used to care for children (e.g. Franzway, 2001; Gallup , 2017) . Another variable indicated by the literature and deduction is the existence of diversity management , which would be predicted to decrease Whiteness and Masculinity in work units . Th ree questions measure diversity management looking at the ability of supervisors to work with employees of different background, the vigorousness of responses to prohibited workplace discrimination, and the level of satisfaction with polic i es and program s that promote diversity. These three questions had an alpha of .803 1, and an Eigenvalue of 1.6. Centering the variables aided in interpretation so that positive numbers indicate better than average perceptions of diversity management. The model finds that w hite men are statistically significantly more likely to feel satis fied with diversity management in the federal government, and that it is both a racial and gendered performance ( T able 4. 2 ). White women federal employees are more satisfied with diversity

PAGE 105

93 management than average, but they are significantly less satisfied than white men at the p <.001 level. Men of color are less satisfied with diversity management than average, and significantly less satisfied than white men at the p <.001 level. But the people who are the least convinced that their units promote diversity, prohibit workplace discrimination, and support employees of different backgrounds are women of color ( p <.001 compared to white men). The last variable captures the level of abstract objectivity found in the work unit . This variable combined four questions from the FEVS. 1) Promotions in my wo rk unit are based on merit. 2) I n my work unit, steps are taken to deal with a poor performer s who cannot or will not improve. 3) M y performance appraisal is a fa ir reflection of my performance. 4) M y work unit is able to recruit people with the right skills. The fourth variable was added to the index after factor analysis identified it as hanging on the same factor. Overall, the index had an Eigenvalue of 1.814, and a n alpha of .779. Centering the variable aided interpr etation. The model indicates that w hite men perceive their organization as statistically significantly less objective tha n any other group (T able 4. 2 ). White women are slightly more satisfied ( p <.05) with the objectivity of their performance metrics, but l ess satisfied than the average federal worker. Women of color are slightly under the mean , and statistically significantly more likely to indicate satisfaction with merit processes than white men at the p <.001 level. Meanwhile, men of color are the only gr oup to indicate a higher than average satisfaction with the merit processes of their organizations ( p <.001), sitting at the pole exactly opposite white men. This difference persists even when accounting for military status as a control variable, as well as time and unit level mixed effects. Taken together, these findings on abstract objectivity and diversity management tend to indicate that white men believe that the system is tilted against them, and that the federal

PAGE 106

94 government has done more than sufficien t work to create opportunities for women and people of color . Individual Level Logit Analysis In the end, the most robust, theoretically sound, and conservative model used to test the effects clustered logit regression . Data pr oblems described in Chapter Three included heteroscedasticity and non normality that limited the validity of those initial tests. To account for those limitations, the dependent variables were made binary by creating dummy variables around the mean, theref ore a 1 observation for job satisfaction is any answer that indicates above the mean response, and 0 indicates the mean or below. Another reason to rerun these tests is that additional years of data have since been released, including both 2016 and 2017, w hich provides more context, variability, and brings the research into the first y ear of the Trump presidency . The updated data includes many more unit identifiers, bringing the total organizational count to between 309 and 288, depending on which units inc luded the relevant demographic data. These updated cluster variables give more opportunities to avoid type I error, and so the findings will be more conservative. Attempts at multilevel modelling on the individual level data failed. M odels including the in dependent variables of interest failed to converge despite exploring many optimizers, centering all variables, simplifying models, and consulting several texts. Models with only the control variables converged, and each time indicated similar findings to f ixed effects models of the same variables, while indicating that the clustering effects had no statistically significant impact on the dependent variables. It may be that models with the independent variables did not converge because the interaction term o ratio of whites/men/white men in management was simultaneously at two levels. The individual

PAGE 107

95 demographic was at the individual level, while race and gender in management was at the unit level. This may have constituted a violation of assumptions that undermine d the validity of multilevel models and prevent ed them from converging. The first tests find that job satisfaction is impacted by race, but not gender (Table 4.3). The base model on job s atisfaction commands a respectable .393 pseudo R squared and finds that respondents who feel that they have more adequate resources are more than 60 percent more likely to indicate above the mean job satisfaction ( p <.001). Given that organizational resourc es is a 20 point scale, and each point represents a 60 percent increase in odds that the respondent is above the mean satisfied with their job conditions, the organizational resources variable is highly statistically and practically significant. Meanwhile, employees who indicate a one point increase on a five point scale of office cleanliness, quiet, comfort, and lighting are about 25 percent more likely to indicate above the man job satisfaction ( p <.001). These two variables explain a great deal of the var iance in job satisfaction, but the analyses do indicate racial differences as well. White people are statistically significantly more likely to indicate above the mean job satisfaction when a larger proportion of their supervisors are white. Identifying as white is statistically significantly related to a 17 percent increase in odds of indicating above the mean job satisfaction after accounting for overrepresentation of whites in management and the effect that has on white people ( p <.001). This finding cont radicts the earlier finding that whites are less satisfied with their jobs than most other groups except women of color and may be related to the effects of white overrepresentation in management. While a larger ratio of white supervisors does not predict job satisfaction in general, it does for white employees. Confirming hypothesis 2a, the model finds that a white employee in a work unit with an even ratio of white supervisors

PAGE 108

96

PAGE 109

97 to white employees has more than a 50 percent lower odds of indicating abov e the mean satisfaction than a white employee with 25 percent too many white supervisors ( p <.001). Interestingly, gender congruence and race/gender congruence among supervisors do not hold statistically significant relationships with job satisfaction in th ese analyses, thereby allowing me to reject hypotheses 2b and 2c. Satisfaction with management is similarly impacted by racial congruence between white supervisors and white employees, but not gender or race/gender congruence of white men (Table 4.4). Agai n, the organizational resources and hygiene variables are highly significant, together accounting for a greater than .19 of the pseudo R squared in each of the models. A one point increase on a 20 point scale of organizational resources is statistically s ignificantly ( p <.001) related to a 30 percent increase in the odds of indicating above the mean satisfaction with management. Hygiene factors like good lighting, quiet, and adequate technology account for an additional 14 percent increase in odds of above the mean satisfaction with management ( p <.001). Identifying as a white/man/white man increases the odds of indicating above the mean satisfaction with management, but overrepresentation of white supervisors increases the odds of above the mean satisfaction with management for white employees only. Overall, identifying as white increases the odds of above the mean satisfaction with management by more than 35 percent ( p <.001), while the increase in odds for men is more than nine percent ( p <.05), and more than 24 percent for white men ( p <.001). This is consistent with the categorical variable findings discussed above (Table 4.1). When white people are in an organization with a 25 percent increase in the proportion of white supervisors, they are statistically si gnificantly ( p <.05) likely to experience an additional 14.75 percent increase in the odds that they will indicate above the

PAGE 110

98

PAGE 111

99 mean satisfaction with their more white managers, thereby giving additional confirmation to hypothesis 2a. Once again, this a nalysis cannot support hypotheses 2b and 2c that Masculinity and White Masculinity in management have similar impacts on men and white men employees. owerThe model on management satisfaction as related to race includes a troubling finding. After accounting for the increase in management satisfaction among whites, and the interaction with the proportion of supervisors who are white, organizations with a 25 percent higher relative proportion of white supervisors is related to a 22.28 percent ( p <.05) increase in odds of indicating above the mean satisfaction with management in the work unit generally. Three theories may explain this finding. Stereotyping, intuitive assessments, and stereotype promise explain how perceptions of white managers could be inflated c ompared to their people of color peers. Also, because of these same psychological phenomena, white managers may be able to use their power to secure better treatment for their employees, people of color and white. Third, while this analysis clusters for or ganizational and time characteristics, the results may be biased by unobserved factors. When it comes to public service motivation (PSM), the findings support hypotheses 1a, 1b, and 1c in Table 3.1. Overall, whites/men/white men indicate less PSM than wom en and men of color, consistent with the categorical findings discussed above. Whites have seven percent lower odds ( p <.01), men have 20 percent lower odds ( p <.001), and white men have more than 27 percent lower odds ( p <.001) of indicating above the mean P SM (Table 4.4). But these lower odds are mitigated when whites/men/white men are in units with more white/men/white men supervisors. I find that a 25 percent increase in the proportion of white/men/white men supervisors is related to an increase in the odd s of above the mean PSM by 12.175 percent for

PAGE 112

100

PAGE 113

101 white employees ( p <.1), 6.6 percent for men employees ( p <.1), and 11.1 percent for white men employees ( p <.05). Overall, the individual level quantitative analyses indicate that Whiteness in management in creases comparative white job and management satisfaction, and both Whiteness and Masculinity in management improve comparative PSM for whites, men, and white men. Therefore, hypotheses H1a c and H2a are confirmed, while the regression testing hypotheses H 2b c failed to reject the null that men and white men over representation in management increase satisfaction for men and white men employees respectively (Table 4.6). These findings support literature indicating that the effects of race are statistically small but broad (Greenwald, Banaji, & Nosek, 2015) , while the effects of gender in management are more nuanced. The next section explores organizational effects in more detail using a machine learning technique commonly referred to as decision trees . Deci sion Trees: Machine Learning about Whiteness and Masculinity Decision trees, or classification and r egression trees (CART) are machine learning technique s for exploring relationships between variables. ANOVA regression trees are a form of CART that use ite rative analysis of variance (ANOVA) on a single dependent variable, exploring multiple independent variable s sequentially for statistically significant correlations with the dependent variable (Therneau & Atkinson, 2018). As the iterative ANOVA regressions run, statistically significant splits emerge that predict the dependent variable, and the observations in each split are subsequently analyzed to see if another split emerges with the same or a different independent variable. The result is a figure that r esembles an upside down tree with the independent variables that are the most relevant to the dependent variable listed on top of the tree . V ariables that have more limited impact are listed further down the tree until no statistically

PAGE 114

102 Table 4.6 Table con firmation or rejection of hypotheses. Hypothesis Finding H1a: When whites are over represented in supervisorial teams, whites in those organizations will exhibit more PSM. Confirmed H1b: When men are over represented in supervisorial teams, men in thos e organizations will exhibit more PSM. Confirmed H1c: When white men are over represented in supervisorial teams, white men in those organizations will exhibit more PSM. Confirmed H2a: When whites are over represented in supervisorial teams, whites in t hose organizations will be more satisfied. Confirmed H2b: When men are over represented in supervisorial teams, men in those organizations will be more satisfied. Failed to Reject Null H2c: When white men are over represented in supervisorial teams, whit e men in those organizations will be more satisfied. Failed to Reject Null

PAGE 115

103 significant relationships remain (for examples, see Figures A1 A9) . At the bottom of the tree are that have the characteristics described in the analysis. Regression trees have several b enefits for this study . The first is that the FEVS is administered on line, and so has missingness issues as respondents are distracted or tire of completing the 95 10 7 questions on the survey. The RPART package in R compensates for this by predicting the missing variable and reporting on patterns in the missingness that are biasing dependent variables (Therneau & Atkinson, 2018) . When it is predicted that missingness i s biasing the predictions, the observation is deleted by the machine learning algorithm. In my analyses, this led to between 404 and 132 observations eliminated due to bias detected in missingness. Therefore, the number of observations ranged from between 1,188 and 1,460 work unit/years per regression tree. A second reason to use decision trees is that they account for curvilinear relationships through the process of splitting the independent variable relationships and breaking down each of the larger spli ts into smaller splits that may go in opposite directions. If there are meaningful curvilinear trends, ANOVA regression trees will reveal them . The third reason to use decision trees is that CART enable s exploration of independent variables that may be mec hanisms of primary variables, facilitating testing on abstract objectivity , diversity management, alternati ve work, and childcare variables, all of which are potential mechanisms of Whiteness and Masculinity in federal work units. Decision trees suffer tw o big limitations. First, the method is not suited to causal analysis. Just as ANOVA is a method that can only capture correlation, so ANOVA regression trees can only capture correlation between the dependent variable and a host of independent variables. T he

PAGE 116

104 qualitative methods described in Chapter Five will do the work of establishing causal connections to the extent possible. The second major limitation is that decision trees are difficult to interpret because of the complexity of the results. For that re ason, a long format study is an ideal medium for piecing apart the detailed findings, with reconstructed trees documented in detail in the Appendix (Figures A1 A9). CART is useful for exploring many independent variables at once without degrading the parsi mony of the model, because at each stage only the most influential independent variable is retained before proceeding with tests on subsets of the data. Taking advantage of this benefit of CART, regression trees were run on unit level job and management sa tisfaction , as well as PSM dependent variables. The mean job satisfaction of people of color employees in each unit was subtracted from the mean job satisfaction of white employees. The process was repeated for men and for white men and similar variables g enerated for management satisfaction and PSM in addition to job satisfaction. While these variables are only proxies, they should theoretically be able to observe institutional benefits to whites and men in work units, and so are useful dependent variables for testing unit level Masculinity and Whiteness. If men are comparatively more satisfied with their jobs and management and more motivated than their women colleagues, this proxy theoretically captures some aspect of masculinity. ANOVA regression trees a nlayzed nine dependent variables including comparative job satisfaction, management satisfaction, and PSM across each race, gender, and raced gender. Overall, the results indicate that race and gender operate differently on job satisfaction, management sat isfaction, and PSM, but that the differences between PSM and satisfaction for each group are less significant (Figure 4.1 )

PAGE 117

105

PAGE 118

106 CART findings on comparative white job satisfaction, management satisfaction, and PSM. Decision trees indicate that the single gr eatest influence on the comparative satisfaction and PSM of white employees is whether they work in a defense unit (Figure 4.1 and Figures A1 A3) . Defense units have less average comparative white job satisfaction than civilian agencies (Figure A1). In civ ilian agencies, units with higher perceptions of diversity management and higher levels of access to alternative work schedules , indicate a net positive comparative white job satisfaction (55.3% of units) . U nits that indicated lower levels of perceived div ersity management and less access to alternative work arrangements joined defense agencies in having a predictive mean white job satisfaction of .068 (Table 4.1) . This finding supports the argument that the combination of high diversity management and low alternative work arrangements is practically as well as statistically significant. Civilian units with high perceived diversity management and low alternative work arrangements have lower levels of Whiteness, at least as measured by comparative job satisf action. When running decision trees on management satisfaction, again the single most significant predictor of comparative white satisfaction was whether the unit was in a defense agency. White perceptions of management in defense agency units were compara tively lower than in civilian agencies (Table A2). The only units that indicated higher perceptions of management for people of color than whites were those defense units that have higher access to alternative work arrangements, and over half of employees in the first ten years of their tenure with the government (3.4% of all units). On the opposite pole, the units that indicated the highest comparative white management satisfaction were those defense units with lower alternative work arrangements , a higher ratio of men in management, and low perceptions of diversity management (1.3% of all units). Contrarily, the plurality of units in the dataset were civilian

PAGE 119

107 units where whites were more than 2.94 percent overrepresented among supervisors with greater acce ss to alternative work arrangements, and higher perceptions of diversity management (30.2%) (Figure 4.1). In these units, the comparative white satisfaction with management is predicted to be relatively high (1.12). The findings for job satisfaction and ma nagement satisfaction both tend to support the proposition that units offer ing higher levels of alternative work arrangements with supervisors who are perceived to work well with diverse employees , do not tolerate prohibited practices, and promote diversit y are among the units with the largest advantages for comparative white satisfaction rates. This presents a puzzle that will be explored further in Chapter Five. Just like job and management satisfaction, units in defense agencies were overall statisticall y significantly more likely to indicate lower comparative levels of white PSM (Figure A3 ). Of those units in defense agencies, younger units (2.1% of all units) indicated substantially lower comparative levels of white PSM than the units that were relative ly older (37.4% of all units) . In civil ian agencies, white PSM was comparatively the lowest in the units with lower perceptions of diversity management (1.2% of all units) , once again confirming that there is a problematic relationship between diversity ma nagement as it is commonly understood and the comparative satisfaction and motivation of whites and people of color . Of those units in civilian agencies with higher perceptions of diversity management, lower perceptions of abstract objectivity are statisti cally significantly correlated to somewhat lower comparative PSM for whites ( 2.92) . This group represents a plurality with 38 percent of all units in the dataset. The decision tree reveals a curvilinear relationship between the overrepresentation of whit es in supervisorial roles and PSM. A small subset of units (35 of 1,188) have more than 29 percent too many white supervisors. C ivilian units in this subset with higher perceptions of

PAGE 120

108 diversity management, and higher perceptions of abstract objectivity , in dicate low l evels of comparative white PSM ( 5.93). However, 19 percent of units have these same features but with fewer than 23 percent too many white supervisors , and they have a comparatively lower predicted white PSM . Interestingly, in the 1.3 percent of units where abstract objectivity and diversity management are higher, and white supervisors are between 23.5 and 29 percent over representative, white employees indicate higher PSM than their people of color co workers. The limited parameters of this cu rvilinear relationship may be the reason that quadratic regression tests did not reveal this finding. Overall, the biggest factors increasing comparative white satisfaction and PSM in the federal workforce are 1) working in a civilian agency, 2) higher per ceptions of diversity management, and 3) greater access to alternative work arrangements. Perhaps surprisingly, the extent to which there were a greater ratio of white supervisors than white employees was a relevant, but secondary influence. Also surprisin gly, working in a defense agency was a strongly mitigating factor in the differences experienced between white and people of color employees. This phenomenon should be further explored in later research. Another relationship that will be further explored i n the institutional ethnography section below is the counter intuitive relationship between alternative work schedules and increased comparative white satisfaction . T he decision trees analyzing racial effects on satisfaction and PSM were considerably diffe rent than those analyzing gender. comparative job satisfaction, management satisfaction, and PSM. was the most complex tree in the study, discovering 10 layers of statistically significant splits in the data, including four curvilinear relationships. The simplest relationship is a straightforward

PAGE 121

109 confirmation of a great deal of the literature on gender suppression in organizations (Lindstead, 2000) . When units report lower levels of abstract objectivity in hiring, promotions, discipline, and performance appraisals, men in that unit indicate lower comparative levels of job satisfaction . For most units (52 percent) the average man in the unit indicates more job satisfaction than the average woman in the same unit (Table A4) . These are unremarkable units in civilian agencies with greater than 24 percent and fewer than 61.2 percent of employees in the government for less than ten years , and more than 11.5 perc ent of t he units under 40 years old. When these unremarkable unit s are among the 97% least over representative of men supervisors , men are more satisfied than their women co in management gets above 30.7 per cent, the overrepresentation of white managers becomes salient to the gendered job satisfaction . In these units, when the race of supervisors is relatively congruent , men show much more comparative job satisfaction . At the same time, in those same units wi th between 3.6 and 14.2 percent overrepresentation of white supervisors , women show greater satisfaction than men. Another interesting finding from this regression tree is t hat a small subset of units (2% of all units ) with a small, but not insubstan tial n umber of employees under 4 0 years old indicate higher job satisfaction for women than for men, particular ly with more men managers (>18.1% overrepresentation). It is noticeable that this decision tree contrast with racial job satisfaction by the primacy of abstract objectivity, and the absence of alternative work arrangements and the military variable. management satisfaction , the first level finding is again abstract objectivity (Figure A5) . Higher levels of abstract objectivit y predict higher levels of management satisfaction. While pay and having started in federal service

PAGE 122

110 fewer than 10 years earlier has an impact, once again the finding that best predicts greater comparative satisfaction for men is whether t he employees in that unit find more objectivity in promotion, hiring, discipl ine, and performance appraisals. As with job and management satisfaction, m by abstract objectivity . Men show significantly more PSM com pared to women when their units perceive promotion, hiring, discipline, and performance appraisals as more objective (Figure A6). Abstract objectivity interacts in multiple ways with other relevant variables . For example, in units where 76.3 percent of emp loyees have more than 10 years of government experience (18.2% of all units) , less abstract objectivity predicts lower comparative PSM for men, and better perceptions of diversity management further diminish the compara tive PSM of men . A plurality of units (34.1 % ) fall into a leaf where women and men are predicted to have relatively similar PSM . These units in defense agencies show the highest levels of abstract objectivity measures, greater than 23.7 percent employees with less than 10 years of experience , and fit in the bottom 97 percent of overrepresentation of whites among managers. Two leaves in this regression tree speak to the limited relevance of some of the factors that are most discussed in the literature on gender in organizations. The leaf that indicates the greatest comparative PSM for men includes civilian units in the top four percent of overrepresentation of men supervisors, with h igher levels of abstract objectivity (.8% of all units) . Two more leaves highlight how child and eldercare are s ignificant contributions to and eldercare facilities is correlated to lower comparative PSM of men in civilian units where 70.9 percent of the workforce is over 4 0, the super visors are in the 96 percent least over representative of men , and abstract objectivity is relatively high (23.2% of all units) . In these same units , but with lower

PAGE 123

111 levels of access to child (5% of all units) . These leaves combine d show that women in roles of authority and access to family care support family are correlated to Decision satisfaction, satisfaction with management, and PSM indicate that abstract objectivity impacts the work place to the benefit of men. This finding quantitatively supports the work of many scholars who argue that merit is by default a masculine concept (J. M artin, 1990; Mastracci & Arreola, 2016; Stivers, 1993) . These findings provide focus for a subsection of the institutional ethnography in Chapter Five . C ART findings on comparative satisfaction, and PSM. As with men in general, white men are less comparatively sa tisfied with their jobs in the five percent of units that are perceived as having the least abstract objectivity (Figure A7) . White men in the 58.5 percent of units that are civilian and have higher levels of abstract objectivity show no statistically significant difference between white men and women/ people of color . Working in a defense unit predicts lower comparative job satisfaction levels for white men, with even lower comparative job satisfaction in unit s with greater than 36.3 percent under 40 years old. When it comes to satisfaction with management, the decision tree for white men looks like a combination of the tree s for whites and for men (Figure A8) . Being in a defense agency was a primary predictor of comparative white man management satisfaction. The 2.2 percent of p eople in defense agencies that show lower levels of abstract objectivity are predicted to have the lowest comparative white man satisfaction with management of all units in the dataset . However, 32.7 percent of all units in the dataset were in defense agencies with fewer than 55 percent newer

PAGE 124

112 employees, more than 29.1 percent higher than GS 9 paygrades, higher abstract objectivity , and relatively high comparative management satisfaction f or white men. Most units (58.4%) are in civilian agencies, have relatively higher levels of diversity management, and have relatively higher access to elder and childcare. These units indicate far more comparative satisfaction with management among white m en than among people of color or white women. Interestingly, the lowest comparative satisfaction for white men is in civilian units with higher levels of diversity management, and relatively lower access to elder and childcare, though these units only repr esent .6 percent of the dataset. When it comes to comparative PSM for white men, abstract objectivity is the most predictive factor, with diversity management coming in to play only for units with lower predicted objectivity in promotions, hiring, discipli ne, and assessment (Figure A9) . In units with lower perceptions of abstract objectivity , white men indicate comparatively low PSM, but diversi ty management somewhat increases their comparative PSM. The largest leaf (35.1% of all units) indicate s somewhat l ess PSM for white men than other people, but still more comparative PSM than any of the units with lower levels of abstract objectivity . In these units, a slightly lower PSM is predicted for white men than for women and people of color when 96.7 percent of those units have more than 10 years of experience, fewer than 47.6 percent are paid at GS 9 or higher, and abstract objectivity is relatively low . Like other decision trees in this study, the 14 percent of units with the highest levels of perceived abstra ct objectivity showed the least people of color and women . Like ANOVA trees on raced and gendered dependent variables, CART tests on white men indicate some themes to explore further . Some of the most prominen t variables are expected and predicted by the literature, such as abstract objectivity , as well as diversity management,

PAGE 125

113 representativeness of managers to employees, and the availability of childcare and alternative work arrangements. Some variables that s eem like control variables, such as the proportion of a unit that is under 30 years old, or that the unit is in a defense agency, require deeper analysis to understand how such factors correlate to differences between the experiences of whi tes and people o f color , and men and women. Conclusion This chapter reveals that Masculinity and Whiteness is having statistically significant effects on how federal workers experience their public service. In answer to the quantitative research question asking what effec work life, the quantitative findings reveal several notable variables that are significant. White men indicate significant differences in perception of job satisfaction, satisfaction with management, P SM, alternative work arrangements, abstract objectivity, diversity management, and childcare supports. Whites are more satisfied when their managers are also white, and white men show higher PSM when their managers look like them. Work units show comparati vely more satisfaction and PSM for whites when they have higher levels of alternative work arrangements, lower levels of diversity management, and higher abstract objectivity. They show more satisfaction and PSM for men when they have more young and new em ployees , and more abstract objectivity. White men are experiencing public service differently than their people of color and women colleagues. In the next chapter, interviews with employees in representative units reveal more about how these differences ar ise.

PAGE 126

114 CHAPTER FIVE INSTITUTIONAL ETHNOG RAPHY OF WHITENESS A ND MASCULINITY IN FEDERAL WORK UNITS Masculinity and Whiteness are not static identities. They are relational and evolve with the extent of social exchange between white men, women, and people of co lor . Inside a room with no communication Masculinity and Whiteness are meaningless. But work in a society requires social exchange, and work for a society is social exchange. This chapter reports the findings of an institutional ethnography exploration of the research question looking at how public organizations, public managers, and public administrators perform Whiteness and Masculinity in federal work units . After quantitative testing, 14 federal workers were identified through a combination of personal and professional networks. The informants were initially identified to maximize the diversity of the type of work performed by their agencies, and the geographic diversity of their work units. The work units focused on IT support, immigration, auditing oth er agencies, environmental management, and labor issues. The 14 informants work ed in eight different work units in the Denver metro area, the Washington metro area, and rural Virginia. The informants were lawyers, IT experts, researchers, engineers, and HR managers. Seven of the informants were white men, five were people of color , and five were women , but revealing intersectional demographics risks revealing their identities. Before conducting interviews, it was confirmed that each work unit was less than 1.6 standard deviations from the mean comparative job satisfaction, management satisfaction, and PSM, and over representative ness of whit es/men/white men in management. The interviews were at various locations identified by the informants as places where

PAGE 127

1 15 t hey felt comfortable. Five interviews were at coffee shops, three were in informants homes , one was in a park, sports team office, and three were in a library. The interviews were reco rded on a digital recorder , with handwritten notes taken on identified copies of the question guide. Additional audio notes were recorded after each interview, and the social media, weblogs, and relevant agency web presence was downloaded and saved to encr ypted files along with the audio recordings. Once all the data were compiled, all audio was transcribed using Dragon Naturally Speaking transcription software with the researcher following along to insure accuracy . All told, the qualitative data adde d up t o more than 150,000 words . After the first pas s , the researcher took notes on overarching themes apparent during the transcription process. On the second pass, the transcripts were imported into QDA Miner Lite and the performances listed in Tables 2.1 2.3 were identified and coded . Single data segments received multiple codes when multiple performances of Whiteness and Masculinity occurred simultaneously, such as when blindness was defended by minimization , displacement, or microaggressions . After the secon d pass, the data was analyzed again for emergent codes and miscoded segments , and interview data was compared to data collected over the internet for triangulation purposes . In the final stages of analysis , the coded data were analyzed for performances ide ntified by the CART findings including the relationship between diversity trainings, alternative work arrangements, and Whiteness , and abstract objectivity and Masculinity . These analyses are the next two subsections below. The end of the chapter report s o n each of the performances identified in Tables 2 .1 2 .3 to identify how Masculinity and Whiteness are maintained and before determining the extent of support for each of the

PAGE 128

116 proposition s. Diversity Management, Color blindness, and Alternative Work Arrangements The CART findings indicate that diversity management was made ineffectual in work units with greater access to alternative work arrangements. The qualitative data supporting the finding that alternative work ar rangements are both necessary for employees who are tasked with caregiving , and contribute to nervousness and color blindness to maintain whiteness by undermining camaraderie . White men express color and gender blindness. The data indicate d that nervousne ss and color blindness are a daily reality in the lives of white men federal workers. White men informants indicated they are generally uncomfortable talking about race and gender. One white Several informants used displacement tactics to avoid talking about race in the work place. Displacement is when one institution, such as gender, is used rhetorically to protect and maintain a second institution, such as race. In one case, when asked about discomfort talking ee in many cases has more to A white man informant displaced the direct question about race and gender to rank . After being asked if he was ever uncomfortable talking about race, he answered,

PAGE 129

117 Not to my knowledge. I mean to me peopl military. He then shifted the conv ersation to frame himself as the victim of discrimination in high taking that he did not feel nervous talking about race in the workplace, his sensemaking process moved through displacement to discrimination based on rank and back to a time when he was accused of racism more than a decade earlier. The level of discomfort while he spoke about race was readily visible in his intense eye contact, and audible as his vocal inflection raised in pitch and tone, and his volume increased. In shor t, he was performing emotion work to align his feelings with his perception of himself as genuinely color blind. H e seemed earnest in his belief that he had never been uncomfortable talking about race. When pressed for an instance when he was comfortable t blindness as the agency for which he worked had been in the news multiple times only one year earlier over accusatio ns of racial discrimination. Perhaps counter intuitively, the data supporting the finding that nervousness is heightened by laws that are meant to create race and gender equality, such as the Equal Employment Opportunity Act (EEOA). Even informants who are knowledgeable and capable of talking about race and gender openly and in ways that get at fundamental problems avoid conversations of race and gender because they do not want to violate laws mandating color and gender blindness. One white man worked to m ake sense of why race was an uncommon

PAGE 130

118 When asked about tra ining, the informant discussed EEO requirements. Another white man reported, In management, government management, we are very careful about things that have to gender, age bias, and things like that when it comes to hiring, and what you can say, what Rules and policies meant to create equality limited the ability of these white men to talk about race. While so me women and people of color mentioned EEO requirements, it was not as thematic in their descriptions of race and gender in the workplace. Among those who do not express nervousness in their conversations about race, the data indicate that color blindness is nevertheless the primary articulation of race in their public service. For example, one white man in a work unit that dealt directly with immigration was able to speak knowledgeably about civil rights history and authored authoritatively on race in immi gration policy. When it came to his work product, race was front and center. Yet, when it I should never mention When I asked why this was the case, he mentioned organizational culture and reiterated how laws add to nervousness . constructively. I guess the first thing that [my supervisor] would think is t He indicated that he was afraid that the emotion work demands on making such a statement would be too hefty, despite his noticing problems and wanting to speak about it. When I asked how it is that a work unit dedicated to immigration issues had never discussed race at work, he

PAGE 131

119 Color b lindness and nervousness were prevalent themes in all interviews, even with people of color who desired to speak about race more than their white colleagues. One woman of color described a time that she wanted to talk more about race but felt unable to do so because Over the He went on to describe an inst ance when he suggested that people of color were not well represented among supervisors in his agency (a fact I verified in my data). He described performing emotion work to predict and negotiate the reactions of white people , saying, over the years I goi Another man of color said that he only brought up race after he performed the emotion work necessary to know say and what somebody else might perceive as either an attack or just something that is at nervousness is not just experienced by white men, but by people of color must perform emotion work to negotiate conversations about race with white colleagues. Women too were nervous to bring up gender in the workplace, to the extent that almost every w oman I interviewed felt they had a legitimate sex discrimination complaint that they did

PAGE 132

120 complaint? Could I have been a pain? Yes. I coul that they felt uncomfortable bringing up their concerns because the emotion work and the risk to their job were too great . Perhaps the most common way that people expressed color blindness and nervousness was t hrough abstract objectivity. Almost every informant responded to the question of whether t, a white woman roblem. No, its, again, because of the way the met the qualifications they were seeking. Although almost every informant acceded that white men had advantages that others did not in the work place, this was the closest that anybody came to asserting that the hiring and promotions process was biased toward white men . This st udy confirms most of the literature in finding that color blindness and gender blindness are the dominant articulations of race and gender in federal government work units . These data also support the finding that people of color and women were less likely to confront racism and sexism because of nervousness and blindness, such as the man of color who suffered

PAGE 133

121 a wall of silence from his co worker, and the women who avoided filing complaints of sex discrimination. Even when white men were able to articulate race and gender in meaningful ways, they avoided the topics because of nervousness that they would make an error that might have emotional, legal , or disciplinary repercussions. Th e relationship between trainings that were meant to represent diversity mana gement but had the effect of increasing nervousness may explain the Chapter Four finding that increased diversity management is related to increased comparative white satisfaction and PSM (Figure 4.1). There was only one set of circumstances that informant s consistently reported overcame this wall of color and gender blindness. Blindness mitigated by camaraderie. This research finds that c olor and gender blindness breaks down when informants develop personal relationships with a person of another race or gender, or when they became close to a person of their own race and gender who expresse s race and gender cognizance. This closeness facilitates easier emotional transactions when discussing race, lowering the amount of emotion work needed to enter convers ation s about race. One man of color described the circumstances under which he openly discusse d race and , and I already have a rapport, and we can The emotion work required of the informant was less This was like another man of color who said that he would have a conversation only after assessing the oth er

PAGE 134

122 White men relied heavily on casual conversations when talking about race , the kinds that are associated with the proverbial water cool er. These conversations required less emotion work than overcoming their own nervousness and the nervousness of their colleagues and managers to speak about race in more formal situations. One white man explained that there was one person with which he dis cusses race and gender regularly. Caucasian. And we talk about it, because, I guess we both have family in [region]. I guess we can just kind of talk about these things. Another white man was able to talk about race and gender by using humor as a buffer. foot two soft six foot when pushed on an instance when race was explicitly discussed in the workplace d escribed attending a baseball game with a colleague of color. The only other time this white man could recall was when a colleague was asked by the Trump Administration to enforce a law that directly and negatively affected the colleague and his family. He reported that this colleague this as a moment of vulnerability between friends, rather than a professional interaction. After describing a time when he discuss ed race, another white man said that he would not have talked The white men were able to overcome the lower emotion work threshold required to tal k about race in informal settings with trusted friends despite their nervousness and color blindness. When white men talk about race with their trusted colleagues, they often express surprise about what they learn. The white man with the colleague who was uncomfortable with an order,

PAGE 135

123 the six foot two black friend who wa s soft spoken appeared surprised by the realization that a large black man would work to avoid replicating stereotypes with his colleagues. Similarly, another white man described being told that one of their supervisors was prejudiced against blacks, It w as just shocking to me how cavalier it was brought to me like it was no big deal. It had met her. That obviously kind of colored my impression of [the supervisor]. Another white man was talking with a friend who was a black man at a conference, when I was like, rency or whatever. I and concerned, and upset that he had never noticed the problem. Camaraderie was not a factor that reached saturation in undermining gender bl indness , p erhaps because men are more likely to have friendships with women generally, including intimate relationships with spouses, daughters, mothers, and sisters. Yet, personal relationships with people of other races are still rare (Lipsitz, 2004). Th ese kinds of friendships lowered the emotional costs for people of color to speak openly about race and brought new insights to the white men I interviewed. T he insights they discovered surprised them and helped them better understand the experiences of th eir colleagues . Alternative work arrangements undermining camaraderie. The study finds that Several informants reported emptier offices with people on compressed schedul es, flexible schedules, or teleworking. Every informant described being affected by telework. Two worked

PAGE 136

124 from home full time, while several others met me on their telework days (during lunch hours) to avoid having to get me through security. Those who did not work from home were still affected by the others who did. One white man who did not work from home was in a unit with colleagues spread throughout the country, so he had friends who worked in the same physical office as he did, but he had no colleagues within 500 miles. The alternative work arrangements are increasing is backed by data from the OPM and from the private sector. OPM (2017) reports that 42 percent of federal workers were eligible to telework in 2016, and that more than half of those employees who were eligible took advantage of the program. This is an increase from 29 percent of eligible employees in 2012, and with the trend going up, it was likely even higher in 2018 when informants participated in the study . This trend is no accident. Telework is widely viewed as saving the government money while increasing worker productivity, decreasing environmental impact, and making employees more satisfied with their working conditions by freeing them from commute times and allowing them to deal with domestic chores during their breaks and lunch hours (OPM, 2017). At least one report from the private sector confirms these findings, while also finding that employees who work from home are more likely to be involved in thei r communities and their (Global Workplace Analytics, 2017) . Gallup (2017) similarly finds many benefits for workers, while qualifying their findings with the warning that remote workers are less engaged, and are less likely to receive fe edback from their colleagues. This study finds that , thereby undermin ing camaraderie. One white man tried to figure out how many people he le tend to telework in the federal government

PAGE 137

125 not socialize with his team because because it reminds me all the time, o h (!) The result of this lack of face to face interaction is a decline in camara derie. One white man complained about bouncing off questions and having a cube mate nex more telework, and that changes the dynamic, just like walking around on a daily basis there are araderie important for breaking down color blindness is being undermined by alternative work arrangements. The finding that alternative work arrangements are undermining camaraderie to reproduce Whiteness in the federal government explains CART findings th at show whites are comparatively more satisfied with their jobs and management and exhibit more PSM than people of color when they have greater access to alternative work arrangements. This leads to an inevitable tension between the needs of caregivers, wh o are more often women and people of color , and the need to dismantle Whiteness and Masculinity in the federal government.

PAGE 138

126 Alternative work arrangements support caregiving. Almost all informants gave the needs of caregiving as a reason that inequality exis ted between men and women. A childless woman of color who presented as white explicitly mentioned this as an example of masculinity in her agency. care of their new One white man explained a choice to sacrifice his caree r to keep his children from having resilient, and all that. I minded moving. I got moved from [State] to [State], and I minded it a eflected in the sentiments of the other white fathers I interviewed. Every one of them met with me immediately before picking up their kids from extracurricular activities. In these data, parents of both genders discussed challenges associated with caregiv ing while working for the federal government. One white man informant described how he advanced stability was important for them whereas it was less important workers made for stability was the opportunity for promotion. The idea that children impeded with career satisfaction was a common theme in the data. One white woman discussed how her gender was less of a setback than hav ing children. When she arrived on the job, she was five months pregnant. I watched my junior colleagues getting assigned completely different work in their first year than I got assigned because I was going to be going out on maternity leave for three mont hs. I got short term projects. I was in large part sort of assigned to babysit a senior

PAGE 139

127 She reports that because she now worked out of her home managing a team of people spread across the c ountry with her mother in the other room taking care of her children. White fathers found it important to have flexible schedules and to be able to work at home. One father said he was in the federal workforce, instead of a private firm because of the flex Overall, telework was more important to parents than child less employees. One childless man of color said that he preferr ed to work in the office for the camaraderie. No parents that I interviewed indicated that they chose not to work from home, even when they said they preferred to work in the office. The tension between alternative work arrangements and childcare needs bec omes more apparent given literature indicating that men are less likely be primary caregivers to children and elderly parents (Doucet, 2018) . The literature also indicates that white people have substantially more wealth to spend on childcare, eldercare, t heir own retirements, and taking care of their grandchildren (Oliver & Shapiro, 2006) , meaning that white people have less need for alternative work schedules to fulfill caregiving roles. This indicates that white men are less dependent on alternative work arrangements, and so telework and compressed schedules are likely to have comparatively lower impacts on satisfaction and PSM for white men. Yet, the quantitative

PAGE 140

128 findings indicate the opposite, at least when it comes to race. White people are comparative ly more satisfied and show higher PSM than people of color when given access to alternative work arrangements. This lends support to the finding that camaraderie is the missing link between alternative work arrangements, color blindness, and increased comp arative white satisfaction and PSM. The qualitative data indicate that the mechanism that cannot be accounted for in the quantitative data is camaraderie, the casual water cooler style conversations that lower the emotional costs for engaging in probing a nd challenging discussions about race. These opportunities arise at baseball games, lunchtime gatherings, on car rides between work locations, When race is discus sed, white men find the insights gained from talking with their friends of color surprising and even shocking. These moments circumvent nervousness by lowering emotion work demands, thereby break ing down color blindness and exposing white men to the lived experien ces of people of color , and thus increasing the job satisfaction, management satisfaction, and PSM of people of color employees. Meanwhile, white men typically have few friendships with people of color (Lipsitz, 2009) , but they have regular close m eaningful relationships with spouses, daughters, mothers, and friends who are women (Bancroft, 2002) . This provides a possible explanation for why alternative work arrangements do not appear to have the same impact on comparative satisfaction and PSM for m en in quantitative or qualitative analysis. Masculinity, (Whiteness,) and Abstract Objectivity CART analysis indicates that abstract objectivity is related to increased comparative satisfaction and PSM for men in the federal workforce. In Chapter Two, the link between

PAGE 141

129 objective ideas of merit and masculinity was well established in the literature, and the CART findings in Chapter Four confirms that literature and indicates that it is having ongoing impacts. Complicating these assessments, Table 4.1 finds th at white men have statistically significantly lower perceptions of objectivity in hiring and promotions, while men of color indicate the highest perceptions of objectivity, with women in between. The qualitative data confirm these findings and provide deep er understanding. The data support the finding that federal employees work diligently to make sense of how to define quality public service. When asked what the government wants in an employee, one of color was exasperated by the inability to understand how you going to bri administration into objective certainties but found the organization lacking in setting expectations. One informant relied on objective facts and numbers to determine merit and use d this sense of objectivity to assure himself that his work unit did not unjustly benefit white men. At the [employee] level everything is driven by your numbers. How much work did you do this past quarter? Did you do the work in the correct order, you kn ow the order it came in? So that we can serve our customers best. And did the quality of work under very very very well as I can envision. This white man informant expressed confidence that hiring and retention was very equal

PAGE 142

130 Most jobs held by informants were not so easily standardized and regimented. Most of the informants, including the white man who reported working for a near meritocracy, valued opportunities for creativity and to serve the public. They described their work as remarkably self directed , and informants described experiencing a great deal of freedom to pursue their interests. One white always made it very clear I love this, this is really fascinating to me. okay? for something that her agency could be doing to pursue the mission, and she was authorized to betwee getting their first interview for a job within two days after going to a job fair. S tudents of bureaucracy and the federal government would likely be surprised by the sheer amount of innovation, autonomy, and self motivation in which these informants were engaged. Little of it could be easily codified, regimented, and standardized. Theref ore, a theme emerged of informants spending a substantial amount of energy and time figuring out whether they were succeeding at these new, innovative, and creative tasks. Because merit is so difficult, if not impossible to pin down in their public service , the informants I interviewed spent a great deal of time and energy attempting to interpret their experiences into calculable determinations of work quality (Stivers, 2018). Some informants

PAGE 143

131 reported that it was their primary job to determine quality. One white man described his job thus, y figuring out how to measure success in their job. Lower level employees were active in working with their supervisors to determine what the goals should be and felt a great deal of satisfaction when clear goals were set and followed up with analysis. One white woman said with energy put in to attempting to understand what merit is and then account for it was a substantial portion of what informa nts talked about during their interviews. Most informants identified ways in which people of certain races or genders were negatively impacted by lack of objectivity in the merit system. One white man reported, It seems like the ones who are targeted with performance issues, they tend to be African the surface. The same white man di scussed a group of coworkers who did not come to a social gathering with everyone else. When the informant asked a colleague why some people were not present, the colleague responded, uld you say

PAGE 144

132 ed with. So this conflicted with Americans i One white man recollects having walked in to a situation where several senior workers were being disciplined as poor performers, and he perceived that they were disproportionately black women. A woman of color who was not bla ck in that same work unit discussed this problem more. Well, I will say in my career, especially having an African American [supervisor], I do feel like there are different standards for African American employees that were not as rigorous as for the rest The white man that mentioned that merit was objective because it was based on numbers, nevertheless perceive d that fur ther up in the agency that objectivity broke He blamed this discrepancy on soft skills. ce will be put in positions on the team person in the room just because I am the best communicator. When pressed about why women were not represented at the top of the age ncy, he Another white ma n struggled with his hidden suspicion that he was being iced out of promotional opportunities because there was a group of women supervisors who had been working together for a long time and mostly promoted others who were almost exclusively women. When as ked if everyone in the unit was treated equally regardless of gender, he answered,

PAGE 145

133 I feel very uncomfortable talking to you and wondering out loud whether I am being treated unfairly as a white male. It has crossed my mind that I am not in the in group bec ause I am male, and I never will be, and because of that it might be better for me to I feel uncomfortable talking to you saying that gender might play a role in that, but it has are after sayin g that, this white woman informant equivocated in her assessment because she was These info rmants understood that their emotional experience of their workplace was not in line with their values as emotional experiences with their perception of themselves , they performed emotion work. Overall, these data reveal a theme that various groups of people feel like merit determinations are not objective, and many of the respondents feel that white men are suffering the most in this false meritocracy, even when mo st express the more socially acceptable view that people of color and women suffer from discrimination. These data support the quantitative findings that white men feel the merit systems are less objective than other groups. Even some women and people of c olor informants perceive the merit systems as disadvantaging white men. Given the amount of effort put in to making sense of how to determine success and merit among these informants, it is no wonder that their perceptions of merit would be colored by the process of sense making, a process injected with race and gender assumptions, implicit biases, stereotypes, and other individual level performances of Masculinity and Whiteness . Despite the Chapter Four findings, the qualitative data indicate that gender i s impacted by abstract objectivity in nearly the same levels as race. Informants discussed subjectivity in merit

PAGE 146

134 systems as both raced and gendered, and the qualitative data support the proposition that both men and whites are comparatively more satisfied and motivated under supervisors that are perceived as more objective in hiring, promoting, and retention. It could be that the data are ill equipped to capture the relationship in either case, or that the relationship between gender and merit is more stati stically significant. It could also be that there is a mechanism canceling out the relationship between abstract objectivity and satisfaction that is racial but not gendered. This puzzle merits further research. Other Performances of Whiteness and Masculin ity Many other performances documented in the literature and recorded in the three tables in Chapter Two were reported by informants. Additionally, some themes that were not in the literature review emerged. Society level performances of Masculinity and Wh iteness. The data collection instruments were not well designed to focus on gathering data at the society level, but some performances reached saturation. While society level noblesse and innocence were mentioned in six interviews, most of those performanc es coincided with organizational and individual level innocence/noblesse and will be discussed in the organizational subsection below. Sacrificial zones, representation, and schooling did not reach saturation in the data. Capitalism, consumerism, wealth an d income were mentioned in several interviews but were not linked to race and gender with consistency. Laws that maintained Whiteness and Masculinity emerged as thematic along with caregiving, and both are discussed above. Stereotype was a theme in six of 14 interviews. Informants mentioned big hats worn by black women, a judge who thought that all Chinese persons were duplicitous, women being expected to do party planning, and men sitting around and talking sports. One man of color said

PAGE 147

135 that he was treated , tereotypes about race and gender impacted the work lives of informants. Ten informants referenced mostly masculine social norms that informed their work place, although less mention was made of white norms. These society level norms were the expected behav ior of all employees. One woman of color noted , n accredited pay Housin g segregation emerged as a theme only in so far as 12 of 14 informants lived in majority white neighborhoods. One white man lived in a majority Latino neighborhood that was rapidly gentrifying. The last informant did not reveal where she lived. Individual level performances of Whiteness and Masculinity. Many individual level performances of Whiteness and Masculinity were reported in the data, including some that were not well documented in the literature. The interviews were not designed to capture intuitiv e assessments and stereotype promise that benefit white men, and only limited evidence of these phenomena emerged. Displacement, and color and gender blindness are discussed in more

PAGE 148

136 detail above. One emergent theme was the surprise that white men experien ced when they talked about race, and it too is discussed above. Technocracy, dress, speech, and mannerisms most consistently emerged as themes at the organizational level, and so will be discussed in the next subsection. The coupled themes of white innocen ce and masculine noblesse emerged from every one of the informants, and overlapped with plausible deniability, an emergent theme not documented in the literature review. The theme of white innocence, and the concomitant ideal that if white people share the ir innocence it will benefit people of color was prevalent in the data. Several informants worked for agencies involved in work that they felt disproportionately helped people of color and were proud of it. These people held up their work as proof of their antiracism. One self described woman of color was particularly aware of this. She was able to pass as white and around and say, you know, are we like little white do gooders? Because the [volunteers] are all Another way in which innocence expressed itself was through plausible deniability. Informants justified racism and sexism by arguing it was unintentional, unimportant, or inevitable. One white man explai predominantly by lawyers were particularly troubled with the fact that the legal community has so many white men. Informants used merit to give their actions plausible deniability. Part of this plausible deniability stemmed from the desire to mitigate short term losses that stood against the long term goal of race and gender equality. One white man des cribed a situation in which he felt accused of racism.

PAGE 149

137 There was a very strong move, rightly so, in the 70s and 80s when these folks were know if they did a great job training them over the years. I think, you know, they may African American candidates that were brought in at the time, which again was probably the right thing to do. He continued, manage ment systems and we started dealing with low performers. Most of them were African employees, though both realized the racial connotations of performance measures that identified mostly blacks as poor performers. Informants used the plausible deniability of their situation to integrate their feelings of being good people with the suspicion that they were participating in a system that was perpetuating discrimination agai nst women and blacks , performing emotion work to maintain their white innocence. Microaggressive behavior was reported by all but two informants, both of whom were ressed that day or something like that, which in my opinion has no place in a for an opinion that he felt was salient to his race. One manager described calling o ut to his administrative assistant across the office , who was a black woman, and feeling like that was hostile. A woman of color working as a professional discussed being asked to make copies for the three other professionals in the room, who were all men.

PAGE 150

138 Another white man described a Americans. This same manager wondered aloud why there was not an Irish History Month to match Black History Month. Some of this behavior was macro aggressive. One white man told about a man who was driving along a road with a woman colleague. He stopped the car in the middle of some rural road. Got out of the car. Physically then drove off and left her in the m iddle of a field. Additionally, three different informants discussed sexual harassment that they did not report , as described above . Another theme that emerged in every interview was minimization. Informants brought up a trivial case of perceived racism an d used that example as evidence that all claims of racism were overblown. One white woman who worked for an agency that has often been in the news for racism described a black co worker who was upset because a municipality they were working ., You can feel that way, but at the same One white man sai d that race and gender discrimination was mostly a thing of the past, anachronism to minimize the impact of discrimination on modern people of color . Others minimize d by narrowing in on a single work unit that was relatively diverse. In the words of one Note that all of the work units were verified by the quantitative data as representative of the f ederal government. Nevertheless, the claim that his work unit was different had the dual effect of both minimizing discrimination and maintaining the innocence of the speaker. As noted above, multiple

PAGE 151

139 informants made sense of discrimination against women b y minimizing disparate impacts as the result of their negotiating style or their having children. In these cases, rather than wondering whether the system was poorly designed for rewarding masculine negotiation techniques, or masculine approaches to childc are, the informants placed the blame on women facing a masculine system. The data support the finding that the perception of a generation gap operates as another performance of minimization. Seven informants discussed their perception that older people in the agency were more likely to exhibit Whiteness and Masculinity than younger people. One white man heard about a supervisor who some felt was discriminating against black workers and ervisor] is very opinionated and very, I would say, kind of old school. Has that very old school mentality, where color reported that her agency was getting more because the older men in the office made inappropriate comments and jokes. One white woman referred to sexis m and racism as a relic of the time when people smoked cigarettes in the office. Another white man reported, I think the newer generations, if they were in the leadership positions when they were hiring us, when they were hiring people, this is my opinion or hope, is that since they were not raised in that [old fashioned] mindset, they will be different. The perception that younger generations are dismantling racism and sexism in federal agencies is another performance of minimization. CART analysis reveal s that units with greater percentages of younger people are associated with higher comparative white masculine PSM, and units with lower percentages of younger employees are associated with lower comparative masculine job satisfaction (Figure 4.1). Also, t he qualitative data and OPM (2016) indicate that

PAGE 152

140 many women and black employees were hired into the federal government in the 1970s and 1980s, yet whites and men are still overrepresented among supervisors decades later , and descriptions of micro aggressiv e behaviors and color /gender blindness continue with frequency. All of these data are confirmed by the literature on critical race and gender theory, which assumes that racism and sexism are durable attributes of race and gender (D. Bell, 1980; hooks, 201 4) . With no evidence showing that younger generations of federal workers will dismantle Masculinity and Whiteness, the ongoing belief that the institutions will fade away with time maintains existing race and gender structures by justifying inaction. Proje ction was a theme that has limited support from the data, as five informants reported white men identifying discrimination against them as a significant problem. As discussed above, white men are more likely than other groups to indicate that merit systems are not objective, and are described the work situation of a colleague that revealed discrimination against men. She recalled talking about a development program, I was tal Pointedly, the white man did not indicate that he applied and was rejected, but instead that he was advised that he would be rejected. One woman of color described projection in her unit, heard any complaints from anybody in, like, a woman being not hired because she was a whatever. But I definitely heard white men make those kinds of comments.

PAGE 153

141 Another white man worked through his feelings that he may be having a hard time because his managers were mostly women, a process of sense making that is discussed in more detail above. Overall, though, projection was a much more limited process than it is in the literature. Most white men informants did not express projection, and those who did expressed a distrust in the impulse to believe they were the subjects of discrimination. One emergent theme that indicates the beginning of the d ismantling of Masculinity and Whiteness was the cognizance of systemic racism and sexism among informants. Awareness of the structural nature of race and gender discrimination was expressed by almost every one of the informants when asked about the root ca uses of discrimination in our society, even when they did not use words like white supremacy and patriarchy. Most informants cited personal behaviors along with structural problems, but they usually started with structural problems. One example is this whi te man, I think there are lots of reasons. I think that, obviously 300 years of slavery and Jim Crow has prevented African Americans from having the same economic and education human the same ways... Only one woman discussed personal attributes to the exclusion of societal and structural discrimination, and she gave the disclaimer that my f structural ways that white men were benefited in society and their agencies. Organizational level performances of Mascu linity and Whiteness. Institutional ethnography focuses primarily on the organization as the unit of analysis, and so it is that organizational level performances were the first to reach saturation and had the most support in the data. Innocence and nobles se were discussed in detail above as individual level performances.

PAGE 154

142 The data collection method created few opportunities to observe differences in mannerisms between white men and other employees, but it did reveal Whiteness and Masculinity in dress and sp eech expectations in work units. Every informant discussed expectations of dress and speech that reinforced Masculinity and Whiteness, and there were 146 my of ar attire. One informant, a man of color, wore a traditional shirt that derived from his ethnicity as part of a cultural awareness month. This small breach of dress norms held significant impact on his experience being people of color in a white work envi ronment. He felt compelled to speak with his boss about wearing the garment in advance, and his boss responded in a joking way that email to have everybody come and wa fortable having breached the that his co workers were welcoming of this small deviation from white dress standards. The most noticeable speech norms came from men who were associated with the military. Their speech was overtly formal, even directly relating to military terms, suddenly

PAGE 155

143 sliding into crude militaristic humor to emphasize certain points. One man of color was not a military veteran, but described his ag ency as being very militaristic. He described his day as eight detailed, as though I was debriefing him. Out of his formal style, the use of profanity took on the feel as an asset for While white speech performances were mostly invisible to informants, there was some passing recognition that the work units were adhering to white and masculine speech. The speech of most informants was not formal. As can be observed in t he quotes selected here, including stall words that are associated with the white suburban mid something positive had happened, as The closest that anyone came to using a slang term that was more closely associated with people of color informal, but there were no terms or uses that were more readily associated with any cultural tradition other than Whiteness, although the interviewer s identity as a white man from the Rocky Mountain West likely impacted this finding.

PAGE 156

144 Related to the dress and speech of work units, informants reported dozens of cultural touchstones that maintained Whiteness and Masculinity. Sports were a constant theme, and most of the sports were of the kind associated with Masculinity. There were golf tournaments, social gatherings to watch football and soccer, and recreational softball games, all of which are reported by the American Time Use Survey as predominantly sports engaged in by men (Woods, 2017). cardiovascular equipment were mentioned by informants (Woods, 2017). A sport that studies show blacks have greater interest in than whites ( R. W. Turner, Perrin, Coyne Beasley, Peterson, & Skinner, 2015) , basketball was only mentioned twice in interviews, both times by men of color. While many work units shared touchstones that incorporated sports, a few mostly technical work units emphasized (Kendall, 2000) . This white masculinity emphasizes comic book characters and technological know how and envisions itself as anti masculine because of the overt rejection of tradition ally masculine sports and machismo. Nevertheless, the literature indicates its strong ce. He reported that his office was not masculine because, what I would consider masculine is more kind of like your kind of more stereotypical the [work unit] is kind of like As with this informant, n erd masculinity was most likely to appear in work units where men felt that they were insufficiently conforming to hegemonic M asculin ity , highlighting the possibility th at the Whiteness of comic book and science fiction culture could be interacting with a perceived deficit in Masculinity. The emergence of nerd masculinity in white dominant Silicon

PAGE 157

145 Valley, and the fact that it is dominated by white men, not just men, conne cts this form of masculinity to institutional Whiteness (Cooper, 2000) . While these data hint at the idea that nerd masculinity is an intersectional performance of Whitenes s protecting Masculinity, further research better focused on nerd masculinity is nee ded to better understand how this intersectionality operates. Celebrations also operated as exclusive touchstones. While celebrations of women and people of color veterans who are m ostly men were considered universal celebrations in which everyone could of women and people of color There were no such limitations on veterans day celebrations, where they celebrated all veterans in a color and gender blind fashion , though more than 70 percent of the veterans in the federal government are men (OPM, 2016). Many of the social gatherings occurred in spaces that were affiliated with Whiteness and Masculinity, including Irish sports bars, upscale foodie restaurants, baseball games, and other places where sports and hamburgers were prevalent. No informants reported going to an R&B club, a rap concert, or a basketball game with their work colleagues. Still other social gatherings were racially segregated, and one white woman and one white man in differ ent work units both noted that black administrative professionals often went out with other black administrative professionals. One man of color noted that he had a regular lunch time group that were mostly men of color, and that they openly spoke about ra ce during their meals. One white man noticed two of his colleagues never attended group social gatherings, and that they both were black.

PAGE 158

146 When he asked a co worker why that might be, she told him that their supervisor was racist. These data support the fin interactions with management and colleagues. There was limited support for the impact of private life taboos on the maintenance of Whiteness and Masculinity. One white man was surprised at how co mfortable people were talking about their personal lives, but most people felt open talking to their boss about their personal needs, and that was further facilitated by the alternative work arrangements discussed in more detail above. One exception was in parenting . While men felt comfortable asking for time last woman discussed above who was told that she had to choose between being a mom and her career. Still, this was a limited theme that did not arise as often as expected , perhaps because of the prevalence of alternative work arrangements. Many technocratic themes appeared in conversations with informants, mostly around merit issues discussed above. However, the amount to which the description of work itself was technocrat ic varied greatly based on the work unit, and only limited support appeared for the idea that technocracy significantly extends beyond promotion, hiring, and retention. As discussed in other parts of this chapter, informants described a workplace that was remarkably innovative and open to change, which created the need to spend time and energy developing so called objective measures of excellence. Entrepreneurialism might better capture the nature of work ongoing in the federal government than technocracy. Adhocracy and bureaucracy worked together to maintain and reinforce Masculinity and

PAGE 159

147 Whiteness among my informants. As discussed above, alternative work arrangements simultaneously undermined Masculinity, and strengthened Whiteness. The adhocratic nature of work units was facilitated by alternative work arrangement s , and in turn alternative work creative as you want, do whatever you want to jus solving required improvised approaches by small teams playing varying roles depending on the out what needs to be The adhocratic nature of the work in most units meant that team members required certain skills that were often associated with educational attainment. Multiple work units struggled to find people that could fill certain roles and found that the candidates they were getting could not do the job without additional training. One white woman who was hiring could not find anybody with minimum qual This supervisor found two people and trained them into the job, and both of those people were black. Another white man noted that most of the people negatively affected by attempts to crack down on poor performers were black, to the extent that he reported

PAGE 160

148 hiring people in the Washington Metro area, where informants report that a substantial portion of the clerical working poo l is black. Adhocracy also became apparent in contracting. Informants in all three locations reported that their contractors were largely black and Latino. Contractors doing building maintenance were often Latino, and security guards were often black along with IT and administrative strative services, like contracting in the mail and all that stuff is African wa s described as both temporary parties, except the contractors who were disproportionately black women. While this qualitative data indicate that contractors may be more often people of color (2017) study of federal contracting makes it clear that there is currently no data on the demographic makeup of federal contract employees. This presents a possible avenue for future research. While work units often worked in highly adhocratic ways, overall agency structures were still strongly vertical and bureaucratic. For example, wh ile one informant described his workday nd to take a hierarchy of military org chart perspective, per Gender and race segregation in job assignments is reported as closely aligned with a dichotomy between experts and non experts, as is predicted by the literature (Burris, 1996) . One

PAGE 161

149 white w informant in another work unit reported, I see the fact that we have African American [professionals] at all? I mean, sure we do. If I sat here and went through each office. Yes, I c ould identify African American and Hispanic [professionals]. One white man talked with his colleague about whether they could think of any cial segregation was present at work units in all three geographical areas. There was also job segregation based on gender, but it was not as clear cut as race. One white man reported that his agency was almost entirely men when he entered it two decades e in recent years, the trend he was experiencing surprise. A white woma n attempted to make sense of her supervisors being predominantly women, go out to lunch with the support staff and do th ings with them. And that seems to be a race thing.

PAGE 162

150 clerical and administrative positions. One white man described the support staff in his work unit, aged African These dichotomies between experts who were disproportionately white men, and non experts who were dispropor tionately women and people of color was intertwined with power shifting that placed most of the authority and freedom in the offices of professionals that were treme ndous amount of authority. Some of the support staff were also given a good amount of she meant by sophistication, she gave an example of a white woman support staffer who was given autonomy , and a black woman who was not. A woman of color noted that a , s o , it would disproportionately a The white man from above in are limiting when it comes to prom These data indicate that informants recognize a trend where whites and men more often take jobs that are considered more expert, jobs that have more promotional opportunities and situate white men to have more power and influence over their work units.

PAGE 163

151 Support for Propositions The data mostly support the propositions laid out in Chapter Three (Tabl e 5.1). Individual Proposition #1 is that public servants perform Whiteness and Masculinity in their daily tasks. Throughout the data, informants indicated that they regularly performed color and gender blindness through multiple forms of minimization, re ferring to white masculine dress, speech, a nd touchstones, all while referencing technocratic ideals of excellence. Likewise Individual Proposition #2 was strongly supported. Informants indicated that they communicated race and gender through Whiteness and Masculinity performances such as discriminatory framing, displacement, projection, innocence, and noblesse, and reported various microa ggressions in the workplace. At the organizational level, informants reported adhocracy and bureaucracy working together to create job segregation around the expert/non expert dichotomy to create power shifting in whites and men, thereby giving strong sup port for Organizational Proposition #1. Similarly, managers and employees reported organizational processes and systems that exhibited Whiteness and Masculinity dress, speech, and touchstones, as well as nervousness to discuss race and gender leading to st atements of plausible deniability to maintain organizational innocence and noble s s e . These performances confirmed Organizational Proposition #2 that managers implement Masculinity and Whiteness performances as operational processes and systems. Organizatio nal Proposition #3 that public organizational culture reflects Whiteness and Masculinity performances is also strongly supported by the data indicating an overwhelming norm of touchstones, dress, speech, and color and gender k units. There was moderate support for Organizational Proposition #4 that lower levels of

PAGE 164

152 Table 5.1 Level of support of propositions in institutional ethnography based on qualitative and quantitative data. Proposition Level of Support Individual Proposition #1 include pe rformances of Whiteness and Masculinity . Strong Support Individual Proposition #2 Public servants make sense of race and gender through performances of Whiteness and Masculinity. Strong Support Organizational Proposition #1 Public organizational struct ures allocate power and resources to whites and men . Strong Support Organizational Proposition #2 Public managers implement Whiteness and Masculinity performances as operational processes and system s . Strong Support Organizational Proposition #3 Public organizational cultures reflect Whiteness and Masculinity performances . Strong Support Organizational Proposition #4 Public organizations with relatively fewer performances of Whiteness and Masculinity will be more welcoming to people of color and women . Moderate Support Societal Proposition #1 Public organizations and public servants refer to society level Whiteness and Masculinity to resolve conflict . Weak Support Societal Proposition #2 Public organizations and public servants refer to society lev el Whiteness and Masculinity to maintain their status with other organizations, public servants, and individuals in the larger society . Little to No Support

PAGE 165

153 Whiteness and Masculinity as exhibited through institutional performances created work units that were more welcoming to women and people of color . Women informants indicated that they left strongly masculine work units because they felt it was hurting their chances to get ahead, and people of color indicated that they felt more comfortable at work when their co workers communicated acceptance of cultural dress and o pinions informed by their race. The link between lower levels of Masculinity and more comfort for women informants was somewhat stronger than for people of color . Women indicated that they had left work units to escape masculine performances. No person ind icated that they had changed jobs or transferred work units because of performances of Whiteness or indicated that they had even considered it . Overall, the support for Organizational Proposition #4 was less compelling than for the other organizational pro positions. As stated above, the data collection process was not well calibrated to capture society level performances, and as a result the societal level propositions were the least supported. There was weak support for Societal Proposition #1 that public servants and organizations refer to society level Whiteness and Masculinity to resolve conflict. Perhaps the strongest support for this proposition derives from the conflict that arises from public servants attempts to make their jobs calculable . In respo nse to the efforts of workers to make sense of the quality of their incalculable work, they turn to abstract objectivity and capitalistic measurements of efficiency that the findings indicate benefit men and white people. Another conflict that informants d iscussed was the tension created by accusations of discrimination, whether it be sexual harassment or discrimination against groups of workers who were disproportionately African American and women. Here too, the informants responded to the conflict by ref erring to abstract objectivity, plausible deniability, or nervousness . Overall, however, the data instrument did not

PAGE 166

154 capture conflicts unrelated to race and gender, which would have presented a better opportunity to test the proposition. Informants did no t speak much about how they related to people outside of their work units, so there was little data to determine the extent of support for Societal Proposition #2. One white woman informant did speak about her use of federal acronyms with people who were n ot in the federal workforce, but the line of inquiry did not provide any insight into how she built or maintained her status with other organizations. One other white woman informant discussed using her profession as a bridge between her own work unit and other agencies with which she worked but again there was little insight into how she did that work. This proposition deserves more focused study. Conclusion The work units analyzed in this research exhibited Whiteness and Masculinity at every level in many of the ways predicted by the literature. White men experienced differences in satisfaction, in human relations, and in relationships between colleagues. White men worked to maintain their perception of themselves as competent using abstract objective meri t calculations that advantaged themselves more than people of color and women, while at the same time struggling to engage in basic conversations about race and gender inequality and discrimination. In the next and final chapter, the work performed to repr oduce Masculinity and Whiteness is mapped and discussed before the implications of this research are laid out for public managers and public administration researchers.

PAGE 167

155 CHAPTER SIX MANAGING WHITENESS A ND MASCULINITY IN PU BLIC ORGANIZATIONS White men inform ants interviewed for this study reported craving authentic engagement with race and gender equality efforts but were stymied about how to start. One white man expressed a desire to do more, but a lack of knowledge about how to do it. I don't want someone h aving anything to say against me that could in any way be construed as a comment that could en gender an EEOC complaint against me. And I say hat as a white male, and if there's a better way that someone would guide me at the office as to how to talk about these issues in a more constructive way rather than avoiding them, I'm open to hearing that , but I haven't heard it yet. White men felt that they could not contribute to solutions for race and gender discrimination . a lot about this just because I knew that I was going to ri ghts activist in the NFL and believed that the treatment of immigrants to America was one of the most important issues of the era. Despite caring deeply about race and gender discrimination , even dedicating their lives to government agencies working on equ ality, these white men felt like their participation in race and gender justice was uninformed and inauthentic. This chapter describes tools and strategies for white men and others desiring authentic engage ment in social equity and social justice efforts in the workplace. The first section uses the federal work units. The next section identifies key locations on that map where Masculinity and Whiteness can be ma naged. Some of these locations can be acted upon by entry level employees, some are the purview of public managers, while some actions must be undertaken by political bodies overseeing federal agencies. The research implications of these findings are discu ssed in

PAGE 168

156 greater detail in the final section of this study. Mapping Whiteness and Masculinity in Federal Agencies Institution mapping provides a simplified visualization tool to track institutional work as it moves through organizations (S. Turner, 2006) . F igure 6.1 presents a map of the work performed by the work units analyzed in this research. White boxes represent Whiteness and/or Masculinity performances, black boxes represent performances that dismantle some aspect of Whiteness and Masculinity, and gre y boxes indicate performances that both maintain and dismantle Masculinity and Whiteness. The thick solid arrows indicate linkages between two performances of Masculinity and Whiteness and the direction of work maintaining the linkage. A dashed thin line i ndicates linkages dismantling Masculinity and Whiteness. Complexes are combinations of linkages and performances that work together to strengthen institutional stability. The map is incomplete, representing only the empirical findings of this one study. St arting in the upper left corner of the map, white men supervisors establish norms, including exclusive touchstones that benefit whites and men who more easily engage in the work place. These same supervisors maintain expectations of dress and speech that a re more easily met by men and whites. To the right of the norms box in Figure 6.1, diversity management undermines some of the norms that support Masculinity and Whiteness by making the assumptions behind those norms explicit. Simultaneously, the findings indicate that diversity management promotes open conversation and dialogue about race and gender, thereby increasing color and gender cognizance. Diversity management also requires making accommodations for caregiving that help women and people of color t o more fully participate in the work place.

PAGE 169

157

PAGE 170

158 The invisibility of Masculinity and Whiteness norms supports masculine noblesse and white innocence as men and whites participate more effortlessly in the work place leading to the assumption that they have skills and abilities that people of color and women lack. When they are confronted with color and gender cognizance that reveals these advantages as linked to their race and gender, they respond with nervousness and/or microaggressions. The data also indi cate that most informants have some knowledge that gender and race provide systemic benefits to men and whites, and this poses a challenge to innocence and noblesse that increas es nervousness surrounding conversations about race and gender. Similarly, my f indings indicate that laws meant to create equality contribute to this nervousness, because not only are men and whites nervous that their innocence and noblesse is challenged, but they also may be running afoul of work policies and the law. White men, whi tes, and men develop friendships with people of color and women, and this camaraderie dismantles some of the nervousness and helps white men achieve surprising revelations about the race and gender relationships surrounding their everyday lives. They laugh , talk, and determine to do better. This camaraderie is undermined by alternative work arrangements that increase the emotion work required to discuss race , even as those same work arrangements facilitate women and people of color to more fully participate in the work place. Nervousness, compounded by alternative work arrangements and equality laws, overwhelms camaraderie, leaving race and gender articulated primarily through color and gender blindness. Another reaction to challenges to white innocence and masculine noblesse is the microaggression (DiAngelo, 2011; Sue, 2010) . I find three common microaggressions in my data. Stereotyping assigns whites and men according to common societal images. Federal workers resort to projection when they respond to an o bservation of racism or sexism by

PAGE 171

159 accusing people of color and women of the very performances in which they are accused of engaging . Displacement protects one institution, either Masculinity or Whiteness by turning focus to another institution as a bigger wrongdoer, such as disability discrimination, or the bureaucratic hierarchy. My findings indicate that these macroaggressions have minimal direct impact on the rest of the system, but the presence of so many microaggressive performances reveals a conflict between espoused organizational innocence and noblesse, and a reality that is encumbered by assumptions of white and masculine superiority. While microaggressions often rely on espoused blindness, true blindness is better maintained through minimization, o f which I find three distinct flavors. Plausible deniability argues for inaction by focusing on ways that people in the work unit are not responsible for inequity or focusing on the colossal effort required to dismantle Masculinity and Whiteness, an effort possible. The generation gap myth takes anachronism into the 21st cent ury by justifying inaction through the argument that the arrival of a younger generation will inevitably change things. Blindness funnels into the human relations process through a constant struggle to make public service work calculable, effort that resul ts in the myth of abstract objectivity in hiring, promotions, and retention. This abstract objectivity is the primary driver behind a complex of job placement that is segregated by race and gender, a dichotomy between expert and non expert jobs, and a shif t in power toward expert jobs, those that are more often occupied by whites and men. In this complex, equality laws have limited impact dismantling Masculinity and Whiteness by providing guard rails to limit the amount to which job segregation can overlap with race and gender. Through power shifting, men and whites come to be overrepresented in adhocratic work

PAGE 172

160 units often overseeing more rigidly bureaucratic work units occupied disproportionately by women and people of color , and contractors who are perceiv ed as more often people of color and women. In this complex of performances, white men find themselves more often in positions of autonomy where they are encouraged to use their creativity and ultimately placed in authority over a work force that is less w hite and masculine than they are. Thus, the generation of workers that some informants believe will begin the process of dismantling Masculinity and Whiteness are reformed to reflect the institutions already occupying the system. The old performances may i ncorporate a new twist, like alternative work arrangements or the generation gap myth, but they ultimately recreate the same social hierarchy, a hierarchy with men and whites at the apex. These men and whites in management then establish the norms for a ne w generation of public servants, thus beginning the cycle anew. Using the Map to Diagnose and Manage Whiteness and Masculinity This simplified map (Figure 6.1) provides a tool for diagnoses and prescriptions to manage Masculinity and Whiteness in federal work units. Referring to the map, Masculinity and Whiteness management can 1) interrupt linkages between performances, 2) disrupt identified performances, and 3) radically transform well entrenched complexes of performances and linkages. While interrupting a linkage between performances is a relatively minor shift requiring few resources and buy in, it is only a temporary fix that minimally impacts the institutional arrangements that maintain white supremacy and patriarchy. At the other extreme, radical tra nsformation of complexes of Masculinity and Whiteness is likely only possible with a critical mass of support and an understanding of invisible institutions. This section presents strategies for public servants who wish to begin managing Masculinity and Wh iteness in their work units.

PAGE 173

161 Managing Linkages The benefit of managing linkages without attempting to change performances is low er demand for resources. Any member of the work unit can help manage a linkage, from the entry level employee to the President ial appointee. Managing linkages strategically provides an immediate and low cost opportunity to begin the project of managing Masculinity and Whiteness in work units. One place where managing linkages can be effective is in developing camaraderie. This re search finds that alternative work arrangements both assists women and people of color to more fully participate in the work place and increases nervousness that strengthens Whiteness. As a performance, nervousness is challenging to change because any atte mpt at confronting it is generally received with more nervousness (Gooden, 2014) . However, the data indicate that camaraderie is one way that nervousness is dispersed as individuals use greater levels of color and gender cognizance with their close collea gues. Two recent articles argue that this gender and color cognizance is a necessary first step to undermining Masculinity and Whiteness in public service (Bishu, Guy, & Heckler, In Press ; Starke, Heckler, & Mackey, 2018) . The findings indicate that not j ust inter racial camaraderie, but all camaraderie helps with race cognizance as people more comfortably discuss issues they feel nervous talking about with other colleagues. Alternative work arrangements undermine the camaraderie required to develop race c ognizance by staggering schedules and facilitating fewer informal meetings between colleagues. The linkages surrounding cognizance, camaraderie, nervousness, and alternative work arrangements are mostly the kind that help to manage the influence of Whiten ess in the work place, except for the arrow that undermines camaraderie and limits the latter effect on nervousness (Table 6.1) . Managing this one Whiteness linkage could be as simple as an entry -

PAGE 174

162 level employee inviting his colleagues to his home for dinne r or developing a routine of taking walks around the block for coffee with peers. White men and others seeking authentic engagement in social justice can use these informal opportunities to bring up media, news items, or personal stories involving race and gender. Meanwhile, managers can facilitate informal camaraderie through encouraging their subordinates to take lunches together by offering to cover the phones, scheduling work meetings at times that facilitate a gathering afterward, or simply giving empl oyees permission. As the dichotomy between work and home becomes less meaningful with telework and alternative work schedules, cultivating workplace camaraderie must adjust. An entry level employee seeking to authentically engage in social justice can use technology to cultivate camaraderie by establishing a private social media group on platforms such as WhatsApp or Signal. To facilitate the kind of comfort that is necessary to overcome nervousness means using these encrypted platforms in ways that might conflict with the interests of management, such as by complaining about unfair treatment in the workplace or identifying ways the office is benefiting white men. Managers can support this effort by encouraging such private groups, even though it means that communications between employees will be outside of their perception. Because of laws like the Hatch Act that increase nervousness on work platforms, these social media groups should be privately operated by the employees outside the control of supervisor s. Any attempt at supervisorial control can recreate the same nervousness that these platforms are positioned to minimize. The research indicate s that even occasional gatherings, like annual conferences and baseball games can have the intended effect of u ndermining nervousness. As funding is withdrawn from public service, it is often these kinds of social opportunities that are the first to

PAGE 175

163 be cut. A supervisor looking to better manage Masculinity and Whiteness will understand that the importance of a conf erence is not just training in the latest technologies, law, and techniques, but also opportunities to explore ideas about race and gender. This manager will be more reluctant to eliminate such expenses, which are often comparatively small in federal agenc ies. But when forced to cut these funds, managers can still cultivate camaraderie through supporting an in office retreat where employees take it upon themselves to research and share the latest influences in their field and take a couple of meals in betwe en where they are given informal opportunities to cultivate camaraderie. Once camaraderie is better developed, managers and diversity trainers can make use of that camaraderie to facilitate more open and honest dialogues. One way that diversity trainers m ay be undermining their own efforts could be in facilitating so people are encouraged to avoid making others feel uncomfortable. Literature indicates that t hese safe spaces increase nervousness and allow blindness to reinforce it self (Starke, Heckler, & Mackey, 2018). Additionally, this study finds that trainings given by equal opportunity officers are sources of nervousness, as the discomfort already present in conversations of race and gender is heightened and justified by the f ear of legal accountability. Nevertheless, the law makes it clear that openly speaking about race and gender in the workplace does not increase legal liability (EEOA, 1974) . Diversity management efforts should emphasize brave spaces over safe spaces where discomfort is an important part of learning to engage in authentic efforts for social justice (Arao & Clemens, 2013) . Additionally, these equal opportunity officers could better manage Masculinity and Whiteness in their organizations by instructing public servants on affirmative ways of openly conversing about diversity in the work place, thereby using legal training as an opportunity to undermine nervousness. This same tactic undermines the

PAGE 176

164 nervousness of women facing gender discrimination, so women are en couraged to bring up sex discrimination problems through informal and formal channels. Strategically weakening some linkages while strengthening others increases color and gender cognizance to better manage Whiteness and Masculinity. Managing Performances Managing Masculinity and Whiteness performances is more challenging than managing the linkages between them. Much ink has been spilled discussing the difficulty of making meaningful changes in organization s (e.g Argyris, 1999) . The effort and authority n eeded to manage performances can be intimidating, but such work can start small. The literature and this converge on the inevitable point that performances are changed through white and masculine identity development in individual managers . In Figure 6.1, the two focal points for generating awareness of Whiteness and Masculinity are at workplace norms and blindness. Gender and racial identity development in white men. Blindness and norms that benefit whites and men are critical points in th e institutions of Masculinity and Whiteness. They are both lynchpins without which the institutions suffer. The two performances are related. As Argyris (1999) points out, invisibility maintains organizational norms. Color and gender blindness are part of the complex maintaining norms that benefit men and whites. Dismantling one performance is likely to impact the other. In both instances, the key to changing the performance is for white men and others to engage in racial and gender identity development as individuals and use that awareness to identify the invisible assumptions of Masculinity and Whiteness. Public managers who wish to authentically engage in race and gender equality issues must first engage with their own race and gender. Scholars have stu died white and masculine

PAGE 177

165 identity development, including Helms (1990) and Doucet (2018) . These scholars find that whites and men work to make sense of their identities when they no longer can assume that their identities are the default, such as when white s are in rooms where they are the only white person, or men undertake the job of mothering. Relatedly, Starke, et al. (2018) and Bishu et al. ( In Press ) argue that awareness of Whiteness and Masculinity can spark self understanding among public servants th at benefit equality efforts and effectiveness of public service delivery. Through this process of self discovery, white men and others sensitize themselves to the influences of Masculinity and Whiteness. Public servants who are thus sensitized can point ou t white masculine assumptions in the work place, thereby sharing their awareness with their colleagues. Public managers can then modify workplace norms and decision making processes and begin to dismantle Whiteness and Masculinity. As they move through th e development of their racial and gender identities, public managers can use their learned sensitivity to identify and change norms that benefit whites and men. This study finds that diversity management is one way to help develop sensitivity to these norm s. The linkage between norms and innocence/noblesse is one of the keystones of the Masculinity and Whiteness system, where the sense of purpose of the work unit is created and defended, even as whites and men are more freely able to engage in their work pl ace. The findings of this study indicate that touchstones like sports and comic books reproduce Masculinity and Whiteness, as do dress and speech expectations. As white men and others develop their racial and gender identities, they become aware of the ef fects of these norm and have the opportunity to present alternatives. Instead of going to a baseball game at Nationals stadium, the next outing might be to a hip hop dance concert in Northeast D.C. ( www.danceplace.org ), as dance was one of the leading exer cise activities reported by women in

PAGE 178

166 ( https://mystics.wnba.com/ ), which is a sport with far more women and black spectators (R. W. Turner et al., 2015) . Any employee in a work unit that is invested in nerd masculinity, as were three of the units in this study, could create opportunities to discuss the implications of Black Panther and Wonder Woman on race and gender in their agency, thereby questioning the innocence and nob lesse assumptions that drive much of public service. Racial and gender identity development will also sensitize public servants to gender and color blindness, thereby limiting the effects of minimization and abstract objectivity. Minimization justifies i naction by arguing that the problems created by Masculinity and Whiteness are inconsiderable, anachronistic, or inevitably diminishing (Bonilla Silva, 2003c) . I n this study, research indicate s that minimization is a common performance of Masculinity and Wh iteness in federal work units. However, as race and gender cognizance grows, the erroneous bases of minimization become more apparent. As white men become more aware of their advantages, they can speak out with confidence about the ongoing existence of ra cial and gender discrimination, and support women and people of color who choose to do so. Thus color and gender cognizance created by race and gender identity development effects minimization and abstract objectivity. Abstract objectivity is seen as an antidote to racism and sexism, a way of cutting through Masculinity and Whiteness to hire, promote, and retain the best employees. Unfortunately, the data in this study support a large amount of literature in finding that calculable objectivity is impossib le for many public service jobs. Merit is not, in fact, objective, but a normative concept. Merit determinations are attempts to determine what the work unit believes should be valued, and workers in my data are as eager as organizations to make that deter mination so that it can

PAGE 179

167 guide their work. Because of the ubiquity of Masculinity and Whiteness, all merit determinations will not only be incalculable, but raced and gendered white and masculine. Race and gender identity development in public managers will not lead to management that is color and gender blind, but instead an arrangement of hiring, promotions, and retention that is color and gender cognizant. Such a public manager will not hire a black woman candidate solely because of her identity but wil l critically interrogate how the experience and education of a black woman candidate translates into the white masculine expectations of the federal government. From this perspective, Whiteness and Masculinity give white men a hiring preference that likely decreases the quality the federal workforce. This hiring preference must be identified and managed. Managing Complexes In Figure 6.1, the data reveal complexes of performances and linkages that work together. One complex is that surrounding blindness, an d includes nervousness, minimization, and abstract objectivity. Another complex appears where adhocracy and bureaucracy link through power shifting to the expert dichotomy and job segregation. This complex of performances and linkages appears very stable i n my data, with many informants noting the existence of each of these performances, and few if any suggestions about how to make it less discriminatory. In general, the data indicate that people believe in these structures at the core of how the government is run. Changing abstract objectivity as a performance can modify the belief on which the entire system stands, but it is only one piece of the complex. Literature on evolutionary institutions confirms scholarship on Masculinity and Whiteness indicating that these structures quickly evolve to protect core functions (Ferguson, 1984; Fürstenberg, 2016; Ray, In Press) . Changing any one performance in such a complex will fail to significantly dismantle the system. Instead, it will result in new institutions o f racial and gender hierarchy with better defenses. Just as civil

PAGE 180

168 rights era equal opportunity laws have become a slippery piece of the color and gender blindness upholding Whiteness and Masculinity today, modifications to individual performances are like ly to result in more resilient iterations of white supremacy and patriarchy in the future. Interrupting linkages and modifying performances are necessary for social justice and social equity, but only radical change is sufficient to meaningfully begin the process of dismantling Whiteness and Masculinity. Radical change to work structures and hierarchies. Feminist and critical race (1972) (1979) bureaucracy and adhocracy that currently dominate the s tructure of the work units in this research . These data, and critical race and gender literature (Acker, 2006; Ray, In Press) indicate that these systems are being used in conjunction with job segregation, the expert/non expert dichotomy, and power shiftin g to maintain Masculinity and Whiteness in public service organizations. Ferguson (1984) argues that highly democratic flat structures can counter masculine influences adopted from the (1924) power in their employees and support formal and informal employee associations that bring perspectives from varied angles into the co nversation. Similarly, Schmidt (1993) finds that hierarchical knowledge structures can lead to catastrophic failure, and argues that the job of a manager is to incorporate perspectives from every level of the organization, not just designated experts. Thes e feminist perspectives of management are drastically different from either vertical bureaucracy, or the small team adhocracy model present in the data. To implement these kinds of systems would require a radical overhaul of the workplace structures Simil ar themes of power sharing surface in the literature on indigenous American

PAGE 181

169 leadership. For example, Gambrell (2016) expanding self interest until the interests of the collective are reflected in the lead ers own interests . One example of indigenous leadership includes councils of elders from diverse backgrounds (Harvard Project, 2008) . Contrary to the white organizational structures that were foisted upon them by federal agents and entrepreneurialism (Bane rjee & Tedmanson, 2010; Otis, 1973) , these councils were not dominated by any one profession, but a conglomerate of leaders with decades of experience in many different areas. The result is an efficient system of organizational management that more seamles sly incorporates information derived from multiple knowledge bases (Carlson, 1981 ; Harvard Project, 2008 ) . More research on pre Columbian indigenous governance structures could provide more effective and equitable models for public managers (Ronquillo, 201 1) . The incorporation of radically new forms of management and organizational structure require far more resources than disruptions to individual performances of Masculinity and Whiteness . Nevertheless, such radical change is necessary to secure substanti al movements toward social equity and social justice. Complexes of Masculinity and Whiteness performances and linkages are self reproducing and stable in the system, and only radical change will significantly undermine them. As Masculinity and Whiteness ha ve adapted to ongoing equality movements, so too have feminist and civil rights organizations developed structures that resist white supremacy and patriarchy (Banerjee & Tedmanson, 2010; Ferguson, 1984) . These organizational structures and leadership philo sophies serve as models for white men and others to adopt as replacements to the structures currently in place that were developed under the influence of centuries of congresses, presidents, and civil servants who were overwhelmingly white men.

PAGE 182

170 Implicati ons of Institutional Masculinity and Whiteness for Research The theory and findings in this study contribute to an ongoing conversation about racial and gender justice in American public administration that includes thinkers like Gooden (2014) , Stivers (2 000) , Blessett (2018) , Guy (1993) and many others who are listed in the references (as well as many who are not). This research contributes to the growing body of evidence supporting their work. This study is intended to span institutionalism, representati ve bureaucracy, social equity, feminist, critical race theory, and human resources management, all of which had profound effects on the study. The next steps include expanding the context of research to other settings. Qualitative research can examine how Masculinity and Whiteness performances operate differently in local governments, state governments, nonprofits, and social enterprises. Findings indicating that military units show relatively more equal satisfaction and PSM rates between whites and people of color , but not men and women have yet to be explored. Theoretical development around the culture, history, and structures in the military may reveal that early efforts at racial integration in military units worked, or that racial discrimination is alm ost entirely displaced by rigid gender hierarchies, or some other explanation. Additional research could explore intuitive assessments more directly by using the implicit assessment test (Banaji & Greenwald, 2013) , and experiments have the potential to rev eal much about how public service collects and spends moral license and develops stereotype threats and promises (Cascio & Plant, 2015; Steele, 2010) . Each performance documented here can anchor a small research agenda. This study used two methods that ar e relatively new to public administration research. Institutional ethnography and decision trees are both well suited to explore the multitudinous influences on public administration and management. This research shows that institutional ethnography and in stitutional maps have the capacity to provide understanding of public

PAGE 183

171 administration phenomena. Decision trees also are well suited to explore quantitative data for correlations that can be more closely analyzed through qualitative methods. The combination of the two methods provides a methodological model for future research. While institutionalism is hardly new, this study is part of a modern trend of thinkers that are narrowing in on race and gender institutions, as well as other related institutions (e .g. Portillo & Humphrey, 2018; Ray, In Press) . Replicating the theoretical development process and the empirical methods tested in this dissertation can provide insight into other institutional concepts and their intersections. S imilar work could gain insi ght on the intersection between blackness and femininity, or Latino ness and upper class institutional influences. Research into the conceptual institutions that maintain nonprofits and social enterprises could provide insight into the publicness puzzle po sited by thinkers like Bozeman and Bretschneider (1994) , and Moulton (2009) . Overall, this study suggests on ongoing research agenda that is rich with possibilities. Conclusion: Discovering and Pulling Social Justice Levers in Public Service Organizations Addams (1902) remind ours that for which many men and women of every generation have hungered and thirsted, and without which and thirst. White men and others seek to authentically engage in that which gives so many lives meaning, a search for social justice and social equity that is in a crescendo in newspapers, movies, television shows, and on streaming music services. Yet my data indicate that white men are among man y who do not feel like they know how to join this growing conversation. Because whites and men are at the apex of their respective racial and gender hierarchies, without their

PAGE 184

172 participation social equity and justice are likely to remain elusive. The resear ch here hopes to give public servants more levers to pull in their work for social justice. This study presents an institutional theory of Whiteness and Masculinity that situates men and whites, not simply as inactive participants in race and gender, but as central figures maintaining institutions that benefit us. We have disproportional power over the system, and many of us feel a moral obligation to act. Yet, the same invisibility that guards Masculinity and Whiteness from the observation of researchers, veils our race and gender from ourselves. By tracing performances of Whiteness and Masculinity in federal work units, this study works to unveil the race and gender of those who still hold the most power in many public organizations. Using the knowledge p rovided here, white men who so desire can abate some of our hunger and thirst for social justice , while women and people of color can begin to trust our wish to do better.

PAGE 185

173 REFERENCES Acker, J. (2006). Inequality regimes: gender, class, and race in organiz ations. Gender & Society , 20 (4), 441 464. Adams, G. B., & Balfour, D. L. (2004). Unmasking administrative evil . New York: M.E. Sharpe. Addams, J. (1902). Democracy and social ethics . New York: MacMillan. Addams, J. (1911). Twenty years at Hull House . New Y ork: MacMillan. Alexander, J., & Stivers, C. (2010). An ethic of race for public administration. Administrative Theory & Praxis , 32 (4), 578 597. Ali, S. R., Yamada, T., & Mahmood, A. (2015). Relationships of the practice of hijab, workplace discrimination, social class, job stress, and job satisfaction among Muslim American women. Journal of Employment Counseling , 52 (4), 146 157. Allen, R. L. (2008). What about poor white people? In W. T. Ayers, T. Quinn, & D. Stovall (Eds.), The handbook of social justice in education (pp. 209 230). New York: Routledge. Andersen, C. (2016). White rage: The unspoken truth of our racial divide . New York: Bloomsbury. Arao, B., & Clemens, K. (2013). From safe to brave spaces: A new way to frame dialogue around diversity and soc ial justice. In L. M. Landreman (Ed.), The art of effective facilitation: Reflections from social justice educators (pp. 135 150). Sterling, VA: Stylus Publishing. Argyris, C. (1999). On Organizational Learning . New York: Wiley. Arminen, I. (2008). Scienti F rom incompatible paradigms to ethnomethodological sociology. Philosophy of the Social Sciences , 38 (2), 167 191. Ashkenas, J., Park, H., & Pearce, A. (2017, August 24). Even with affirmative action, blacks and Hispanics are more underrepresented at top colleges than 35 years ago. New York Times . Ayers, W. (2004). Teaching the personal and the political: essays on hope and justice. In The teaching for social justice series Bachrach, P., & Baratz, M. S. (1962). Two faces of power. The American Political Science Review , 56 (4), 947 952. Baldwin, J. (1962). The fire next time . New York: Dell. Black on white: Bla ck writers on what it means to be white (pp. 177 180). New York: Schocken Books.

PAGE 186

174 Banaji, M. R., & Greenwald, A. G. (2013). Blindspot: Hidden biases of good people. New York: Delacorte Press. Bancroft, L. (2002). Why does he do that? Inside the minds of ang ry and controlling men . New York: Berkley Books. Banerjee, S., & Tedmanson, D. (2010). Grass burning under our feet: I ndigenous enterprise development in a political economy of whiteness. Management Learning , 41 (2), 147 165. Bar On, B. A. (1993). Marginali ty and epistemic privilege. In L. Alcoff & E. Potter (Eds.), Feminist epistemologies (p. 199). New York: Routledge. Bartky, S. L. (1990). Femininity and domination: Studies in the phenomenology of oppression . New York: Routledge. Bartley, T., Koos, S., Sam el, H., Setrini, G., & Summers, N. (2015). Looking behind the label: Global industries and the conscientious consumer . Bloomington, IN: Indiana University Press. Bell, D. (1980). Brown v. Board of Education and the interest convergence dilemma. Harvard Law Review , 93 , 518 533. Bell, L. A. (2013). Theoretical foundations. In M. Adams, W. J. Blumenfeld, C. (Rosie) Castañeda, H. W. Hackman, M. L. Peters, & X. Zúñiga (Eds.), Readings for diversity and social justice: An anthology on racism, antisemitism, sexism , heterosexism, ableism, and classism (pp. 21 26). New York: Routledge. Beneito Montagut, R. (2011). Ethnography goes online: towards a user cent e red methodology to resea r ch interpersonal communication on the internet. Qualitative Research , 11 (6), 716 735. Berman, J. (2015). Soon, not even 1 percent of Fortune 500 companies will have black CEOs. Retrieved May 1, 2017, from http://www.huffingtonpost.com/2015/01/29/black ceos fortune 500_n_6572074.html Bertelli, A. M., & Lynn, Jr., L. E. (2006). agers: Public administration and the Constitution . Baltimore, M.D.: The Johns Hopkins University Press. Bishu, S. G., Guy, M. E., & Heckler, N. (In Press). Seeing gender and its consequences. Journal of Public Affairs Education . https://doi.org/10.1080/152 36803.2018.1565039 Blessett, B. (2018). Symposium on cultural competence, accountability, and social justice: Administrative responsibility and the legitimacy of United States democracy. Public Integrity , 20 (4), 231 234. https://doi.org/10.1080/10999922.20 18.1439566 B.L.S. (2017). USDL 17 0105: U sual weekly earnings of wage and salary workers, fourth quarter 2016 . Wa s hington D.C.

PAGE 187

175 Blyth, M., Hodgson, G. M., Lewis, O., & Steinmo, S. (2011). Introduction to the special issue on the evolution of institutions. J ournal of Institutional Economics , 7 (3), 299 315. Bonilla Silva, E. (2003a). New racism, color blind racism, and the future of Whiteness in America. In A. W. Doane & E. Bonilla Silva (Eds.), White out: The continuing signifi c ance of racism (pp. 271 284). N ew York: Routledge. Bonilla Silva, E. (2003b). Racism without racists: Color blind racism and the persistence of racial inequality in the United States. Lanham, MD: Roman & Littlefield Publishers, Inc. Bonilla Silva, E. (2003c). Racism without racists: Col or blind racism and the persistence of racial inequality in the United States. Lanham, MD: Roman & Littlefield Publishers, Inc. Bonilla Silva, E. (2012). The invisible weight of Whiteness: T he racial grammar of everyday life in contemporary America. Ethnic and Racial Studies , 35 (2), 173 194. Bonilla Silva, E. (2015). More than prejudice: R estatement, reflections, and new directions in critical race theory. Sociology of Race and Ethnicity , 1 (1), 73 87. Bonilla Silva, E., & Dietrich, D. (2011). The sweet ench antment of color blind racism in Obamerica. The ANNALS of the American Academy of Political and Social Science , 634 (1), 190 206. Bourgois, P. (1999). Theory, method, and power in drug and HIV prevention research: A participant Substanc e Use & M i suse , 34 (14), 2155 2172. Boxcryptor.com. (2017). Retrieved June 20, 2017, from https://www.boxcryptor.com/en N egotiating multiple organisational masculinities. Women in Management Review , 17 (3/4), 131 141. https://doi.org/10.1108/09649420210425273 Bozeman, B. (1987). All organizations are public: Bridging public and private organizational theories . San Francisco, CA: Jossey Bass. A test of alternative explanations of differences between public and private organizations. Journal of Public Administration Research and Theory (J PART) , 4 (2), 197 223. Brodkin, K. (1994). How Jews became white folks and what that says about race in America (7th ed.). New Brunswick, NJ: Rutgers University Press. Brown et al. v. Board of Ed. of Topeka et al. , 347 U.S. 483 (1954). Bryer, T. A., & Zavattaro, S. M. (2011). Social media and public admin i stration. Administrative Theo ry & Praxis , 33 (3), 325 340. Burris, B. H. (1993). Technocracy at work . Albany, NY: SUNY Press.

PAGE 188

176 Burris, B. H. (1996). Technocracy, patriarchy and management. In D. L. Collinson & J. Hearn (Eds.), Men as managers, managers as men: Critical perspectives on m en, masculinities, and management (pp. 61 77). Thousand Oaks, CA: Sage. Butler, J. (1990). Gender trouble: Feminism and subversion of identity . New York: Routledge. Cabrera, N. L. (2016). When racism and masculinity collide: S ome methodological considerati ons from a Man of Colour studying Whiteness. Whiteness and Education , 3406 (January), 1 11. https://doi.org/10.1080/13613324.2015.1122662 Campbel, M. L., & Gregor, F. (2002). Mapping social relations: A primer in doing institutional ethnography . Aurora, Ont ario: Garamond Press. Campbell, C. (1987). The romantic ethic and the spirit of modern consumerism . London: Blackwell Publishing Ltd. Carbado, D. W., & Gulati, M. (2013). . Oxford: Oxford University Pre ss. Carlson, L. (1981). Indians, bureaucrats, and land: The Dawes act and the decline of Indian farming . Westport, CT: Greenwood Press. Carrigan, T., Connell, B., & Lee, J. (1985). Toward a new sociology of masculinity. Theory and Society , 14 (5), 551 604. Carter, R. T., Helms, J. E., & Juby, H. L. (2004). The relationship between racism and racial identity for white Americans: a profile an a lysis. Journal of Multicultural Counseling and Development , 32 (1), 2 17. Cascio, J., & Plant, E. A. (2015). Prospective moral licensing: D oes anticipating doing good later allow you to be bad now? Journal of Experimental Social Psychology , 56 , 110 116. Case, A., & Deaton, A. (2015). Rising morbidity and mortality in midlife among white non Hispanic Americans in the 21st ce ntury. Proceedings of the National Academy of Sciences , 112 (49), 15078 15083. https://doi.org/10.1073/pnas.1518393112 Case, A., & Deaton, A. (2017). Mortality and morbidity in the 21st century . Washington, D.C. Center for American Women and Politics. (2017 ). Women in state legislatures 2017. Retrieved December 4, 2017, from http://www.cawp.rutgers.edu/women state legislature 2017 Chamberlain, L. (2017). Pantsuit Nation . New York: Flatiron Books. Chen, K. K. (2009). Enabling creative chaos: The organization behind the burning man event . Chicago: University of Chicago Press.

PAGE 189

177 Choudhury, N. (2007). From the Stasi Commission to the European Court of Human Rights: L Columbia Journal of Gender and Law , 16 (1), 199 296. Clark v. Jeter , 486 U.S. 456 (1988). CNN. (2004). Election results. Retrieved April 13, 2017, from www.cnn.com/ERLECTION/2004/pages/results/states/US/P/00/epolls.0.html CNN. (2008). Exit polls. Retrieved April 13, 2017, from www.cnn.com/ELECTION/2008/results/polls/#USP00p1 E lection center. Retrieved April 13, 2017, from www.cnn.com/election/2012/results/race/president/ CNN. (2016). Election 2016: E xit polls. Retrieved April 13, 2017, from w ww.cnn.com/election/results/exit polls Collinson, D., & Hearn, J. (2003). Naming men as men: I mplications for work, organization and management., 1 (1), 2 22. https://doi.org/10.1111/j.1468 0432.1994.tb00002.x Collinson, D. L., & Hearn, J. (1996). Men as ma nagers, managers as men: Critical perspectives on men, masculinities, and managements . Thousand Oaks, CA: Sage. Connell, R. W., & Messerschmidt, J. W. (2005). Hegemonic masculinity: R ethinking the concept. Gender & Society , 19 (6), 829 859. Cooper, M. (2000 F atherhood, masculinity, and the organization of work in Silicon Valley. Qualitative Sociology , 23 (4), 379 405. Cowles, M. K. (2013). Applied Bayesian statistics: W ith R and OpenBUGS examples . New York: Springer. Crenshaw, K. (199 1). Mapping the margins: I ntersectionality, identity politics, and violence against women of color. Stanford Law Review , 43 (6), 1241 1299. Danso, H. A., & Esses, V. M. (2001). Black experimenters and the intellectual test performance of White participants: T he tables are turned. Journal of Experimental Social Psychology , 37 , 158 165. Darby, S. (2017, September). The rise of the V alkyries: I n the alt right, women are the future, and the problem. , 25 33. Davis, G. F., Diekmann, K. A., & Tins ley, C. H. (1994). The decline and fall of the conglomerate firm in the 1980s: T he deinstitutionalization of an organizational form. American Sociological Review , 59 (4), 547 570.

PAGE 190

178 De Beauvoir, S. (2009). The second sex . (C. Borde & S. Malovany Chevallier, E ds.). New York: Vintage Books. De Croon, E. M., Sluiter, J. K., Kuijer, P. P. F. M., & Frings Dresen, M. H. W. (2005). The effect of office concepts on worker health and perfo r mance: A systematic review of the literature. Ergonomics , 48 (2), 119 134. Debesa y, J., Nåden, D., & Slettebø, Å. (2008). How do we close the hermeneutic circle? A Gadamerian approach to justification in interpretation in qualitative studies. Nursing Inquiry , 15 (1), 57 66. Dellinger, K., & Williams, C. L. (1997). Makeup at work: N egoti ating appearance rules in the workplace. Gender & Society , 11 (2), 151 177. DiAngelo, R. (2011). White fragility. International Journal of Critical Pedagogy , 3 (3), 54 70. DiMaggio, P. J., & Powell, W. W. (1983). The iron cage revisited: I nstitutional isomor phism and collective rationality in organizational fields. American Sociological Review , 48 (2), 147 160. Dobbin, F. (2005). Comparative and historical approaches to economic sociology. In N. J. Smelser & R. Swedberg (Eds.), The handbook of economic sociolo gy (2nd ed.). Princeton, NJ: Princeton University Press. Dorie, V. (2015). Bayesian linear mixed effects models (Version 1.0 4) [R]. CRN. Retrieved from https://github.com/vdorie/blme/issues Doucet, A. (2018). Do men mother?: Fathering, care, and parental responsibility (2d ed.). Toronto, Canada: University of Toronto Press. Dragon Professional. (2017). (Version 15). Retrieved from https://www.nuance.com/dragon/dragon for pc.html Du Bois, W. E. B. (1903). The souls of black folk: Essays and sketches (3rd ed .). Chicago: A.C. McClurg & Co. Du Bois, W. E. B. (1935). Black reconstruction in America: Toward a history of the part which black folk played in the attempt to reconstruct democracy in America, 1860 1880 . New York: Harcourt, Brace, and Co. Du Bois, W. E. B. (1980). W. E. B. Du Bois on sociology and the black community . (D. S. Green & E. D. Driver, Eds.). Chicago: University of Chicago Press. Dunleavy, P., Margetts, H., Bastow, S., & Tinkler, J. (2006). New public management is dead long live digital era governance. Journal of Public Administration Research and Theory (J PART) , 16 (3), 467 494. Edsal, M., & Edsal, T. B. (1991). Chain reaction . London: Norton.

PAGE 191

179 Eikenberry, A. M. (2009). The present and (normative) future of public administration and implicat ions for ASPA. Public Administration Review , 69 (6), 1060 1067. Remedial and Sp ecial Education , 26 (1), 25 31. Eliot, L. (2009). Pink brain, blue brain: H ow small differences grow into troublesome gaps -and what we can do about it . New York: Houghton Mifflin Harcourt. Equal Employment Opportunities Act, 42 USC § 2000e (1974). Esterbe rg, K. G. (2002). Qualitative methods in social research . Boston, MA: McGraw Hill. Fanon, F. (1963). The wretched of the earth . New York: Grove Press. Ferguson, K. (1984). The feminist case against bureaucracy . Philadelphia, PA: Temple University Press. Fl and the debate around it. National Public Radio . Foldy, E. G., & Buckley, T. R. (2014). The color bind: Talking (and not talking) about race at work . New York: Russel Sa ge. Follett, M. P. (1924). Creative experience . London: Longmans, Green and Co. Fortin, J. (2017, February 17). Who is Sebastian Gorka? A Trump adviser comes out of the shadows. New York Times . Foucault, M. (1977). Discipline and punish: The birth of the p rison . (A. Sheridan, Ed.). New York: Vintage Books. Frankenberg, R. (2001). The mirage of an unmarked whiteness. In B. B. Rasmussen, E. Klinenberg, I. J. Nexica, & M. Wray (Eds.), The making and unmaking of whiteness (pp. 72 96). Durham, NC, NC: Duke Unive rsity Press. Franzway, S. (2001). Sexual politics and greedy institutions. Sydney, Australia: Pluto. Fredrickson, G. M. (1988). The arrogance of race: Historical perspectives on slavery, racism, and social inequality . Middletown, CT: Wesleyan University Pr ess. Fryar, A. H., & Hawes, D. P. (2012). Competing explanations for minority enrollments in higher education. Journal of Public Administration Research and Theory , 22 (1), 83 99. https://doi.org/10.1093/jopart/mur009 Fürstenberg, K. (2016). Evolutionary in stitutionalism. Politics & the Life Sciences , 35 (1), 48 60.

PAGE 192

180 Gadamer, H. G. (1976). Philosophical hermeneutics . (D. E. Linge, Ed.). Berkeley, CA: University of California Press. Gallagher, C. A. (2003). Playing the white ethnic card: using ethnic identity t o deny contemporary racism. In A. W. Doane & E. Bonilla Silva (Eds.), White out: The continuing signifi c ance of racism (pp. 146 158). New York: Routledge. Gallup. (2017). State of the American workplace . Washington, D.C. Gambrell, K. M. (2016). Lakota wome n leaders: Getting things done quietly. Leadership , 12 (3), 293 307. https://doi.org/10.1177/1742715015608234 Garcia, A. C., Standlee, A. I., Bechkoff, J., & Cui, Y. (2009). Ethnographic approaches to the internet and computer mediated communication. Journa l of Contemporary Ethnography , 38 (1), 52 84. The production of reality: Essays and readings on social interaction (pp. 381 3 92). Thousand Oaks, CA: Sage. Geduldig v. Aiello , 417 U.S. 484 (1974). Gerring, J. (2012). Social science methodology: A unified framework. Cambridge, England: Cambridge Univer s i ty Press. United States. Gill, J., & Meier, K. J. (2000). Public administration research and practice: A m ethodological manifesto. Journal of Public Administration Research and Theory (J PART) , 10 (1), 159 199. Gilligan, J. (1996). Violence: Reflections on a national epidemic . New York: Vintage Books. Gilligan, J. (2003). Shame, guilt, and violence. Social Rese arch , 70 (4), 1149 1180. Glauber, R. (2008). Race and gender in families and at work: T he fatherhood wage premium. Gender & Society , 22 (1), 8 30. Global Workplace Analytics. (2017). 2017 state of telecommuting in the U.S. employee workforce . Carlsbad, CA. G oertz, G. (2006). . Princeton, NJ: Princeton University Press. Goldstein, J. S. (2001). War and gender . Cambridge, England: Cambridge University Press. Gooden, S. T. (2014). Race and social equity: A nervous area of g overnment . Armonk, NY: M.E. Sharpe.

PAGE 193

181 Gooden, S. T. (2015). From equality to social equity. In M. E. Guy & M. M. Rubin (Eds.), Public Administration Evolving (pp. 210 230). New York: Routledge. Goodsell, C. T. (1983). The case for bureaucracy: A public admin istration polemic (2nd ed.). Chatham, NJ: Chatham House Publishers, Inc. Granovetter, M. (1985). Economic action and social structure: T he problem of embeddedness. American Journal of Sociology , 91 (3), 481 510. Greenwald, A. G., & Banaji, M. R. (1995). Imp licit social cognition: A ttitudes, self esteem, and stereotypes. Psychological Review , 102 (1), 4 27. Greenwald, A. G., Banaji, M. R., & Nosek, B. A. (2015). Statistically small effects of the implicit association test can have societally large effects. Jou rnal of Personality and Social Psychology , 108 (4), 553 561. Guglielmo, T. A. (2003). White on arrival: Italians, race, color, and power in Chicago, 1890 1945 . Oxford: Oxford University Press. Gulick, L. H. (1937). Notes on the theory of organization . New Y ork: Columbia University Press. Guy, M. E. (1993). Three steps forward, two steps backward: T into public management. Public Administration Review , 53 (4), 285 292. Guy, M. E., & Mccandless, S. A. (2012). Social equity: I ts l egacy, its promise. Public Administration Review , 72 (S1), S5 S13. Guy, M. E., & Newman, M. A. (2005). Valuing diversity. In Handbook of human resource management and government (pp. 143 163). San Francisco, CA, CA: Jossey Bass. Guy, M. E., Newman, M. A., & Mastracci, S. H. (2008). Emotional labor: Putting the service in public service . Armonk, NY: M.E. Sharpe. Hallett, R. E., & Barber, K. (2014). Ethnographic resea r ch in a cyber era. Journal of Contemporary Ethnography , 43 (3), 306 330. Hamilton, A., Madison , J., & Jay, J. (1894). The federalist and other constitutional papers . (E. H. Scott, Ed.). Chicago: Albert, Scott, & Company. Hanisch, C. (2006). The personal is political. Carol Hanish.Org . Harris, C. I. (1993). Whiteness as property. Harvard Law Review , 106 (8), 1707 1791. Harro, B. (2013). The cycle of liberation. In M. Adams, W. J. Blumenfeld, R. Castañeda, H. W. Hackman, M. L. Peters, & X. Zúñiga (Eds.), Readings for diversity and social justice: An anthology on racism, antisemitism, sexism, heterosexi sm, ableism, and classism (pp. 463 469). New York: Routledge.

PAGE 194

182 Harvard Project on American Indian Economic Development. (2008). The s t ate of the native nations: Conditions under U.S. policies of self determination . New York, N.Y.: Oxford University Press. H eckler, N. (In Press). Whiteness and Masculinity in nonprofit organizations: L aw, money, and institutional race and gender. Administrative Theory & Praxis . Heckler, N. (2017). Publicly desired color blindness: W hiteness as a realized public value. Administ rative Theory & Praxis , Forthcomin , 1 29. http://dx.doi.org/10.1080/10841806.2017.1345510 Helms, J. E. (1990). Black and white racial identity: Theory, research, and practice. Contributions in Afro American and African studies, No. 129 . New York: Greenwood Press. Henderikse, W., van Beek, A., Pouwels, B., & Vandoornehuiskes en Partners. (2017). The efficacy of strategies to elevate gender equality in leadership: A andell, & A. M. Dare (Eds.), Gender, communication, and the leadership gap (pp. 127 149). Charlotte, N.C.: Information Age Publishing Inc. Henry, G. T. (1990). Practical sampling: Applied social research methods series vol. 21 . Newbury Park, CA, CA: Sage. Herrnstein, R. J., & Murray, C. (1994). The bell curve: Intelligence and class structure in American life . Ann Arbor, MI: University of Michigan Press. Hoffman, M. C. (2002). Paradigm lost: P ublic administration at Johns Hopkins University, 1884 96. Public Administration Review , 62 (1), 12 23. Hooks, B. (1984). Feminist theory: From margin to center . Boston, MA: South End Press. hooks, B. (2014). Talking back: Thinking feminist, thinking black . New York: Routledge. Horton, S. B. (2016). They leave their kidn eys in the fields: Illness, injury and illegality among U.S. farmworkers . Oakland, CA: University of California Press. Ignatiev, N. (1995). How the Irish became white . New York: Routledge. Ignatiev, N. (1997). Treason to whiteness is loyalty to humanity. I n R. Delgado & J. Stefancic (Eds.), Critical whiteness studies: Looking behind the mirror (eBook, pp. 607 ). Philadelphia, PA: Temple University Press. Ignatiev, N., & Garvey, J. (1996). Race traitor . New York: Routledge. Ingraham, P. W. (1993). Of pigs in pokes and policy diffusion: A nother look at pay for performance. Public Administration Review , 53 (4), 348 356.

PAGE 195

183 Ingraham, P. W., Thompson, J. R., & Sanders, R. P. (1998). Transforming government: Lessons from the reinvention laboratories . San Francisco, CA : Jossey Bass. Jones, B. D. (2003). Bounded rationality and political science: L essons from public administration and public policy. Journal of Public Administration Research and Theory , 13 (4), 395 412. https://doi.org/10.1093/jpart/mug028 Jones, J. M. (20 16). Six in 10 Americans say racism against blacks is widespread. Retrieved April 17, 2017, from http://www.gallup.com/poll/194657/six americans say racism against blacks widespread.aspx Jones, M. D., Shanahan, E. A., & McBeth, M. K. (2014). The science of stories: Applications of the narrative policy framework in public policy analysis . New York: Palgrave Macmillan. Jun, S., Lowery, B. S., & Guillory, L. (2017). Keeping minorities happy: H ierarchy maintenance fied white politicians. Personality and Social Psychology Bulletin , Pending . Kahneman, D. (2003). A perspective on judgment and choice: M apping bounded rationality. The American Psychologist , 58 (9), 697 720. Kahneman, D. (2011). Thinking fast and slow . New York: Farrar, Straus and Giroux. Kahneman, D., & Tversky, A. (1984). Choices, values, and frames. American Psychologist , 39 (4), 341 350. Kanter, R. M. (1977). Men and women of the corporation . New York: Basic Books. Katz, J. (2013). Leading men: President ial campaigns and the politics of manhood . Northampton, MA: Interlink Publishing. Katz, J., Heisterkamp, H. A., & Fleming, W. M. (2011). The social justice roots of the mentors in violence prevention model and its application in a high school setting. Viol ence Against Women , 17 (6), 684 702. Kaufman, H. (2006). The forest ranger: A study in administrative behavior (Specal Rep). Washington, D.C.: Resources for the Future. Keiser, L. R., Wilkins, V. M., Meier, K. J., & Holland, C. A. (2002). Lipstick and logar ithms: G ender, institutional context, and representative bureaucracy. American Political Science Review , 96 (3), 553 564. m asculinity on an o nline f orum. Gender & Society , 14 (2), 256 274.

PAGE 196

184 Kerlin, J. A. (20 06). Social enterprise in the United States and Europe: U nderstanding and learning from the differences. VOLUNTAS: International Journal of Voluntary and Nonprofit Organizations , 17 (3), 246 262. Kingsley, J. D. (1944). Representative bureaucracy: An interp retation of the British Civil Service . Antioch, NY: Antioch Press. Kline, R. B. (2004). Beyond significance testing: Statistics reform in the behavioral sciences. Was hington, D.C.: American Psychological Association. Klinesmith, J., Kasser, T., & McAndrew, F. T. (2006). Guns, testosterone, and aggression: An experimental test of a mediational hypothesis. Psychological Science , 17 (7), 568 571. Kochan, T. effects of diversity on business performance: Report of the diversity research network. Human Resource Management , 42 (1), 3 21. https://doi.org/10.1002/hrm.10061 Koenig, A . M., & Eagly, A. H. (2005). Stereotype threat in men on a test of social sensitivity. Sex Roles , 52 (7/8), 489 496. Kruschke, J. K. (2012). Bayesian estimation supersedes the t test. Journal of Experimental Psychology: General , 142 (2), 573 603. https://doi .org/10.1037/a0029146 Kuhn, D. P. (2009). What white women want, surprisingly the GOP. Retrieved April 13, 2017, from www.realclearpolitics.com/articles/2009/11/13/what_white_women_wnat_surprisingly_t he_gop_99144.html Lawrence, P. R., & Lorsch, J. W. (1967 ). Organization and environment . Boston, MA: Harvard Business School Press. LeCompte, M. D., & Schensul, J. J. (2010). Designing and conducting ethnographic research: An introduction (2nd ed.). Lanham, MD: Rowman & Littlefield Publishers, Inc. Lee, J. C. ( New York Times . W hat children of color learn from white women and what this says about race, whiten ess, and gender. In M. Lynn & A. D. Dixson (Eds.), Handbook of c ritical r ace t heory in e ducation (pp. 313 324). New York: Routledge. Leonardo, Z., & Hunter, M. (2009). Race, class, and imagining the urban. In Z. Leonardo (Ed.), Race, Whiteness, and e ducati on (eBook, pp. 228 262). New York: Routledge. Liberman, K. (2013). More studies in ethnomethodology . Purchase, NY: SUNY press.

PAGE 197

185 Light, P. C. (1997). The tides of reform: Making government work, 1945 1995 . New Haven, CT: Yale University Press. Light, P. C. ( 2017). 1984 2015 (Issue Paper). New York, N.Y.: The Volcker Alliance. Retrieved from https://www.volckeralliance.org/sites/default/files/attachments/Issue%20Paper_True%20 Size%20of%20Gove rnment.pdf Lindblom, C. E. (1977). . New York: Basic Books. Lindstead, S. (2000). Comment: Gender blindness or gender suppression? A comment on Fiona Organization Studies , 21 (1), 297. https://doi.org/10.1177/0170840600211007 Lindzon, J. (2016). There are now more Fortune 500 companies with all male boards than in 2015. Retrieved May 1, 2017, from http://fortune.com/2016/03/28/fortune 500 all male board 2016/ Lippa, R. a., Ma rtin, L. R., & Friedman, H. S. (2000). Gender related individual differences and mortality in the Terman longitudinal study: I s masculinity hazardous to your health? Personality and Social Psychology Bulletin , 26 (12), 1560 1570. Lipsitz, G. (2009). The pos sessive investment in whiteness: How white people profit from identity politics (2nd ed.). Philadelphia, PA: Temple University Press. Lynn Jr., L. E. (2006). Public management: Old and new . New York: Routledge. E xploring whiteness and power in organizations. Management Learning , 36 (4), 429 450. Mackinnon, C. A. (1989). Sex P Ethics , 99 (2), 314 346. Majors, R., & Billson, J. M. (1992). Cool pose: T he dilemmas of black manhood in America . New York: Touchstone. Marchenko, Y. (2015). Bayesian analysis using Stata. Mars hall, C., & Rossman, G. B. (2011). Qualitative research genres. In Designing qualitative research (2nd ed., pp. 17 37). Thousand Oaks, CA: Sage. Martin, D. D. (2000). Organizational approaches to shame: A vowal, management, and contestation. The Sociologica l Quarterly , 41 (1), 125 150. Martin, J. (1990). Deconstructing organizational taboos: T he suppression of gender conflict in organizations. Organization Science , 1 (4), 339 359.

PAGE 198

186 Marx, K. (2004). The German ideology. In F. Dobbin (Ed.), The New Economic Socio logy: A Reader . Princeton, NJ: Princeton University Press. Masson, M. E. J. (2011). A tutorial on a practical Bayesian alternative to null hypothesis significance testing. Behavior Research Methods , 679 690. Mastracci, S., & Arreola, V. I. (2016). Gendered organizations: How human resource management practices produce and reproduce administrative man. Administrative Theory & Praxis , 38 (2), 137 149. https://doi.org/10.1080/10841806.2015.1130523 A teacher educator of color unveiling the dangerous minds of white teacher candidates. Teacher Education Quarterly , 40 (2), 53 73. Matias, C. E. (2016). Feeling white: Whiteness, emotionality, and education . Rotterdam, Netherlands: Sense Publishers. Matias, C. E., & Mackey, J. ( I ntroducing c ritical Whiteness p edagogy. The Urban Review , 48 (1), 32 50. https://doi.org/10.1007/s11256 015 0344 7 Jeff Sessions. New York Times . McClain, C. (2012). How presidential elections are impacted by a 100 million year old coastline. Retrieved January 4, 2017, from http://www.deepseanews.com/2012/06/how presidential elections are impacted by a 100 million yea r old coastline/ Meghani, S. H., Byun, E., & Gallagher, R. M. (2012). Time to take stock: A meta analysis and systematic review of analgesic treatment disparities for pain in the United States. Pain Medicine , 13 (2), 150 174. Meier, K. J., Mastracci, S. H., & Wilson, K. (2013). Gender and emotional labor in public organizations: An empirical examina tion of the link to performance . Public Administration Review, 66 (6), 899 909. Meier, K. J., & O Toole, L. J. (2013) . Subjective organizational performance and measurement error: Common source bias and spurious relationships. Journal of Public Administration Research and Theory (J PART), 23 (2), 429 456. Meyer, J. W., & Rowan, B. (1977). Institutionalized organizations: F ormal structure as myth and ceremony. America n Journal of Sociology , 83 (2), 340 363. New York Times , p. A3. Mills, C. W. (1997). The racial contract (eBook). Cornell, NY: Cornell University Press.

PAGE 199

187 Mills, C. W. (2007). White ignorance. In S. Sullivan & N. Tuana (Eds.), Race and the epistemologies of ignorance (pp. 13 35). Albany, NY: SUNY Press. Mintzberg, H. (1979). The structuring of organizations . Englewood Cliffs, NJ: Prentice Hall. Mintzberg, H. (1983). Power in and around organizations . Englewood Cliffs, NJ: Prentice Hall. Morgan, D. (1996). The gender of bureaucracy. In D. L. Collinson & J. Hearn (Eds.), Men as managers, managers as men: Critical perspectives on men, masculinities, and management (pp. 43 60). Thous and Oaks, CA: Sage. Moulton, S. (2009). Putting together the publicness puzzle: A framework for realized publicness. Public Administration Review , 69 (5), 889 900. Mumby, D. K. (2006). Organizing men: P ower, discourse, and the social construction of masculi nity(s) in the workplace. Communication Theory , 8 (2), 164 183. Mumby, D. K., & Putnam, L. L. (1992). The politics of emotion: A feminist reading of bounded rationality. Academy of Management Review , 17 (3), 465 486. Murray, C. (2012). Coming apart: The stat e of white America, 1960 2010 . New York: Cox and Murray, Inc. NBC News/Wall Street Journal Survey . (2016). New York. Nelson, T. E., Acker, M., & Manis, M. (1996). Irrepressible stereotypes. Journal of Experimental Social Psychology , 32 (2), 13 38. Never, B. , & Westberg, D. (2016). Place matters: T he spatial effects of human service expenditures. Nonprofit Policy Forum , 6 (1), 1 20. Neville, H. A., Coleman, M. N., Falconer, J. W., & Holmes, D. (2005). Color blind racial ideology and psychological false conscio usness among African Americans. Journal of Black Psychology , 31 (1), 27 45. Newman, M. A., Guy, M. E., & Mastracci, S. H. (2009). Beyond cognition: A ffective leadership and emotional labor. Public Administration Review , 69 (1), 6 20. he political is personal: T he influence of white supremacy on white Silva (Eds.), White out: The continuing signifi c ance of racism (pp. 254 267). New York: Routledge. 2012). Collaborative public management: W here have we been and where are we going? The American Review of Public Administration , 42 (5), 507 522. Oliver, M. L., & Shapiro, T. M. (2006). Black wealth/white wealth: A new perspective on racial inequality . New York: Routledge.

PAGE 200

188 Office of Personnel Management (OPM) . (2015). Diversity & Inclusion: Federal Workforce AT A Glance OPM.gov. Retrieved September 26, 2017, from https://www.opm.gov/policy data oversight/diversity and inclusion/federal workforce at a glanc e/ Office of Personnel Management (OPM) . (201 8 ). Federal employee viewpoint survey (FEVS) . Washington, D.C. Office of Personnel Management ( OPM ) . (2017). Status of telework in the federal government: Report to Congress (Report to Congress). Washington, D. C.: United States Office of Personnel Management. Retrieved from https://www.telework.gov/reports studies/reports to congress/2017 report to congress.pdf Orfield, G., Frankenberg, E., Ee, J., & Kuscera, J. (2014). Brown at 60: G reat progress, a long retrea t and an uncertain future . Berkeley, CA. Ostrom. (2005). Understanding institutional diversity . Princeton, NJ: Princeton University Press. Ostrom, E. (1990). Governing the commons . Cambridge, England: Cambridge University Press. Ostrom, E. (2000). Collecti ve action and the evolution of social norms. The Journal of Economic Perspectives , 14 (3), 137 158. Otis, D. S. (1973). The Dawes Act and the allotment of Indian lands . (F. P. Prucha, Ed.). Norman, Oklahoma: University of Oklahoma Press. Page, S. (2005). Wh A dministrative change in the human services. Public Administration Review , 65 (6), 713 727. shield. The Washington Pos t . Pascoe, C. J. (2007). M asculinity and sexuality in high school . Berkeley, CA: University of California Press. Pelizza, A., & Hoppe, R. (2015). Birth of a failure: C onsequences of framing ICT projects for centralization of inter depar tmental relations. Administration & Society , (Pending), 1 30. Perry, J. L. (1996). Measuring public service motivation: A n assessment of construct reliability and validity. Journal of Public Administration Research and Theory (J PART) , 6 (1), 5 22. Perry, J . L., & Wise, L. R. (1990). The motivational bases of public service. Public Administration Review , 50 (3), 367 373. Pigott, R. L., & Cowen, E. L. (2000). Teacher race, child race, racial congruence, and teacher Jour nal of School Psychology , 38 (2), 177 195.

PAGE 201

189 Pini, B., & Pease, B. (2013). Men, masculinities, and methodologies . New York: Palgrave Macmillan. Pitts, D. W. (2009). Diversity management, job satisfaction and performance: E vidence from U.S. federal agencies. P ublic Administration Review , 69 (2), 328 338. Pollner, M. (2012). The end(s) of ethnomethodology. American Sociologist , 43 (1), 7 20. Portillo, S., & Humphrey, N. (2018). Institutionalism and assumptions: Institutionalizing race and gender in public administ ration scholarship. In Handbook of American Public Administration (pp. 289 303). Cheltenham, UK: Edward Elgar Publishing. Ray, V. (In Press). A theory of racialized organizations. American Sociological Review . https://doi.org/10.1177/0003122418822335 Regen ts of the Univ. of Calif. v. Bakke , 438 U.S. 265 (1978). Rethemeyer, R. K. (2006). Policymaking in the age of internet: I s the internet tending to make policy networks more or less inclusive. Journal of Public Administration Research and Theory (J PART) , 1 7 (2), 259 284. Riccucci, N. M. (1997). Cultural diversity programs to prepare for work force 2000: W wrong? Public Personnel Management , 26 (1), 35 41. Riccucci, N. M. (2002). Managing diversity in public sector workforces . Boulder, Colorado: West view Press. Riccucci, N. M. (2010). Public administration: Traditions of inquiry and philosophies of knowledge (eBook). Washington D.C.: Georgetown University Press. Riccucci, N. M., & Van Ryzin, G. G. (2017). Representative bureaucracy: A lever to enhance social equity, coproduction, and democracy. Public Administration Review , 77 (1), 21 30. Riccucci, N. M., Van Ryzin, G. G., & Li, H. (2015). Representative bureaucracy and the willingness to coproduce: A n experimental study. Public Administration Review , 7 6 (1), 121 130. Rice, K. (2014). Bayesian Statistics (a very brief introduction). Roberts, A. (2014). Large Forces . Seattle, WA: CreateSpace Independent Publishing Platform. Roediger, D. R. (1999). The wages of whiteness: Race and the making of the American working class . (Revised, Ed.). London: Verso. Rohr, J. A. (1990). The constitutional case for public administration. In G. L. Wamsley, R. N. Refounding Public Administr ation . Newberry Park, CA: Sage.

PAGE 202

190 Romzek, B. S., & Johnston, J. M. (2005). State social services contracting: E xploring the of effective determinants contract accountability. Public Administration Review , 65 (4), 436 449. Ronquillo, J. C. (2011). American I nd ian tribal governance and management: P ublic administration promise or pretense? Public Administration Review , 71 (2), 285 292. Rosenbloom, D. H., & Piotrowski, S. J. (2005). Outsourcing the constitution and administrative law norms. The American Review of Public Administration , 35 (2), 103 121. Rousseau, J. J. (1896). Du contrat social: Ou principes du droit politique . Roy, W. G. (1997). Socializing capital: The rise of the large industrial corporation in America . Princeton, NJ: Princeton University Press. S andberg, B. (2016). Against the cult(ure) of the entrepreneur for the nonprofit sector. Administrative Theory & Praxis , 38 (1), 52 67. Sandberg, S. (2013). Lean in: Women, work, and the will to lead . New York: Alfred A. Knopf. Savage, C. (2017, August 1). J ustice Dept. to take on affirmative action in college admissions. New York Times , p. A1. Scales, A. (1992). Feminist legal method: N ot so scary. , 2 (1), 1 34. Scales, A. (2006). Legal feminism: Activism, lawyering, and legal theory (e book). New York: New York University Press. Scales, A. (2009). Student gladiators and sexual assault: A new analysis of liability for injuries inflicted by college athletes. Michigan Journal of Gender & Law , 15 (2), 205 289. Scales, A. (2013). Soft on de fense: T he failure to confront militarism. Berkeley Journal of Gender, Law & Justice , 20 (1), 369 393. Schein, E. H. (2010). Organizational culture and leadership (4th ed.). San Francisco, CA: Jossey Bass. Schmader, T. (2002). Gender identification moderate performance. Journal of Experimental Social Psychology , 38 , 194 201. Schmidt, M. R. (1993). Grout: alternative kinds of knowledge and why they are ignored. Public Administration Review , 53 (6), 525 530. Schneider, A. L., & Ingram, H. (1993). Social construction of target populations: implications for politics and policy. American Political Science Review , 87 (2), 334 347. Scott, W. R. (2001). Institutions and organizations: Ideas and interests (2nd ed.). Thousand Oa ks, CA: Sage.

PAGE 203

191 Scott, W. R. (2005). Institutional theory: Contributing to a theoretical research program. Great , (June), 460 484. Selden, S. C. (1997). The promise of representative bureaucracy: Divers ity and responsiveness in a government agency . Armonk, NY: M.E. Sharpe. Sherman, H. J. (1996). A holistic evolutionary view of racism, sexism, and class inequality. In W. M. Dugger (Ed.), Inequality: Radical institutoinal views on race, gender, class, and nation (pp. 37 50). Westport, CT: Greenwood Press. Silverman, J. M. (1987). Gender differences in attitudes toward nuclear war and disarmament. Sex Roles , 16 (3/4), 189 203. Simon, H. A. (1959). Theories of decision making in economics and behavioral scienc e. American Economic Review , 49 (3), 253 283. Singer, M. (2000). A dose of drugs, a touch of violence, a case of AIDS: conceptualizing the SAVA syndemic. Free Inquiry in Creative Sociology , 28 (1), 13 24. Smith, D. E. (1987). The everyday world as problemati c: A feminist sociology . Toronto: University of Toronto Press. Smith, D. E. (2005). Institutional ethnography: A sociology for people . Lanham, MD: Altamira. Smith, D. E. (2006). Institutional ethnography as practice . Oxford: Rowman & Littlefield Publishers , Inc. Smith, J. L., & White, P. H. (2002). An examination of implicitly activated, explicitly activated, and nullified sterotypes on mathematical performance: I Sex Roles , 47 (3/4), 179 191. Solorzano, D. G. (1997). Images and words that wound: C ritical race theory, racial stereotyping, and teacher education. Teacher Education Quarterly , 24 (3), 5 19. Starke, A., Heckler, N., & Mackey, J. (2018). Administrative racism: public administration education and race. Journal of Public A ffairs Education , 24 (4), 469 489. https://doi.org/10.1080/15236803.2018.1426428 Status of women in the States. (2016). Political participation. Retrieved December 1, 2016, from http://statusofwomendata.org/explore the data/political participation/ Steele, C. M. (2010). Whistling Vivaldi . New York: W.W. Norton & Co. Stepler, R. (2016). 5 key takeaways about views of race and inequality in America . Washington, D.C. Stivers, C. (1993). Gender images in public administration: Legitimacy and the administrative s tate . Newbury Park, CA: Sage.

PAGE 204

192 Stivers, C. (2000). Bureau men, settlement women: Constructing public administration in the progressive era . Lawrence, KA: University of Kansas Press. Stivers, C. (2008). Governance in dark times: Practical philosophy for publ ic service . Washington, D.C.: Georgetown University Press. Stivers, C. (2018). Beyond calculation: Experience and language in administrative situations. Administrative Theory & Praxis , 40 (3), 200 210. https://doi.org/10.1080/10841806.2018.1485448 Sue, D. W . (2004). Whiteness and ethnocentric monoculturalism: M The American Psychologist , 59 (8), 761 769. Sue, D. W. (2010). Microaggressions in everyday life: Race, gender, and sexual orientation . Hoboken, NJ: John Wiley & Sons, Inc . Thandeka. (1999). Learning to be White: Money, race, and God in America . New York: Continuum. Theobald, N. A., & Haider Markel, D. P. (2009). Race, bureaucracy, and symbolic representation: Interactions between citizens and police. Journal of Public Admi nistration Research and Theory , 19 (2), 409 426. https://doi.org/10.1093/jopart/mun006 Therneau, T. M., & Atkinson, E. J. (2018). An introduction to recursive partitioning using the RPART routines . CRAN R project. Retrieved from https://cran.r project.org/p ackage=rpart Trump policies. New York Times , p. A20. Turner, R. W., Perrin, E. M., Coyne Beasley, T., Peterson, C. J., & Skinner, A. C. (2015). Reported sports participation, race, sex, ethnicity, and obesity in US adole s cents from NHANES physical activity. Global Pediatric Health , 2 (1), 1 9. Turner, S. M. (2006). Mapping institutions as work and texts. In D. E. Smith (Ed.), Institutional ethnography as practice (pp. 139 161). Lanham, MD: Rowman & Littlefield Publishers, Inc. Ulin, P. R., Robinson, E. T., Tolley, E. E., & McNeill, E. T. (2002). Qualitative methods: A field guide for applied research in sexual and reproductive health . Triangle Park, NC: Family Hea lth International. Uzzi, B. (1999). Embeddedness in the m aking of f inancial c apital: How s ocial r elations and n etworks b enefit f irms s eeking f inancing. American Sociological Review , 64 (4), 481 505.

PAGE 205

193 Van Ryzin, G. G., Riccucci, N. M., & Li, H. (2016). Repres entative bureaucracy and its symbolic effect on citizens: A conceptual replication. Public Management Review , Publicatio , 1 16. Vance, J. D. (2016). Hillbilly elegy: A memoir of a family and a culture in crisis . New York: HarperCollins. Vera, H., & Gordon, A. M. (2003). The beautiful American: S incere fictions of the White messiah in Hollywood movies. In A. W. Doane & E. Bonilla Silva (Eds.), White out: The continuing signifi c ance of racism (pp. 113 125). New York: Routledge. Wamsley, G. L., Bacher, R. N., Wolf, J. F. (1990). Refounding public administration . Newberry Park, CA: Sage. Watkins Hayes, C. (2011). Race, respect, and red tape: I nside the black box of racially representative bureaucrac ies. Journal of Public Administration Research and Theory (J PART) , 21 (SUPPL. 2), 233 251. Weaver, T. L. (2015). Intent to exit: W hy do US federal employees leave? International Journal of Public Administration , 38 (6), 442 452. Weber, M. (1972). From Max W eber: Essays in sociology . (H. H. Gerth & C. W. Mills, Eds.). Oxford: Oxford University Press. Weber, M. (2004). From t he p rotestant e thic and t he s pirit of c apitalism. In F. Dobbin (Ed.), The n ew e conomic s ociology: A r eader (pp. 49 85). Princeton, NJ: Pr inceton University Press. Weber, M. (2005). The protestant ethic and the spirit of capitalism . (T. Parsons, Ed.). Mineola, New York: Dover Publications Inc. https://doi.org/10.2307/3211491 Weick, K. E. (1995). Sensemaking in organizations . Thousand Oaks, C A: Sage. Weiss, J. W. (2014). Business ethics: A stakeholder and issues management approach . San Francisco, CA: Berrett Koehler Publishers, Inc. Weitz, R. (2001). Women and their hair: seeking power through resistance and accommodation. Gender & Society , 1 5 (5), 667 686. Wilson, J. Q. (1989). Bureaucracy: What government agencies do and why they do it . New York: Basic Books. Wilson, W. (1887). The study of administration. Political Science Quarterly , 2 (2), 197 222. Wise, T. (2011). White like me: Reflections on race from a privileged son . Berkeley, CA: Soft Skull Press.

PAGE 206

194 Witt, M. T. (2006). Notes from the margin: R ace, relevance and the making of public administration. Administrative Theory & Praxis , 28 (1), 36 68. Woods, R. A. (2017). Sports and exercise ( s pot light on s tatistics). Washington, D.C.: U.S. Bureau of Labor Statistics. Retrieved from https://www.bls.gov/spotlight/2017/sports and exercise/pdf/sports and exercise.pdf Wooldridge, J. M. (2013). Introductory econometrics: A modern approach (5th ed.). And over, MA: Cengage Learning. Yancy, G. (2008). Black bodies, white gazes: The continuing significance of race. Lanham, MD: Roman & Littlefield Publishers, Inc. Yetano, A., & Royo, S. (2017). Keeping citizens engaged: A comparison between online and offline participants. Administration & Society , 49 (3), 394 422. Zuberi, T. (2000). Deracializing social statistics: P roblems in the quantification of race. The ANNALS of the American Academy of Political and Social Science , 568 (1), 127 134. Zuberi, T., & Bonilla S ilva, E. (2008). White logic, white methods: Racism and methodology . Lanham, MD: Rowman & Littlefield Publishers, Inc.

PAGE 207

195 APPENDI X Figure A1: Unit level ANOVA regression tree on Comparative White Job Satisfaction.

PAGE 208

196 Figure A 2 : Unit level ANOVA regression tree on Comparative White Management Satisfaction.

PAGE 209

197 Figure A 3 : Unit level ANOVA regression tree on Comparative White PSM.

PAGE 210

198 Figure A 4 : Unit level ANOVA regression tree on Comparative Job Satisfaction.

PAGE 211

199 F igure A 5 : Unit level ANOVA regression tree on Comparative Satisfaction.

PAGE 212

20 0 Figure A 6 : Unit level ANOVA regression tree on Comparative PSM.

PAGE 213

201 F igure A 7 : Unit level ANOVA reg ression tree on Comparative Satisfaction.

PAGE 214

202 Figure A 8 : Unit level ANOVA regression tree on Comparative Management Satisfaction.

PAGE 215

203 Figure A 9 : Unit level ANOVA regression tree on Comparative PSM.