Citation
The effects of the status characteristic of sex on organizational choice

Material Information

Title:
The effects of the status characteristic of sex on organizational choice
Creator:
Holm, Melissa Jean
Publication Date:
Language:
English
Physical Description:
x, 111 leaves : forms ; 29 cm

Subjects

Subjects / Keywords:
Sex role in the work environment ( lcsh )
Conflict management ( lcsh )
Conflict management ( fast )
Sex role in the work environment ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references.
General Note:
Submitted in partial fulfillment of the requirements for the degree, Master of Arts, Department of Sociology.
Statement of Responsibility:
by Melissa Jean Holm.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
26882449 ( OCLC )
ocm26882449
Classification:
LD1190.L66 1992m .H64 ( lcc )

Full Text
THE EFFECTS OF THE STATUS CHARACTERISTIC
j OF SEX
:j ON ORGANIZATIONAL CHOICE
I
by
I Melissa Jean Holm
BJ.S. University of Arizona, 1981
|
B|.S. University of Arizona, 1984
'i
i
i1 A thesis submitted to the
i
Facility of the Graduate School of the
ii
University of Colorado at Denver
in partial fulfillment
I
of jthe requirements for the degree of
] Master of Arts
:j Sociology
1992
l


Th
>1
is thesis for the Master of Arts
degree by
Melissa Jean Holm
has been approved for the
Department of
Sociology
by
s/i b2.
Date


Holm, Melissa Jean (M.A., Sociology)
The Effects; of the Status Characteristic of Sex on
Organizational Choice
Thesis directed by Assistant Professor Cruz C.
Torres
This research examined the extent to which the
; ;l
status characteristic of sex affected the behavior
of subjects] in an organizational experiment. A 2 x
i
2x2 incomplete factorial design was employed to
test the effect, if any, the sex of the supervisor
I
,|
had on the pourse of action chosen by a subordinate
when in conflict with the supervisor. Also varied
was the sex: of the subordinate and historical
context of the situation. Same-sex and mixed-sex
I
dyads were explored.
i
i
Subjects were asked to place themselves in the
1
I ,
position of. the subordinate and respond to vignettes
l
posing a scenario involving a dispute between the
subordinate, and the supervisor over the
subordinate's performance appraisal. The subject
was then asked to rank three strategies for course
of action. Subjects were also instructed to rate
the likelihood of employing each of the strategies


and the effectiveness of each strategy in resolving
the dispute!.
J
I
The da^ba revealed no statistically significant
differences in course of action chosen based on the
sex of the supervisor, the sex of the subordinate or
!
the historical context of the situation.
Explanations of these findings from status
ji
characteristics paradigms are offered. Implications
of these findings for women in management positions
and organizations interested in promoting women
within their ranks are discussed.
The form and content of this abstract are approved.
I recommend!its publication.


To Cruz for her encouragement and guidance:
and to my buddies at work for all their
help and support.


I
CONTENTS
Tables:
CHAPTER
I. INTRODUCTION
II. REVI
W OF THE LITERATURE . .
Introduction .................
Status Characteristics Theory
Scope Conditions .........
Basic Assumptions ....
Sex
Characteristics
Management Characteristics
i Sex' of the Supervisor .
Sex!of the Subordinate
' j
History of Conflict .
Sex
and
of Subordinate
Sex of Supervisor
Conclusions
III. METHODOLOGY
Introduction .........................
Research Methods ......................
:Research Method Limitations . .
i
Research Method Advantages . .
Research Setting ......................
Research Objectives ...................
ix
1
11
11
12
16
18
22
27
30
33
36
37
38
40
40
41
41
43
43
44
vi


Method of Study
and Data Collection
J
................46
Operationalization of Variables ... 49
Propositions .......................... 51
Summary.................................52
i
IV. RESULTS.....................................54
Introduction ........................... 54
Sam]pie Demographics.....................55
Sex' of the Supervisor...................59
Sex: of the Subordinate ...............65
, i
History of Conflict......................69
Sex
and
of Subordinate
Sex of Supervisor..................72
Summary...................................73
V. CONCLUSIONS...................................75
Introduction ............................. 75
Sex of the Supervisor......................76
Sex of the Subordinate.....................79
i
j
History of Conflict........................81
Contributions ............................ 83
Recommendations .......................... 84
i
j
Conclusions................................86
vii


APPENDIX
A. LETTER OF INTRODUCTION
AND CONSENT FORM . .
i
B. DEMOGRAPHIC INFORMATION
I
i
C. INTRODUCTION MESSAGE .
|j
D. VIGNETTES ...............
BIBLIOGRAPHY ...................
i
88
91
92
93
107
i
,i
viii


TABLES
Tables <; |
1 Demographic Sample Information
2 Tenure with the Organization .
3 Level of|| Employment Within
the Organization .............
4 Preference for Strategy by
Sex of Supervisor ............
5 Strategy;for Course of Action
6 Strategy by Sex of Supervisor
7 Likelihood of Employing Each
Strategy by Sex of Supervisor
8 Mean Rating for Likelihood
of Employing Strategy ....
9 Effectiveness of Each Strategy
by Sex of Supervisor .........
10 Mean Rating for Effectiveness
of Each Strategy .............
1 j
11 Preference for Strategy by
Sex of Subordinate ...........
12 Strategy,by Sex of Subordinate
13 Likelihood of Employing
Strategy)by Sex of Subordinate
14 Effectiveness of Each Strategy
by Sex of Subordinate ....
15 Preferences for Strategy by
History of Conflict ..........
16 Strategy |by History of Conflict
57
58
58
59
60
61
62
63
64
64
65
66
67
68
69
70


17 Likelihojod of Employing
Strategy] by History of Conflict
71
18 Effectiveness of Each Strategy
by History of Conflict ....
72
19 Strategy by Interaction Between
Sex of Subordinate and Sex of Supervisor
. 73
x


CHAPTER I
INTRODUCTION
Rosabeth Moss Kanter (1977,3) states in her
! Il
classic; work, Men and Women of the Corporation, that
"if jobs 'create' people, then the corporation is
the quintessential contemporary people-producer".
If modern organizations create people, what happens
to individuals within corporations has a profound
effect not only on the individuals directly involved
I-
i
but also on" the whole social structure.
In the past four decades, the labor force
participation of women in the United States has
risen to over 50 percent (Sidel 1986) Women are
i
expanding their work horizons although during this
century, women have mainly concentrated their labor
force participation in the clerical and helping
professionsi(Sidel 1986). Despite female domination
in specific professions and female emergence in most
others, males still govern the organizations
surrounding-most professions. Women have
traditionally occupied the bottom rung of the
organizational ladder. Although this pattern is
still pervasive, a plethora of legislation enacted
in the 1970'
s catapulted many women into management


positions within organizations (Kanter 1977).
Understandably, these dramatic changes in labor
l
force composition have altered the dynamics of most
organizations. The ramifications of female entry
into and advancement within organizations has become
! il
i
a favored topic of research. Although organizations
range in size and complexity from large and complex
in nature to small and specialized, most of the
interaction! within these organizations still takes
place in small groups.
The importance of small groups to organizations
should not be underestimated. Most major
organizational decisions take place in small groups
as does much of the productive work. Small groups
also provide the conduit through which most of the
formal and informal organizational communication
flows. Individual task performance is dependent
I
upon the information and decisions provided by small
groups. As 'a result, small group interaction has a
profound effect on the ability of individuals to
accomplish performance tasks. Cultivating
i
individual task performance is essential to
attaining the collective goals of any organization.
,1


Understanding the effects that individual
differences among group members has on the
I
interactionj process is tantamount to enhancing
performance. Utilizing the contribution of all
group members is not only equitable to the
individuals] involved but is also expeditious to the
!!
organization.
Volumes of research on small groups have been
conducted. ,j Of the nearly 1400 studies produced
1
between 1927 and 1958, only 10 studies mentioned sex
']
of the group members as relevant to the topic (Dion
1985). When sex was addressed in the research, the
i. # ,
prevalent view of sex differences observed in small
groups was explained by the sex role differentiation
perspective!(Bales and Slater 1955).
:j
Differentiation of the sex roles was rooted in the
1
nuclear family. In this perspective males were seen
,i
as instrumental and task oriented and females were
socially oriented (Bales and Slater 1955). The task
|
specialist was responsible for leading the group and
ij
accomplishing the task while the social-emotional
j
specialist was responsible for group cohesion
1
(Lockheed and Hall 1976; Ridgeway 1982). These
roles were "thought necessary for the stability of
any small group.
3


gender role
Evolving out of the sex role differentiation
perspective (Bales and Slater 1955) came the gender
1 I
role socialization perspective (Wagner 1988).
! 4
Although the premise of this perspective is also
that gender^; roles are gleaned through socialization,
socialization perspective expanded the
range of processes where stereotypes for male and
1 ,1
female roles were learned. The repertoire of
1
behaviors for gender roles continued to be learned
in schools and places of work. In the gender role
socialization perspective, the stereotypes
established: for gender roles allowed for stability
[
i
of masculine and feminine behavior patterns to
appear over'time (Wagner 1988).
Empirical research notes males and females do
tend to behave differently in mixed-sex groups
(Lockheed and Hall 1976; Fennell et al. 1978). The
problem with the sex role differentiation
perspective;and the gender role socialization
ij
'i
perspective I is that the differences noted in the
empirical research of male and female behavior
varied across situations (Wagner 1988). Neither of
these perspectives is able to explain behavioral
variation satisfactorily. As a result,
socialisation perspectives have been challenged by
4


expectation
Zelditch 19
states theory (Berger, Cohen and
)72; Berger et al. 1977; Webster and
: (
Driskell 1978).
i
One brjanch of expectations states theory
providing great insight into group processes is
j
status characteristics theory. The premise of
status characteristics theory is that when
unacquainted individuals form small groups in which
to perform tasks, the ability of group members to
perform the task is evaluated first on the basis of
the external status characteristics the individuals
: i
bring to the group. An external status
I
characteristic is any characteristic on which
individuals: can be differentially rated, such as
sex, race, age or occupation. In the absence of
specific information about the individual's ability
to perform ihe task, status characteristics will be
used by group members to presume task performance.
Group members then interact according to the status
assumptions|formulated by the external status
I
characteristics. In the social psychological
process of status generalization, status
characteristics of individuals affect cognitive and
behavioral outcomes of direct interaction in small
groups (Webster and Driskell 1978; Webster and
5


Foschi 198 8) .
Webster and Foschi (1988) note several
important points about the status generalization
process, in Ifcheir book. Differentiating the ability
of individual group members in terms of task
performance| would not be problematic if the group
had evidence of individual performance on the
I
I
specific task with which to rate all group members
j
(Webster and Foschi 1988). The process is
inequitable)for the individual group members and
'i
inefficient in terms of accomplishing the group task
because,the
contributions of group members are not
weighted according to the worth of the idea but
rather the status of the contributor (Webster and
* j
Foschi 1988). Status generalization is often
utilized by'all members of the group when
: j
formulatingjperformance expectations. Low status
group members process status characteristics
informationijust as do high status members, even
though tlhe process disadvantages them in terms of
interaction : (Webster and Foschi 1988).
I
The status generalization process is not a
conscious thought process. The process is extremely
invidious. ,jUnlike blatant prejudice, the process
operates without group members consciously focusing
6


i
i
animosity tjowards low status members (Webster and
Foschi 1988|!) The final point, states that status
characteristics are peculiar to a particular
'i
I
society, Certain status characteristics may greatly
affect interaction in one society while having no
i
I
relevance in another.
The status generalization process is quite
i, 'i
useful in linking structural sociology, social
psychology, i and behaviorism (Webster and Foschi
I
1988). Status generalization explains the behavior
.1
of individuals in small groups as a social
psychological process. The process is sensitive to
the structural aspect of the situations in which it
occurs.
i
I
Further research is needed in order to
; I
understand if status characteristics theory can be
expanded to I explain group member1s behavior in
establishedjgroups in formal settings. The
explanation|of behavior of individuals in small
groups within organizations has many practical
implications. The ability to maximize the
contributions of all group members is beneficial to
; !'
every organization (Lockheed and Hall 1976; Ridgeway
1982) . '
The purpose of the present study is to examine
7


what effect!, if any, the sex of the supervisor has
on the course of action a subordinate takes when in
conflict with the supervisor. Placing the
supervisor and subordinate in a directly adversarial
position provides the opportunity for greater
. i
understanding of the effects of the status
i
characteristic of sex. With greater understanding
of group dynamics, organizations may be better able
i
to implement policies to aid employees. Greater
opportunities for female supervisors may then
evolve. 1
i
An actual organizational setting was chosen for
the research to lend more credence to the results.
The advantage of this setting was the fact that the
'l
subjects are already labor force participants, thus
have a grealer probability of having had real
.1
experience in handling disputes with superiors. The
; 'i
']
present study attempts to measure behavior rather
than attitudes. Same sex and mixed sex dyads in a
I
corporate setting were examined. Two historical
1 .
contexts were also posed. The first established a
i
history of cooperation; the second established a I
I
history of conflict. The study was conducted in a
corporate organization with greater than 1000
employees.
8


I
The present study was motivated by calls for
further research in the area of status
characteristics (Wagner 1988) An area of
particular interest recommended by many researchers
is in the area of organizational research (Schein
1973; Bartol and Butterfield 1976; Fennell et al.
1978; Stewaft 1988; Martin, 1991).
. i
It is the goal of the study to advance the
state of knowledge in the area of status
'i
characteristics theory. Gender studies and complex
organizations may benefit from this research. The
organizational setting used should provide insight
i
into the interaction of these variables in a real
organization and among a diverse demographic sample.
i
The present study has not only theoretical
implications but also practical implications for
people in organizations. Status characteristics
theory provides a practical guide to intervention in
destructivejgroup patterns (Berger et al. 1972;
!
Freese and Cohen 1973; Berger et al. 1977; Wagner,
Ford and Ford 1986; Wagner 1988).
,1 , ,
Chapter II will provide some insight into the
current literature on status characteristics theory.
ii
Particular attention will be addressed to the
jj
relevant variables in. the study, specifically sex of
9


the supiervisor and sex of the subordinate. Status
generalization theory and a collage of empirical
studies on the effects of sex in the organizational
setting werje used to frame the study.
The methodology used in the study is covered in
Chapter III;. The research was conducted in an
actual organizational setting. In an effort to lend
i
credibility; to the study, an issue of practical
importance to the current employee population was
explored. The issue was a dispute between the
subordinate and the supervisor over the
subordinate's performance appraisal. Subjects'
choices;for course of action taken were recorded
from vignettes for analysis in the next chapter.
, l
Chapter IV provides a thorough analysis of the
data collected. This includes demographic
information related to the sample. An explanation
of the statistical tests used to evaluate the data
will also be included. Finally, the findings from
each of the jhypotheses will be stated.
i-
Chapter V provides a summary of the research.
A discussion of the findings from the study,
conclusions and recommendations for further research
will be presented.
10


CHAPTER II
REVIEW OF LITERATURE
Introduction
During, the past four decades in the United
States femajle labor force participation rose 21.5
percent from 31.8 percent of women in 1947 to 53.3
i
percent in 1984 (Sidel 1986). Unfortunately the
i!
bulk of these female workers have been relegated to
low paying,jlow status jobs. Although the process
has been slow, women have begun to advance within
I, ,
organizations. As they do so, understanding how
subordinates will alter their actions when
: ,i
. S . . ...
responding to women in managerial positions is of
great practical importance.
i
The plight of working women has been of
interest to]many disciplines. Women in management
positions have been the focus of recent studies
i
(Bradley 1989; Kanter 1977; Lunneborg 1990).
1
i
Despite attempts to clarify exactly what factors
determine why women still earn only approximately 65
i
percent of male earnings, a substantial portion of
the variation in the models is still unexplainable
(Cannings 1^91).
This chapter reviews the literature relevant to


i
status characteristics theory. Also included is a
review of the literature pertaining specifically to
the status ^characteristic of sex and literature
developed in the area of management characteristics.
Thus, both the theoretical and empirical literature
: -I
which provides the foundation for each of the
propositions will be examined. There was at the
time of r thej study a scarcity of empirical studies in
the status characteristics theory area concentrating
i
on sex in the work place (Stewart 1988). As a
result,:empirical studies relevant to this study
conducted in other disciplines will be reviewed from
the status Characteristics approach.
Status Characteristics Theory
The development of a more encompassing theory
,, 'l
focusing on the dynamics of small group interaction
began several years ago (Berger et al. 1972).
!' i
Previous theories regarding small group interaction,
' I
specifically the sex role differentiation
perspective^ (Bales and Slater 1955) and the gender
role socialization perspective (Wagner 1988) had
proved inadequate in explaining the results of
empirical work in the area (Eagly 1978; Eagly and
Carli 1981)J Behavior in small groups appears to be
12


I
situational! and positional. The sex role
differentiation perspective and the gender role
socialization perspective are not flexible enough to
account for: these factors (Meeker and Weitzel-
'< 'j
O'Neill 1977; Fennell et al. 1978; Wagner 1988).
Status characteristics theory evolved out of
research on expectation states (Berger et al, 1977).
Status characteristics theory asserts that whenever
unacquainted individuals form groups to accomplish
tasks, group members bring to the group their
external status characteristics. These status
characteristics, such as sex, race and age, all have
acquired meaning from the greater society. These
:i
status characteristics create certain performance
expectations for each of the group members (Berger
et al. 1972). In the absence of additional or
contradictory information, group members tend to
! i
behave according to these performance expectations
rather than;assume nothing about the abilities of
other group members. These assumptions lead to
behavioral inequalities that enhance high status
individuals! opportunities within the group and
simultaneously disadvantage low status group
I ,
members. These behaviors include action
opportunities, performance outputs, reward for l
13
l


I
actions and
the amount of influence group members
possess; In group interaction this is called the
i
power and prestige order (Berger et al. 1972).
Berger
et al. (1977) introduced the idea of the
status-organizing process in which differences in
social and performance attributes becomes the basis
i
i
for stable and observable features of interaction.
'I
The status-organizing process transpires around the
status characteristics of the individuals within the
I
group. The;i idea of a status characteristic is quite
general (Berger et al. 1977). A status
1 i
characteristic is defined as any attribute which
!
could be differentially evaluated, such as age, sex,
race, occupation or reading ability. The status-
ii
organizing process is circular in nature. Status
characteristics evolves out of social interaction,
and once established, subsequent social interaction
is ordered by those characteristics (Berger et al.
1977). Many status characteristics are exported to
!
interactions where the group members were previously
j
unacquainted. The consequence of the emergence of
I
status characteristics is observable power and
'i
prestige order in social interaction (Berger et al.
1977). Status characteristics theory is a theory
about the emergence, maintenance and change process
14


of status characteristics and their effects on
social interaction.
: I
Status1 characteristics theory conceptualizes
inequalitiejs in social interaction as situational
: I
and positional in nature. This situational and
. I
positional approach has three advantages over
transsituatjional perspectives on the subject (Berger
et al. 1977!) The first advantage is the
; ,i
flexibilityj of the theory. Status characteristics
theory is not dependent on any particular type of
interaction,]. Differing status characteristics form
l|
the basis of social interaction in a variety of
situations ranging from jury deliberations to
military task groups (Berger et al. 1972). Another
i
advantage of the theory is the premise that social
operates within the structure of the
situation (Berger et al. 1977). As a result,
individualsipossessing certain status
I. .
characteristics have been known to alter their
; j
behavior depending on the situation and their
!
position. The third advantage of status
!l
characteristic theory is that the theory is
!i
relational and not individual in nature, meaning
both the status of the subject and the object are of
consequence in the interaction (Berger et al. 1977).
interaction
15


\
An example jis that the interaction between an adult,
white female and an adult, white male has an
opposite stjatus order from the interaction of the
same white female and a black child. These
advantages have placed status characteristics theory
' I . . .
ahead of other theories m explaining inequalities
in social interaction.
Scope Conditions
Conditions under which status characteristics
'I
theory can be appropriately applied are known as
scope conditions. Several scope conditions must be
! .1
defined in Order for status characteristics theory
to be utilized effectively.
The first condition required for status
: |
characteristics theory to be applicable is that
I
group members differ in at least one status
characteristic. Two types of status
characteristics, diffuse and specific, have been
identified (Berger et al. 1972; Berger et al. 1977;
'i
Webster and.j Foschi 1988.) .
Diffuse status characteristics have also been
j
called ascribed characteristics in the literature
(Unger 1976)j Examples of diffuse status
characteristics were age, sex, and race. A diffuse
.1
16
I


status characteristic, such as sex, is a
characteristic for which society has defined
differentially evaluated positions for general
perforxnance| expectations (Webster and Foschi 1988) .
i g
These performance expectations have been generalized
,1
I
to influence expectations about the behavior of all
individualsj possessing the particular diffuse status
characteristic. An example of such a generalization
is the assumption that all females are bad at math.
!
Specific status characteristics have been based
i
on performance variables and implies limited
information about an individual (Unger 1976).
Specific status characteristics are also known as
J
achieved characteristics (Unger 1976). Examples of
specific status characteristics are occupation,
i
reading;ability, and income level. Each specific
status characteristic has some general expectations
associated.; These expectations apply to a variety
of specificjtasks.
,j
The individuals in the small group must be task
:|
oriented orjinterested in achieving a performance
goal (Berger et al. 1972; Freese and Cohen 1973;
Berger et al. 1977). Also, the individuals must be
I
collectively oriented, meaning they believe it is
necessary and legitimate to include the group in
17


accomplishing the task (Berger et al. 1972; Freese
!
and Cohen lj973; Berger et al. 1977).
Other ^conditions identified for application of
i
the theory,| included no formal group structure and
;i
no external! evaluation of group processes (Meeker
'i
1981. Thesp conditions obviously limit the range of
situations jin which the theory applies.
j
In the] present study, the status characteristic
of relevancy was sex. Implicit in the
organizational setting were the conditions that the
supervisor and the subordinate were involved in a
task-oriented, collectively-oriented work situation.
, j!
The scope conditions involving informal groups are
; !
not present! in this situation.
Basic Assumptions
i
1
Several basic assumptions are incorporated in
status characteristics theory. The basic
1
assumptions!of this theory include salience, burden
j
of proof, sequencing of connections, aggregating
expectations, and basic performance expectations.
i
The first assumption states that status
i
characteristics are used to discriminate among group
membersi(Berger et al. 1972; Berger et al. 1977;
Webster and
Driskell 1978; Webster and Foschi 1988).
18


If a status characteristic is activated in a
specific situation that characteristic is said to
have bebomej salient and is used to judge the ability
of other group members (Berger et al. 1972; Freese
i
and Cohen li973; Berger et al. 1977; Webster and
'i
Driskell 1978; Webster and Foschi 1988). Once a
status charjacteristic becomes salient or activated
within a situation, the status characteristic is
assumed: to be an important indicator of performance
whether or not the characteristic was generally
believed to! be relevant to the task.
The second assumption, the burden of proof
assumption, means the status characteristic is
considered even if it has no relevance to the task
; |
at hand! Group members use all of the information
'I
they have available when forming performance
expectations about other members whether or not the
characteristic is relevant to performance. Thus,
even if;the characteristics have no relevance to the
j !i
, i- 'I ...
situation at hand, the characteristics will be
utilized if; they are salient. The burden of proof
i
is on demonstrating that the characteristics have no
relevance rather than relevance (Berger et al. 1972;
Berger et al. 1977; Webster and Driskell 1978;
Webster and
Foschi 1988).
19


Third,, status characteristics theory assumes
the concept} of sequencing of connections. This
process occjurs when one individual encounters a
competent individual in another status group.
Through; interaction with this competent individual,
the performance expectations for that individual
changes: the) expectation set for the their status
group (Websjter and Foschi 1988) This assumption
i
complicates the situation when the task involved or
the individuals involved do not remain constant
I
(Webster anci Foschi 1988) Expectations about a
!i
status characteriStic can be altered over time by
the performance of others within that status group.
The fourth assumption, aggregating
i '!
expectations, assumes that all of the status
information^ available to group members will be
utilized when forming performance expectations. No
status information will be arbitrarily excluded when
formulating! performance expectations (Berger et al.
1972; Berger et al. 1977; Webster and Driskell 1978;
Webster and
Foschi 1988). If an individual is
female, black and a medical doctor, all of the
li
differing status characteristics will be included in
the overall status the individual achieves in the
group.
20


Finally, the basic expectation assumption,
;j
concludes tjhe process. Once aggregated expectations
for all; group members are formulated, a power and
prestige order is formulated. This power and
prestige or^er then guides the interaction behaviors
of all group members.
s
Social;; interaction affects the underlying
1
structure of expectations for future performance
(Berger et al. 1972; Freese and Cohen 1973; Berger
et al. 1977; Webster and Driskell 1978; Webster and
1 i
Foschi 1988). Once these expectations are formed
from the status characteristics of group members,
i
I
the subsequent task-related interaction has been
determined.
When a status characteristic has been
activated in a group, the observable power and
prestige order is established (Berger et al. 1972).
Status characteristics are assumed to be activated
when there are individuals with differing statuses
within the group. The order involves four kinds of
!
behaviors including: more action opportunities,
j
more likely;to initiate performance outputs, more
ll
reward actions, and less likely the individual will
|i
accept influence from other group members (Berger et
al. 1972; Berger et al. 1977; Webster and Driskell
1978).
21


Once the power and prestige order of the group
i
is established all group members are either
advantaged jor disadvantaged according to the status
characteristic with regards to interaction in the
task situation.
!
The diffuse status characteristic of interest
to the presjent study is sex. In the United States,
the sex of an individual has many connotations in
different situations. These connotations will be
explored ini an organizational setting in the next
section. i
Sex Characteristics
i1' !
Even though sex is one of the most salient
1
diffuse status characteristics, early research
devoted to the dynamics of small groups rarely
focused on sex differences of the group members.
i
When sex differences were examined and noted, they
were generally attributed to the inherent biological
roles of the sexes or functional role differences
(Bales and Slater 1955). The theory of functional
role differentiation was popular among American
sociologists during the middle of this century when
functionalist theory was a dominant force in the
discipline ([Dion 1985) .
22


As, empirical research in the area of gender
differenced mounted, the biological and functional
role differentiation explanations for gender
differences! became increasingly inadequate in
I
explaining research findings (Wagner 1988).
Differencesj noted among male and female subjects
indicated a| far more situational orientation.
.1
Gender role! socialization theory states males and
females arei socialized into particular stereotypical
behaviors. | Although this theory includes a wider
. , 'i , ,
variety:of ways in which gender stereotypes are
learnedi than the biological and sex role
I
differentiation explanations, it fails to
incorporate;j the transsituational and dispositional
aspects of gender differences (Wagner 1988).
;l
Ini an attempt to explain the varying results
found in many empirical studies (Eagly 1978), a more
comprehensive theory has evolved. Status
i
characteristics theory provides this explanation
(Berger.et al. 1977).
Males and females do behave differently in
mixed-sex groups (Lockheed and Hall 1976). This
fact is harclly surprising in light of the fact that
i|
status does'j organize social interaction (Berger et
al. 1972). !j In this society, males have a higher


status thari females (Lockheed and Hall 1976; Wiley
and Eskilsojn 1976; Unger 1976; Meeker and Weitzel-
0'Neill 1977; Ridgeway and Jacobson 1977; Fennell et
al. 1978; Webster and Driskell 1978; Eagly and Wood
1982). Men) are expected to perform better than
women (Lockheed and Hall 1976). With the higher
status it i's understandable that males would be more
i
task orientbd and more directive in small groups
than would females (Berger et al. 1972; Fennell et
al. 1978; Webster and Driskell 1978; Ridgeway 1982).
;i
Not surprisingly, females who do not conform to
the power and prestige order in the group violate
both the norm of social structure and the group norm
(Wahrmah and Pugh 1974). Even when these females
had a valuable contribution to make to the group,
they had lijttle influence in the task outcome.
Meekerj and Weitzel-0'Neill (1977) introduced
another facet to status characteristics theory, a
i
processcalled legitimacy. This portion of the
theory states that contributions to the group task
raise the status of the individual making the
contribution. Raising an individual's status
relative tojother group members is seen as
appropriatejfor high status individuals (males) but
not for,lowJstatus contributors (females). In order


for low status individuals to have their
I
I
1
contributions to the group accepted, low status
contributors must demonstrate that they have the
'l
best interests of the group in mind.
The contribution to the group by a low status
contributor may be viewed in one of two ways. One
way the contribution may be evaluated is as an
j
attempt by the low status group member to compete
for higher jlstatus within the group. If the
,i
contribution is viewed in this light, the
i
contribution is likely to be shunned by the group as
illegitimat|e.
The other way the contribution may be viewed is
I #
as a way to,; help the group accomplish their task. A
contribution to the goal accompanied by a positive
i
attitude abjout contributing to the group may result
i|
in the contribution being evaluated as legitimate.
As a result], low status individuals seek to have
their contributions to the group accepted by
demonstrating positive social behavior. This should
j
help to have their contributions seen as aiding the
ii
; i
group goals! rather than enhancing their competitive
advantage within the group.
Females are noted for demonstrating larger
i
amounts of positive social behavior in group
25


interaction (Ridgeway 1982; Ridgeway and Berger
1988). Legitimacy has been used to explain the
j
difference^ between male and female behavior in
small groupjs. Rather than female behavior in small
j
group interaction demonstrating a female trait, the
behavior isj illustrative of the behavior of low
j .
status individuals.
i
Behavior in same-sex dyads has also been
'i
studied. Rjidgeway (1988) added to status
characterisjtics theory a situational activation
assumption.1 This assumption states that a diffuse
status char
homogenous
acteristic may be activated for a
group (i.e. all female members) when the
ii
surrounding organizational structure is of a state
opposite thjat of the group. This of course explains
why the stajtus characteristic of sex would be
activated in all female groups. The ramifications
of this! assumption means that even though
j
interacting! m all female groups, the group members
are cognizant of the fact all occupy a low status.
With a
greater understanding of how gender
orders social interaction in task groups, it is time
i
i;
to look at yomen in managerial positions. The
J
following sjection will provide some of the research
.1
on managerial characteristics and on women in
26


I
I .
managerial jpositions.
I Management Characteristics
A study of industrial organizations concluded
that 87- percent of the corporations polled had fewer
than 5 perqent women in management (Schein 1973).
Although majnagers are no longer always men, women in
management [appeared to violate social norms in
i
supervisory positions (Rosen and Jerdee 1973;
i
Eskilson and Wiley 1976). Men generally hold more
prestigiousj, higher paying occupations, for
instance, managerial positions (Eagly and Wood 1982;
I
Wiley and Eskilson 1983). Unger (1976) noted that
legitimate power based on organizational authority
or contractual relationship is characterized as
i
male. Several studies examined the relationship
between sex role stereotypes and ascribed middle
level management characteristics (Day and Stogdill
1972; Scheib 1973, 1975; Cann and Siegfried 1987;
Cann and Siegfried 1990).
Scheirij (1975) selected female middle managers
employed ini the insurance industry as subjects.
i
Earlier, Schein (1973) had examined male middle
|
managers ini) a similar study. The purpose of the
'i
study was t;!o rate characteristics of men in general,
27


women in general, and successful middle managers on
a descriptive index containing 92 items. Subjects
j
were asked jto rank each item as either being
characteristic or not characteristic of three
J
categories.;; The categories included males, females
and successful middle managers.
As was' hypothesized, men in general and
i i
successful middle managers were perceived to possess
more characteristics in common than were women in
general andj successful middle managers. Schein
'I
(1973, 1975) concluded from this research that for
< i
I
women tp achieve success in these positions it may
be necessary for them to display stereotypical male
characteristics. Using status characteristics
theory, thijs conclusion may be valid if one assumes
!!
that task competence is a male characteristic. In
general, a ifemale with high task performance will
change the expectations other group members hold
: 1
with regard to gender. Findings of a statistically
significant| relationship between women in general
and successful middle managers lends further support
;
to the theojry.
AnothJr interesting finding of the studies was
that length of service of the subjects also
significantly impacted the results (Schein 1973,
28


'I
1975) Forj managers with greater than 5 years of
experience,] the relationship between female
i
characteristics and successful middle management
characteristics was statistically significant.
I
Among the managers with less than 5 years of
experience,j this relationship was not statistically
significant*. This finding seems to bare out the
I
theory that more exposure to task competent
individuals' within a group, in this case females,
i
does alter |the aggregate expectation for that group.
When comparing the results of the all female
managers st;udy (Schein 1975) with the previous study
j
of all male managers (Schein 1973), significant
differences! among the subjects did arise. While
i
there was a statistically significant relationship
between womien in general and successful middle
'i
managers in the female sample, the relationship was
not significant in the male sample.
The ddsign of the present study allowed testing
the interaction between individuals working in small
groups at a collective task. Empirical studies
focusing oh various aspects of the issue of gender
differencesj in the work place have been appearing
for some time. Several of these studies have
focused on jthe effect that sex of the group members
29


has on the dynamics of small groups. Although these
studies1 did not emphasize status characteristics
theory, the results of these studies illustrate
portions of the theory. The next sections examine
the three main hypotheses for the study.
Sex of the Supervisor
The fjjirst proposition addressed in the study
ii
sought to determine the effect, if any, sex of the
supervisor
ihad on the course of action chosen by the
J
subordinate when in conflict with the supervisor.
The proposition states:
I
I. Subordinates are more likely to circumvent
their supervisor's conflicting decision if the
supervisor is female.
Utilizing the status characteristics theory
developed previously in this chapter, the
expectation was that females in the group have lower
status tharj males. Also, from the previous
empirical literature, managerial traits are thought
to be particularly masculine in nature. Therefore
women in management positions violate the norm.
j :
Several informative studies have concentrated
:i
on the effectiveness of managers, and whether
evaluation
of their effectiveness varies based on
30


the sex of ithe supervisor. Research on task groups
has concluded sex is salient initially as a status
characteristic but could be overridden when group
performance evaluations were positive (Seifert and
I
Miller 1988j|) Seifert and Miller (1988) examined
whether; subordinates react to the method of leader
selection ojr sex of leader when group performance
was evaluated. The study design varied the sex of
the leader,;
the sex of the subordinate, and
performance feedback.
Surpri
singly when evaluating performance, there
was a tendency for both males and females to rate
the leader
more favorably when the leader was of the
opposite sex. A less direct measure of performance
was tested
Miller 1988
in the evaluation of notes (Seifert and
). Unlike the direct measurement, in
this category both males and females did tend to
evaluate ma!le leaders relatively more positively
than female leaders. Although the direct measure of
performance may be indicative of an unwillingness to
j '
blatantly discriminate against females, examination
of the indirect measure indicates a bias may still
l|
exist. Stajtus characteristics theory would expect
subordinates of both sexes to evaluate the high
status member more positively. Subordinates were
31


I
willing to jwork with the leader again if the task
performance! was successful regardless of the sex of
i
the leader.j Apparently the successful performance
outcome' raised the aggregate status of even the low
; i
: -i
status supervisors.
Rosen jand Jerdee (1973) studied the effect of
sex-role stereotypes on evaluations of male and
j
female supervisory behavior. Their conclusion was
that subordinates evaluation of their supervisors
i
effectiveness did not depend on the sex of the
supervisor.1
i
The ddsign of this study (Rosen and Jerdee
1973), varied the sex of the supervisor, the sex of
the subordinates, the subjects sex and the subjects
']
occupational status. Once again, in this study,
subjects evaluated the supervisors effectiveness.
There jwas no statistically significant
; i
difference jin the evaluation of effectiveness
between: the male and female supervisors (Rosen and
i
Jerdee 1973j. However, there were statistically
significants differences between the managerial
I
styles deemed appropriate for male and female
managers, jpemale managers once again were expected
to exhibit (higher levels of socioemotional behavior
i
in order toj legitimize their task performance.
32


In;a similar study on leadership style (Kushell
and Newton
altered by
. i
1986), subordinate satisfaction was
leadership style. Subordinates are most
satisfied wjith a democratic leadership style. The
sex of the (supervisor had no statistically
significant effect on subordinate satisfaction. As
was demonstrated in the Rosen and Jerdee (1973)
; i
study, female supervisors are evaluated more
negatively If their leadership style is deemed
inappropriate. The legitimacy process continued to
be of paramount importance to low status managers.
i
Another study that evaluated the effects of
ii
sex-role stereotyping on the evaluations of male and
female supervisory behavior was conducted by Bartol
and Butterfield (1976) This study arrived at a
conclusion isimilar to the Rosen and Jerdee study
;i
;l
(1973).: The behavior of female managers was
I
considered 'legitimate if their leadership style
incorpofatdd high degrees of socioemotional
i
behavior.
| Sex of the Subordinate
The purpose of the second proposition addressed
, j
in the study was to determine the effect, if any,
the sex of !the subordinate has on the course of
33


!j
}
i,
i I
i> j
i
action Chosjen by the subordinate when in conflict
with the supervisor. The proposition states:
: 1
Hi Male subordinates are more likely to
i
circumvent [their supervisors conflicting decisions
than femalej subordinates.
Status1 characteristics theory predicts that the
I
sex of the jsubordinate will not alter the behavior
of the individuals with respect to the situation.
!
Individuals react similarly to the external status
characteristics of other group members regardless of
their own status within the group. "Since everyone
knows the repertoires associated with status
'i
superiority and status inferiority, those with a
i
status advantage and those with a status
disadvantage will adopt the appropriate repertoires
i
of attitudes and behaviors, regardless of their sex"
,1
(Wagner 1988,62).
. I
In the Seifert and Miller (1988) study of
i
subordinates' perceptions of leaders, subjects'
responses were mixed. In the successful scenario,
male subjects tended to give themselves more credit
for success than did females subjects (Seifert and
Miller 1988). In the unsuccessful scenario, equal
responsibility was attributed to the leader and the
subject. In the successful scenario, female
I


subjects exhibited a higher amount of socioemotional
!
behavior in; order to have their task contribution
!i
recognized tas legitimate. The higher status male
subjects hdd no need to demonstrate this behavior.
i
When Evaluating performance, both male and
:i
female subjects evaluated themselves and their
leaders; more favorably when the task performance
evaluation was positive (Seifert and Miller 1988).
1
In a study on leadership style and subordinate
'I
satisfaction (Kushell and Newton 1986), the sex of
the subordinate was significant in the satisfaction
of the subordinate in the autocratic leadership.
; i
Among female subjects, autocratic leaders were
perceived more negatively then democratic leaders.
i
Male subjects seemed to presume the autocratic
i
leadership istyle to be expected in a leader.
. i
In anqther interesting study on supervisory
i 1
styles, Rosen and Jerdee (1973) found that in the
5
case of the friendly-dependent approach the sex of
the subordinate was significant when the supervisor
was of the 'opposite sex.
! I
Bartol and Butterfield (1976) studied attitudes
about leadership styles and sex of the manager among
j
male and female subjects. Subjects were asked to
i i
evaluate four leadership categories on each of eight
35



items of interest. The four categories included
; -I
initiating structure, production emphasis,
consideration and tolerance for freedom.
I
In thd initiating structure category,
j
regardless |of the sex of the manager, female
I
subjects evaluated the style more favorably than
;l
male subjects. The consideration category was
;i
evaluated favorably for female managers especially
i
when the subject was female. Female subjects rated
i
male managers lower in the tolerance for freedom
1
category than did male subjects.
I History of Conflict
'.i
The third proposition dealt sought to discover
1 ;i
the effect,, if any, the historical context of the
1 1
relationship between the supervisor and the
i
subordinate would have on the course of action
chosen |by tjhe subordinate when in conflict with the
;l
supervisor.. The proposition states:
; !i
III. Subordinates are more likely to
circumvent
their supervisors conflicting decisions
when there;| is a history of conflict between the
i
supervisor |and the subordinate.
t; ,i
Status characteristics theory generally
1
i,
specifies that the individuals in the task group
i
36


were previously unacquainted (Webster and Driskell
l I
1978). ; Despite this assumption, the original
' i
formulation of status characteristics theory merely
h
!
specified that the theory worked best when the group
'!
did not,; have a long history of interaction (Berger
;k i
i
et al. 1972). Realistically, it is at best rare
I
that individuals form groups in which they will only
' :i
interact orice. Most of the groups within which
1 i
individuals participate in tasks are ongoing in
' I
nature. As a result, the history of interaction
1
between thd individuals in groups has great
practical importance.
I
; I
Sex of Subordinate and Sex of Supervisor
i
The research in the area of interaction between
different sex individuals has produced conflicting
,1
predictions. In a study examining the effects of
i
the status (characteristics of gender in pay
l
distribution (Stewart 1988), males tended to be more
; ii
sensitive to the characteristic than did females.
j
Rosen and Jerdee (1973) found that subjects reacted
more favorably to particular management style if the
I
supervisor[was of the opposite sex. Other studies
j
have reached varying conclusions about the
i
interaction between sex of subordinate and sex of
> i


supervisor l(Kushell and Newton 1986; Seifert and
! !
Miller 19881) In the Seifert and Miller study
'i
(1988), subjects tended to rate leaders of the same
sex more positively than opposite sex leaders.
, |i
As: a result of the mixed predictions on the
interaction between sex of the supervisor and sex of
1
the subordinate, the present study examined the
i
situation most likely to produce the most dramatic
:: '!
result. This situation involves a high diffuse
! I
status individual in a low status position in
,1
conflict with an individual with low diffuse status
in a high status position.
Conclusions
Status characteristics theory has provided a
I
more flexible approach to the interaction in small
i
groups. Tiiis approach has provided a situational
: i
and positional aspect to the study of group
i
interaction (Wagner 1988). Explanations of group
; I
behavior is no longer tied to the personality of
individuals or the socialization of all people.
J
The empirical research on sex differences among
I
supervisors and subordinates in organizational
j|
settings has produced mixed results. Empirical
\ :!
research examining the effects of sex of the


subordinate included one study finding subordinates
react more
favorably to supervisors of the opposite
sex, two studies found no relationship based on sex
'i
of the supervisor and one study concluded the
acceptance
management
of female managers is dependent on the
style of the supervisor. Empirical
research examining the effects of sex of the
subordinate! included two studies finding no
1
relationship based on sex of the subordinate, one
ii
study concluding subordinates react more favorably
,i
to a supervisor of the opposite sex and one study
.1
finding the acceptance of the supervisor is effected
by the success of the task. These results represent
the varied
situations and positions group members
find themselves in within organizations.
39


CHAPTER III
METHODOLOGY
Introduction
The study focused on the effects the sex of the
supervisor thad on course of action chosen by
i
subordinates when in conflict with their supervisor.
To test the status characteristic of sex the study
was conducted in an organizational setting. The
variables under study included sex of the
supervisor;! sex of the subordinate and historical
context. Control variables included the demographic
information collected from subjects. A 2x2x2
incomplete ||factorial design was utilized for data
'I
analysis. |The independent variables under study
i
;l
were sex of supervisor, sex of subordinate and
history of conflict. Four dyads were analyzed:
1
male supervisor and male subordinate, (2) male
supervisor
and female subordinate, (3) female
(1)
supervisor;and male subordinate and, (4) female
supervisor]and female subordinate.
This chapter will describe the method and
I
procedures,) followed in conducting the study. The
!|
measurement instrument will be discussed. The


chapter wilil conclude with reiterating the research
hypotheses under study.
Research Method
In designing the study for the effects of the
1
status characteristic of sex, all of the research
'l
methods; were considered. Upon review of the
literature and theory, an experimental research
design using vignettes was deemed most appropriate.
j
The study was a field experiment in a corporation
with greater than 1000 employees. This research
method was ;jchosen because of its ability to
,1
establish a cause and effect relationship between
sex of the supervisor and course of action chosen
(SingletonJet al. 1988). Due to the nature of the
, ] ,
field experiment, the ability to control for
extraneous
variables through random treatment was
also an important factor. Although it is far more
difficult tlo control field experimentation than it
ii
is to control laboratory experimentation, the
limitations are outweighed by the advantages.
Research Method Limitations
'j
The experimental method used in the present
41


study does ihave several weaknesses. A major
, 1
weakness in experimental research is external
: J
validity. jThe research was conducted in an
t' 1;
organizational setting in one specific location.
Care should be taken in generalizing the results.
* i
The us^e of vignettes in experimental design
i
' 1
minimiz.es experimental realism. In the present
i,
study, an actual organizational setting was provided
i
in an effort to offset the effects of the vignettes
on experimental realism.
i
There jis always the danger that the subjects
will react ,to some subtle cues the experimenter is
j
giving inadvertently. In order to mitigate the
demand characteristics in the experimental
situation, [the experimenter was unaware of which
treatment any particular subject was assigned. Also
i.
there is evidence (Singleton et al. 1988) the
subjects alter their behavior simply because they
. j
realize, they are being evaluated. There is also the
conceivable difference between the volunteers that
. . 'i
participate in the experiment and those that chose
ij
not to participate (Singleton et al. 1988).
42


I
Research Method Advantages

A major advantage of using field
experimentation is the ability to study the behavior
I
of individuals in an actual setting. Inherent in
laboratory experimentation is that the sample of
J
subjects is1from a small, restricted population
(Singleton et al. 1988). The present study was able
to incorporate a demographically diverse sample.
Another major advantage of the experimental
research method is the assumption of a cause and
effect relationship between sex of the supervisor
and course of action chosen. The experimental
'I
method attempts to control for extraneous variables.
!l
'I
Control of extraneous variables is achieved by
random assignment of the treatments.
!
The vignettes utilized in the study depicted a
dispute between a subordinate and a supervisor over
the subordinates' performance appraisal. This type
of dispute is common in organizations. Thus, the
1 i
study provided an experimental setting high in
mundane'realism.
All of,
Research Setting
the subjects under study worked for a
43
.1
I
:i


corporation!with greater than 1000 employees. This
corporation jwas highly formalized and hierarchical
in structure. Although all of the subjects who
volunteered to participate in the study work in the
I
j
corporate office building in one location, the
corporation itself is decentralized with
international ventures.
Most of the subjects gathered in one of the
large conference rooms for the experiment. When
time constraints made this impossible for the
j
participants, the experimenter scheduled specific
arrangements for that individual.
J
Research Objectives
The premise of status characteristics theory is
that the external status characteristics individual
!i
group members bring from society in general to the
group form the basis for interaction within the
group. In order to test behavioral inequalities in
small groupii interaction, the behavior examined was
i |
the course of action chosen by the subordinate. The
status characteristic of interest to the study is
sex. As a result, situations with both male and
' 1
female supervisors and subordinates was designed.
44


In an attempt to portray an issue of relevance and
importance to most individuals, a situation in which
the supervisor and the subordinate were in
disagreement!: over the subordinate's performance
appraisal was described. It appears that effects of
the status characteristic of sex is quite dependent
on the situation in which the individuals are
i
involved. Although the ability to extend status
i
characteristics theory beyond ad hoc groups is
'i
questionable, individuals in organizational groups
i
establish historical relationships. As a result,
I
history of conflict was also added to the situations
1
to give subijj ects an opportunity to respond to
I
mitigating circumstances. One historical condition
was a history of conflict between the supervisor and
the subordinate and the other was no history of
conflict.
i
One objective of the research was to determine
the effect {that sex of the supervisor had on the
i
course of aption chosen by the subordinate when in
conflict with the supervisor. A second objective
was to disc'over if the sex of the subordinate had an
effect on tlhe course of action the subordinate
I
45
,!
i
i


chose. The final objective of the present study was
to examine if historical context of the situation
J
between the;! supervisor and subordinate altered the
'i
I
course of action chosen by the subordinate.
i
Subjects were also asked the likelihood of employing
'I
each strategy for course of action and the
j
effectiveness of each of the courses of action.
Method of Study and Data Collection
I
Procedures used to accomplish the research
i
J
objectives included sampling, data collection, and
data analyses. After the research proposal had been
]
formulated,| the Human Resources Department in the
corporation! under study was contacted. Upon
receiving Corporate approval for conducting the
research ini the corporate office building, the
ij
proposal wa!s sent to the Human Research Committee at
i .
the University of Colorado at Denver. The Human
Research Committee also approved the study.
The proposed study was introduced to
,1
prospective participants through a letters of
introduction distributed to employees via company
mail baskets. If employees returned the signed
' :!
consent form distributed with the letter of


I
introduction, they were then contacted about times
for participation. Participation in the experiment
!
was completely voluntary. Consent forms were
i
, i
collected from all participants. Thirty-three
'I
I!
percent of the employees given consent forms agreed
to participate. Participants were asked to gather
at a specific location or the experimenter met them
at a designated time and location. The letter of
i '
introduction and consent form are contained in
Appendix A.j
When the subjects had congregated at the
experimental location, subjects were first asked to
complete the top page of the instrument which
contained demographic information about themselves
1
(See Appendix B). Included in this was: (1) job
title, whiqh indicated the organizational level, (2)
: j
years of service at this title, (3) years of service
' I
with this company, (4) total years of full-time
employment I of any kind, (5) age and, (6) sex. Upon
completion
of this page, the subjects were then read
a short passage explaining that they were to place
themselves!in the position of the subordinate (see
Appendix C\
The experimenter then asked the
47


; j
subjects to11 read each of the two vignettes they were
given arid respond as though they were the
subordinate in the situations described. The
vignettes utilized in the study are included in
Appendix D. ; The manipulated independent variables
were the sex of the supervisor, the sex of the
subordinate and history of conflict between the
supervisor and the subordinate. Each subject was
randomly assigned one vignette with a male or female
supervisor in one historical context. The subject
was also givjen one of the filler vignettes. A
filler vignette was added to obscure suspicion about
the purpose of the study. The filler vignettes also
j
portrayed an organizational situation in which the
subordinate was unhappy with their job. This
vignette: lerigthened the study and diverted attention
from the! primary vignette.
The subjects completed the vignettes. The
completed vijgnettes were then collected and a
I
document number was assigned to each. A coding
1
I
procedure was developed for input into a computer
flat file.
48


Operationalization of Variables
Four vignettes were used in the study. In each
case the subject was asked to place himself/herself
in the position of the subordinate. The first
vignette posed a conflict with a female supervisor
over a subordinate's appraisal. In this vignette,
the subject] was told the supervisor and the
subordinate had previously had an amiable
relationship for historical context. The second
vignette posed an identical situation with a male
supervisor. In the third vignette, the subordinate
was once again in conflict with a female supervisor
over their appraisal and the subject was informed
that the supervisor and subordinate had a history of
disagreements. The fourth vignette posed a
situation identical to the third vignette with the
exception qf a male supervisor. Thus four vignettes
I
were analyzed in the study. These included: (1)
male supervisor with positive history, (2) male
i
supervisor {with a history of conflicts, (3) female
supervisor with positive history, and (4) female
supervisor with a history of conflicts. Disputes
i
between supervisors and subordinates over
i 49
i
I


performance appraisals depict a common
organizational issue.
In the first part of each of the vignettes,
subjects were asked to rank three strategies for
action. In Strategy A, the subordinate disputes
their appraisal with their supervisor's supervisor.
This constitutes the strongest action from which the
subject has to choose. In the second course of
action, Strategy B, the employee chooses inaction.
By choosing this strategy, the subject admits there
is a disagreement but chooses to do nothing. In
Strategy C,j the subject chooses to write a rebuttal
that will be appended to their appraisal. Although
this strategy does involve action, it is not
directly confrontational with the supervisor. A
ranking of ,one (1) indicated the most likely course
of action. A ranking of three (3) indicated the
least likely course of action. All of the courses
of action are legitimate options every employee has
available tio them.
For each of the strategies presented, the
subjects were next questioned about the likelihood
of employing each strategy. The subjects were asked
50
i


to rate on a five point Likert Scale, how likely
'i
they were to employ each of the strategies with
which they were presented. The scale ranged from a
rating of one (1), which indicated not at all likely
to a rating of five (5) which indicated very likely.
A rating of, three (3) indicated the subject was
undecided about their likelihood of employing the
strategy.
Finally, for each of the strategies presented,
the subjects were asked to rate the effectiveness of
the strategjy in resolving their dispute with their
supervisor. Once again, a five (5) point Likert
Scale was utilized. The scale ranged from a rating
of one (1) which indicated the strategy was not at
all effective to a rating of five (5) which
indicated the strategy was very effective. A rating
of three (3) indicated the subject was undecided
about the effectiveness of the strategy.
Propositions
This study provides an organizational setting
for the analysis of the following propositions:
I. Subordinates are more likely to circumvent
their female supervisor's conflicting decisions
I
51
i


rather than a male supervisors conflicting
decisions.
II. Male subordinates are more likely than
female subordinates to circumvent their supervisors'
conflicting1 decisions.
III. Both male and female subordinates are
more likely to circumvent their supervisor's
conflicting; decisions when there is a history of
conflict between the supervisor and the subordinate
rather than no history of conflict.
Except for hypothesis III, these hypotheses
were derived, from previous research in the area of
the status characteristic of sex. Chapter II
reviews the literature relevant to the topic.
I
, Summary
This chapter provided a brief discussion of why
the experimental method was chosen for the study.
Presented next are the advantages and disadvantages
of field experimentation. A description of the
research setting followed. An organization with
greater thajn 1000 employees was the setting for the
study. The' objective of the study was to test the
j
effects that sex of supervisor, sex of subordinate
52


and history,' of conflict between the supervisor and
the subordinate had on the course of action chosen
by the subordinate in a dispute over the
subordinates performance appraisal. The three
strategies examined in this study were to dispute
the appraisal with the supervisor's supervisor, to
write a rebuttal and to give up. Vignettes were
used to provide an estimate of the variables for
course of action chosen. The chapter concluded with
a synopsis pf the propositions under study.
53


CHAPTER IV
RESULTS
Introduction
The previous chapters offered the relevant
literature, I and the model for testing the effects of
I
the status characteristic of sex in an
.]
organizational setting. The present study examined
the effect that the sex of the supervisor had on the
course of action a subordinate took when in conflict
. | .
with their supervisor. The likelihood of employing
each of the^ three strategies for course of action
1
was asked. Also, asked was the effectiveness of
each strategy at resolving the dispute. Both male
and female subjects were asked to assume the
position of| the subordinate and respond as if they
were in the! situation posed. Two historical
i
contexts were probed: a cooperative history between
the supervisor and subordinate prior to the episode
under evaluation; and a history of conflict between
the supervisor and the subordinate prior to the
episode.
The present study employed a dual approach to
the analysis of the data. A multinomial logit model
was used to identify differing patterns amongst the


i
i
j
i
1
i
strategies for course of action. A chi-square test
was used to'' determine if there were any
statistically significant differences between
combinations of the strategies, likelihood of
employing the strategies and the effectiveness of
the strategies. These tests were chosen because the
data consisted of three (3) independent dichotomous
variables and nine (9) dependent classification
variables. The level of measurement for the
i
I
. 'I
independent. variables was nominal. These variables
J
included sex of subordinate, sex of supervisor and
history of conflict. The samples were independent
I
and random.; The level of measurement for the
i
dependent variables was ordinal. These variables
1
included the ranking for each strategy, a
measurement' for likelihood of employing each
strategy and the effectiveness of each strategy.
One dependent nominal variable included a
classification of the ranking for each of three
1
strategies provided. The alpha level for this study
was set at ;. 05.
1
Sample Demographics
As a rjesult of the present study being
55


conducted in an organization with greater than 1000
employees, the subjects participating in this study
represent ajreasonable cross section of the
population. Demographic variables controlled for in
this study were occupational level, years of service
at this level, tenure with the organization, years
of labor force participation and age. Table 1
illustrates the mean, standard deviation and the
range for the demographic variables.
The occupational level was measured by 5
categories. An occupational level of 1 indicated
i
the subject was a non-management employee. Level 2
was the lowest level of manager, typically the level
for a staff; position. A level 3 was a middle
management position. An occupational level of 4
signified a director position and level 5 was an
executive director.
Far more management personnel participated in
the study than non-management. The average
occupational level of the subjects was 2.4. Average
years of service at this level was 5.42. The
subjects had an average of 14.64 years of tenure
with the organization. The amount of tenure with
the organization ranged from one (1) year to twenty-
six (26) years. The average length of participation
56


Table 1 j
Demographic|Sample Information
Total
Range
Mean SD Min Max
Level 2.4 .9 1 5
Years at Title 5.42 4.32 1.00 19.00
Tenure 14.64 6.07 1.00 26.00
Total Employment 19.64 6.68 1.00 39.00
Age 40.52 6.81 23.00 64.00
Female
1 Range
i Mean SD Min Max
I Level 2.2 .8 1 3
Years at Title 5.35 3.97 1.00 16.00
Tenure ^ 13.81 5.95 1.00 26.00
Total Employment 18.85 6.46 5.00 39.00
Age 39.97 7.26 27.00 64.00
Male
Range
] 1 Mean SD Min Max
Level i 2.7 .9 1 5
Years at Title 5.49 4.69 1.00 19.00
Tenure 15.50 6.11 1.00 26.00
1 Total Employment 20.44 6.86 1.00 35.00
< Age i 41.08 6.33 23.00 54.00
in the labor force was 19.64 years. Average age of
the subjects participating in the study was 40.5
57


years. None of the control variables proved to be
statistically significant.
i
Table 2 illustrates the distribution of tenure
in the organization for the subjects in the study.
Table 3 illustrates the distribution of subjects by
l
level of employment within the organization.
Further scrutiny of the data reveals some
interesting'distributional patterns. The females in
Table 2
Tenure with;the Organization
Tenure Male Female
5 Years and Less 4 6
6 10 Years 11 17
11-20 Years 40 36
20 Years! and More 17 15
the sample tended to be clustered towards the bottom
organizational levels, despite approximately equal
tenure with! the males in the organization.
Table 3 !
Level of Employment Within the Organization
Level Male Female
1 j 6 17
2 24 27
3 36 30
4 1 0
1 5 5 0


! Sex of the Supervisor
The first proposition suggested that sex of
supervisor would have an effect on whether
i
subordinates challenge their supervisor's
performance!appraisal. This proposition was tested
using same-sex and mixed-sex dyads. Each of the
strategies for course of action was evaluated using
a multinomial logit model. This technique relies on
testing the odds ratio of observed frequencies
between the variables. Table 4 contains the cell
frequencies,for each strategy.
Table 4
Preference for Strategy by Sex of Supervisor
Strategy A
Preference 1 2 3
Male 33 23 11
Female 37 33 6
1 Strategy B
Preference 1 2 3
1 Male 5 6 56
1 Female 5 4 67
i Strategy C
Preference 1 2 3
Male 29 38 0
Female 34 39 2
59
I


Analysis of!the data did not support the
proposition. The statistic for preference of
strategy based on the sex of supervisor was (xz =
0.14, p < 0.70).
Table 5 illustrates the responses to the
combinations of strategies for course of action.
Table 5
Strategy for Course of Action
Sub Male Female
Sup Male Female Male Female
History i G B G B G B G B
ABC 0 0 1 0 0 0 0 0
ACB 8 7 12 9 10 9 5 10
BCA 4 6 9 6 7 5 7 9
CBA ' 2 1 1 0 1 2 2 0
BAC 1 0 0 1 0 0 0 1 0
CAB i 1 1 1 1 3 0 1 0
Analysis of| combinations of subjects' choices for
subordinate's course of action when in conflict with
i
a supervisor did not support the proposition. Once
again there1 was no statistically significant
difference tound on the basis of combinations of
i
strategies.1 The study analyzed individually the
i
number of first choice responses for each of the
strategies. The statistic for the strategies by sex
of the supervisor was (**= 0.14, p < 0.93). Table 6
60


illustrates!the number of subjects in each category.
'I
The course of action chosen by subordinates did not
vary based pn the sex of the supervisor.
A cross check of the first proposition
suggested that subordinates would be more likely to
challenge their supervisors conflicting decision
Table 6
Strategy by Sex of Supervisor
1 Male Female
! Dispute Appraisal 34 37
Give Up 1 5 5
\ Write Rebpttal 28 34
based on the sex of the supervisor. Analysis of
4
subjects' choices for likelihood of employing each
strategy when in conflict with their supervisor
i
provided partial support for this hypothesis. A
statistically significant difference was found for
the likelihbod of employing Strategy A. The chi-
I
square statistic for Strategy A was (x* = 6.40, p <
0.04). There was no statistically significant
I
difference found for the likelihood of employing
Strategy B.i The statistic for Strategy B was (x^ =
1.00, p < 0.59). There was insufficient data for
I
analyzing Strategy C. The chi-square test should
I
not by employed when more than twenty percent of the
61


cell frequencies contain less than five, or any cell
frequency is less than one (Blalock 1979; Siegel and
Castellan 1988). Table 7 illustrates the responses
for the likelihood of employing the strategies. As
a result of small numbers of responses in several
categories,; the categories of not at all and not
likely were aggregated into the category of not
likely. Also, the categories of likely and very
Table 7 '
Likelihood of Employing Each Strategy
by Sex of Supervisor
Strategy A Male Female
Not Likely 11 9
Undecided 13 5
Likely 45 63
Strategy B Male Female
Not Likely 46 57
Undecided 11 11
Likely 12 9
Strategy C Male Female
Not Likely 1 9
Undecided 4 5
Likely 64 63
likely were aggregated into the category of likely.
The likelihood of employing specific strategies by
subordinates did vary based on the sex of the
62


supervisor. Table 8 gives a summary of the mean
ratings for the likelihood of employing each of the
strategies., Considering the level of the data, the
i
means for the strategies are used for illustrative
purposes only.
Table 8
Mean Rating for Likelihood of Employing Strategy
Supervisor V [ale Female
Subordinate Male Female Male Female
Strategy A 3.77 3.87 4.07 3.97
Strategy B 2.39 2.42 2.22 2.78
Strategy C 4.65 4.42 4.20 4.31
Another cross check used to examine the first
proposition suggested that subordinates believed it
was more effective to challenge their supervisors
conflicting decisions based on the sex of the
supervisor (See Table 9). Again data analysis
failed to support this prediction. There were no
statistically significant differences found for
Strategies A and C. The statistic for Strategy A
was (x1 = 3.50, p < 0.15). For Strategy C, the
statistic was (x^ 1.70, p < 0.45). There was
insufficient data for analysis of Strategy B.
Subordinates did not believe resolving a conflict
with their supervisor was more effective
63


Table 9
Effectiveness of Each Strategy
by Sex of Supervisor
Strategy A Male Female
Not Effective 20 34
Undecided 25 20
Effective 23 22
1 Strategy B Male Female
Not Effective 65 67
1 Undecided 3 2
Effective 1 6
Strategy C Male Female
Not Effective 32 47
Undecided 18 13
Effective 19 15
based on the sex of the supervisor. The
effectiveness of the course of action chosen by
Table 10 1
Mean Rating'for Effectiveness of Each Strategy
Supervisor V [ale Female
Subordinate Male Female Male Female
1 Strategy A 2.90 3.05 2.80 2.87
Strategy B 1.35 1.45 1.38 1.47
] Strategy C 2.74 2.66 2.18 2.67
subordinates did not vary based on the sex of the
supervisor.! Table 10 gives a summary of the mean
I
64


ratings for the effectiveness of each strategy.
i Sex of the Subordinate
The second proposition indicated that sex of
the subordinate will have an effect on subordinates'
course of action when in conflict with their
supervisor. Again data analysis failed to support
!
this prediction. There was no statistically
significant difference found for the strategies.
The statistic for course of action based on sex of
the subordinate was = 0.01, p < 0.93). Table 11
Table 11 \
Preference for Strategy by Sex of Subordinate
i Strategy A
Preference 1 2 3
Male 37 26 8
1 Female 33 30 9
Strategy B
i Preference 1 2 3
Male 5 5 61
Female 5 5 62
1 Strategy C
Preference 1 2 3
Male 29 40 1
Female 34 37 1
contains the cell frequencies for each strategy.
The statistic for the combinations of strategies by
65


I
sex of the subordinate was (xz= 0.38, p < 0.82).
Table 12 illustrates the number of subjects in each
category.
Table 12
Strategy by,Sex of Subordinate
Male Female
1 Dispute Appraisal 37 34
Give Up 5 5
Write Rebuttal 29 33
Subordinated sex did not alter the action chosen by
subordinates.
As with the first proposition, the second
proposition was also examined by a cross check
i
procedure. This procedure suggested that
subordinates would be more likely to challenge their
supervisors'conflicting decisions based on the sex
of the subordinate. Again data analysis failed to
support this prediction. There were no
statistically significant differences found for the
likelihood of employing any of the strategies. The
I
statistic for Strategy A was (x^ = 0.01, p < 0.99).
For Strategies B and C, respectively, the statistics
i
were (x^= 1.20, p < 0.52) and (xz= 0.20, p < 0.90).
J
Table 13 illustrates the responses for the
j
likelihood of employing the strategies.
66


I
Table 13
Likelihood of Employing Strategy
by Sex of Subordinate
i
Strategy A Male Female
Not Likely 10 10
Undecided 8 10
Likely 54 54
Strategy B Male Female
Not Likely 50 53
Undecidek 13 9
Likely i 9 12
Strategy C Male Female
Not Likely 5 5
Undecided 5 4
Likely 62 65
Subordinated sex did not alter the likelihood of
employing any of the strategies.
Table 8 provides the mean ratings for the
I
likelihood of employing each of the strategies. As
i
in previousjdisplays, the means are for illustrative
purposes only.
1
The effectiveness cross check for the second
1
proposition 1 suggested that subordinates believe it
I
is more effective to challenge their supervisors
conflictingidecisions based on the sex of the
67
I


subordinate; Again data analysis failed to support
this prediction. (See Table 14).
Table 14
Effectiveness of Each Strategy
by Sex of Subordinate
Strategy A Male Female
Not Effective 28 26
Undecided 21 24
Effective 21 24
Strategy B Male Female
Not Effective 64 68
Undecided 3 2
Effective 3 4
1 Strategy C Male Female
Not Effective 41 38
Undecided 13 18
Effective 16 18
There were no statistically significant differences
found for the effectiveness of any of the
strategies. The statistics for Strategies A and C,
respectively, were (x^= 0.10, p < 0.95) and =
1.00, p < 0.58). There was insufficient data for
analysis of Strategy C. Subordinate's sex did not
alter the effectiveness of any of the strategies.
For the mean ratings of effectiveness of each
strategy see Table 10.
68


Historical of Conflict
Finally, the third proposition suggested that
subordinates will challenge their supervisor's
negative performance appraisal when there is a
I
history of conflict between the supervisor and the
subordinate: There was no statistically significant
difference based on history of conflict. The
1
statistic for strategy based on history of conflict
was (x^= 3.13, p < 0.08). Table 15 contains the
cell frequencies for the strategies based on history
of conflict.
Table 15 '
Preference for Strategy by History of Conflict
Strategy A
Preference 1 2 3
Good 1 36 29 12
i Bad 34 27 5
' Strategy B
Preference 1 2 3
Good 8 7 62
Bad 2 3 61
Strategy C
Preference 1 2 3
Good , 33 41 3
Bad 30 36 0
Once again there was also no statistical
69


difference based on the combinations of the
(
strategies.1 The statistic for history of conflict
was (x^= 3.05, p < 0.18). Table 16 illustrates the
frequencies for historical context.
Table 16
Strategy by History of Conflict
i
i Good Bad
Dispute 36 35
Give Up _ 8 2
Write Rebuttal 33 29
The responses for history of conflict were highly
skewed to dispute the appraisal and writing rebuttal
as the first choice. As a result, further analysis
of the data is difficult.
Again the proposition was cross checked by
examining tlie likelihood of employing the strategy
that subordinates are more likely to challenge their
supervisor's negative performance appraisal when
there is a history of conflict between the
supervisor and the subordinate. The data did not
i
support this prediction. History of conflict did
not affect the likelihood of employing Strategies A
or B. The statistic for the likelihood of employing
Strategy A was (xz'= 3.20, p < 0.20). The statistic
for Strategy B was {xz= 2.80, p < 0.25). There was
I
70


insufficient data for analysis of Strategy C.
Table 17
Likelihood of Employing Strategy
by History of Conflict
Strategy A Good Bad
Not Likely 14 6
Undecideh 10 8
Likely 53 55
Strategy B Good Bad
Not Likely 56 47
Undecided 9 13
Likely 14 7
Strategy C Good Bad
i Not Likely 7 3
Undecided 7 2
Likely 63 64
The third proposition was also cross checked
with the effectiveness of the strategy. This
technigue implied that subordinates believed it was
more effective to challenge their supervisor's
negative performance appraisal when there was a
history of conflict between the supervisor and the
subordinate. Once again, the data did not support
the prediction. Table 18 illustrates the responses
j
for effectiveness by history of conflict.
71


Table 18
Effectiveness of Each Strategy
by History of Conflict
Strategy A Male Female
Not Effective 27 27
Undecided 23 22
Effective 25 20
Strategy B Male Female
Not Effective 72 60
| Undecided 2 3
Effective 1 6
Strategy C Male Female
I Not Effective 44 35
i Undecided 13 18
Effective 18 16
Apparently history of conflict did not affect the
I
effectiveness of Strategies A and C. Both
strategies A and C require some course of action.
The statistics for the effectiveness of Strategies A
and C were (x^= 0.20, p < 0.90) and (x2= 1.60, p <
0.47), respectively. Insufficient data precluded
the analysis of Strategy B.
Sex of subordinate and Sex of Supervisor
Also of interest to the study was the
interaction1between sex of the subordinate and sex
of supervisor on the course of action chosen by the


subordinate when in conflict with the supervisor.
Unfortunately, there was insufficient data for an
accurate interpretation of the interaction. Table
]
19 summarizes the data collected.
Table 19
Strategy by,Interaction Between Sex of Subordinate
and Sex of Supervisor
Subordinate Male Female
Supervisor Male Female Male Female
Give Up 2 3 3 2
Dispute 15 22 19 15
Write Rebuttal 13 16 15 18
These findings encouraged further investigation
of this type with larger amounts of data.
Summary
This chapter has presented available
demographic sample information and the statistical
analyses for the data collected. Initially, the
demographic variables extracted from the sample were
analyzed. The demographic variables were used as
control variables in the study. There were no
statistically significant differences among the
control variables. The dependent variables
collected were modeled using a multinomial logit
model and a chi-square technique. Although the
73


rankings for each strategy did not differ, data
analysis partially supported the proposition that
?
subordinates are more likely to alter the course of
action chosen when in conflict with their supervisor
based on the sex of the supervisor. The likelihood
of employing Strategy A was statistically
significant. There was no statistically significant
difference for Strategy B. There was insufficient
data for analysis of Strategy C. Neither the sex of
the subordinate nor the history of conflict of the
situation provided statistically significant
variations. Interpretation and implications of the
findings will be discussed in the next chapter.
I I
I
I
74


CHAPTER V
CONCLUSIONS
Introduction
The plight of women in organizations has been a
topic of concern for some time. This concern has
amplified as women continue to increase their labor
force participation. As women continue to advance
into the management levels of organizations, it is
of extreme importance to understand how their
presence affects the dynamics of interaction within
organizations.
The purpose of the research was to test the
behavior of individuals in conflict within an
organizational setting. In the present study, the
status characteristic of relevance was sex.
Implicit in the organizational setting were the
conditions that the supervisor and the subordinate
were involved in a task-oriented, collectively-
oriented work situation. Lacking in the present
study was an informal group structure. The research
question posed was what effect, if any, the sex of
the supervisor had on the course of action a
subordinate took when in conflict with the
supervisor. Also of interest was the effect the sex


of the subordinate had on the course of action
chosen. Finally, the history of conflict between
the supervisor and the subordinate was examined for
a possible mitigating effect.
Sex of the Supervisor
The first assumption of status characteristics
theory was that the status characteristic of
interest was salient. Despite the fact that mixed-
sex as well as same-sex dyads were examined, it does
not appear that sex was a salient characteristic.
Gender differences are apparently not as fundamental
as traditional theories claimed and so are not
always salient in every situation (Wagner 1988).
The flexibility of status characteristics theory
allows for the fact that sex differences may not be
salient because of the situation or the position of
the individuals involved (Wagner 1988).
The leadership position of the supervisor
apparently reduced or eliminated inequities in the
power and prestige order within the group and
legitimated the supervisors' decisions about the
group members (Meeker and Weitzel-O'Neill 1977;
Ridgeway 1982; Wagner 1988). The effects of sex
differences appeared to be overstated (Wagner 1988).
76


The present study and other research evidence leads
to the suggestion that there was no systematic
advantage fOr males and disadvantage for females in
the formal structure of the organization (Rosen and
Jerdee 1973; Wiley and Eskilson 1976; Bartol and
Butterfield 1976; Fennell et al. 1978; Eagly and
i
Wood 1982; Kushell and Newton 1986; Molm 1988;
Seifert and^Miller 1988; Wagner 1988). In many
situations males exhibit status-inferior behavior
while females demonstrate status-superior behavior
(Wagner 1988).
The results of the present study showed that
sex of the supervisor did not alter the course of
i
action chosen. No statistically significant
difference based on the sex of the supervisor was
recorded. The only indication that sex of the
i
supervisor affected the course of action chosen by
subordinates was the likelihood of employing
Strategy A. It does appear that subordinates were
more likely to employ a strategy of disputing their
appraisal when the supervisor was female. This was
the only difference based on sex of the individual
involved, noted by the study. Although no other
studies have explicitly examined conflictual
situations between supervisors and subordinates, the
77


results are,consistent with the findings of similar
studies. Leader's sex does not affect subordinates'
satisfaction (Kushell and Newton 1986), leadership
effectiveness (Rosen and Jerdee 1973), nor leader's
performance'evaluations by subordinates' (Seifert
and Miller 1988). In the Seifert and Miller (1988)
study there was no tendency for the performance of
male leaders to be evaluated more favorably than
i
female leaders.
Organizations have been thought to be gendered,
which should result in a disadvantage for women
I
(Martin 1991). Fortunately for women in managerial
positions, the results of the present study did not
confirm that prediction. Apparently the
I
organizational structure is the dominate force in
determining the action of individuals within its
boundaries. Ridgeway (1988) explains the process in
which the position within an organization dictates
the behavior of the individuals within the
organization as:
There are two sources of an authority's
ability to mobilize such support. One is
resources granted to the authority from
outside the system, usually from a larger
organization or the larger society. The
other is the direct, personal approval
given tihe authority by subordinates within
the system. Their review of the evidence
suggests that the first is more important,
78


and indeed appears to be a necessary
condition for authority's ability to
achieve compliance. (Ridgeway 1988,205)
The organizational structure dictates to
individuals the range of behaviors for addressing
i
legitimate authority figures regardless of the other
characteristics of the individual within the
position.
Sex of the Subordinate
Individuals often behave in similar ways in
similar situations provided their status, authority
or organizational positions are comparable (Unger
1976; Kanter 1977; Eagly and Wood 1982; Wagner
1988). These findings are consistent with similar
studies featuring varying aspects of the supervisor-
subordinate relationship. The results of the
present study also showed there was no statistically
significant difference in the course of action
I
chosen based on the sex of the subordinate.
Unfortunately, making assumptions about the
reaction of male and female subjects was at best
difficult. jThe research on small groups has been
greatly biased by the fact that most of the
researchers jhave been male (Dion 1985). Also much
of the original research in the area of small groups
79


made assumptions about group interaction based
solely on male subjects (Dion 1985).
Of the;studies that have focused on sex, the
results of the theoretical and empirical literature
on reactions based on sex of the subordinate have
been mixed. This study agrees with findings by
Webster and Foschi (1988) who state that both high
and low status individuals tend to react in similar
ways to the status of others. Despite this,
empirical research provides inconsistent
conclusions. In one study, male and female
subordinates evaluated leaders of the opposite sex
more favorably than same sex leaders (Seifert and
Miller 1988). Stewart (1988), in studying status
characteristics involving pay equity, found
differencesibetween male and female subjects. The
results in this area seem to be inconclusive.
Similarly, there was insufficient data to evaluate
the effect of the interaction between sex of the
supervisor and sex of the subordinate on the course
of action chosen by the subordinate when in conflict
I
with the supervisor.
Many of the subjects participating in the study
explained that they had previously been involved in
disputes with supervisors over their performance
80


appraisal in the past. As a result, some of the
subjects had established personal procedures for
dealing with similar situations. Even for subjects
with no preconceived standard procedure, personal
experience was likely to have colored their
response. In such instances, it is reasonable to
assume that neither the sex of the supervisor nor
the sex of the subordinate altered the course of
action chosen by the subject. Probably subjects
i
responded as they had behaved in the past.
History of Conflict
The majority of status characteristics theory
has evolved!under the assumption that the power and
prestige order is formed by status characteristics
in ad hoc groups (Meeker 1981). The study examined
a situation'which would extend the utilization of
the theory to previously established groups. This
extension to the theory is of great practical
importance due to the fact that most organizational
groups have some longevity. It appears that in
providing historical context to the situation there
was a failure to extend the use of status
I
characteristics theory beyond the scope conditions
of informal group structure.
81


History of conflict was irrelevant to the
strategies for course of action. These strategies
were disputing the appraisal with the supervisor's
supervisor, giving up and writing a rebuttal to
attach to the appraisal. There were no significant
differences in any of these strategies.
It is possible that the history of conflict
may alter the course of action chosen by a
subordinate1 when in conflict with a supervisor in
the strategy involving inaction. In this situation,
it appears that if the historical relationship
between the subordinate and the supervisor had been
good in thetpast, the subordinate was more likely to
i
choose to do nothing. Unfortunately, there was
insufficient data available to make statistically
i
valid evaluations of these relationships. Probably
the subordinate did not feel the need to strain the
relationship with the supervisor even though they
strongly disagreed with the appraisal. Because the
past relationship with the supervisor had been good,
subordinates were inclined to blame inequities in
the system for their negative appraisal.
It appears that the structure of the
organization dictates the course of action chosen by
subordinates. Despite previous assumptions that
82


women in leadership positions are judged foremost as
women and then as leaders (Martin 1991), it appears
that the structural authority position dictates the
course of action subordinates' chose. Other than
the possibility of personal history with a
supervisor, structural authority appears to be the
overriding legitimating force in the study.
Apparently in the present organizational-oriented
society, the status derived from the organizational
position occupied mediates the status characteristic
of sex.
Contr ibut ions
One of the major contributions of the research
is that the hypotheses were tested in an actual
organizational setting. As a result, the sample
drawn from the organization varies in terms of age,
occupational level and years of work experience.
The sample is most likely more representative of the
general population than studies involving strictly
college students. Despite this a word of caution is
necessary in attempting to generalize the results.
The organization which formed the setting for the
research was involved in reorganization at the time
of the study. This re-organization had caused a
83


great amount of turmoil for the employee population.
The present; study presents an a priori analysis of
gender differences in interaction within small
groups. This type of study within an organizational
setting has been recommended frequently in the past
(Fennell et al. 1978; Stewart 1988).
Another major contribution of the research was
I
the situation between the supervisor and subordinate
was a concrete situation which posed a conflict.
Many of the other studies interested in sex of the
supervisor and management have concentrated on more
abstract concepts such as subordinate satisfaction,
management characteristics and styles, or emergent
i
leadership. The present study was interested in the
effect of the status characteristic of gender, but
focused on the behavior the characteristic evoked.
Also, the leadership this study focused on was
already in place in the organization. As Schein
(1973) recommended, the present study provided a
measure of behavior rather than of attitude.
Recommendations
A major recommendation for further research is
to test the difference between behavior in ad hoc
groups and groups with previous history. In this
84


study, a positive or negative history of conflict
did not alter the course of action chosen by
subordinates.
Other recommendations for further research
include involve a more diverse group of subjects and
replicate the study in a more stable organizational
setting. The subjects included in this study were
volunteers from mostly staff management positions.
Non-management employees were under-represented as
subjects. Including more non-management employees
should provide for a more diverse group of subjects
in terms of income level and probably educational
level. Constraints in resources including time and
cost prohibited a more elaborate experimental
setting in conducting this study. Further research
should attempt a better control for extraneous
variables.
Drawing subjects from a more stable
organizational environment would provide an
interesting comparison to this study. It is unclear
if the subjects in this study were more likely to
take action or not due to the uncertainty of the
I
organizational situation.
The present study focused on the formal
organizational procedures which employees have
85


available in resolving disputes. These formal
procedures revealed no differences in action between
males and females. Further investigation is needed
in order to examine if informal procedures for
resolving disputes would lead to the same
conclusions.
Conclusions
The present study did not find significant
differences in behavior based on the sex of the
individuals in the situation. This study does not
wish to suggest that women encounter no obstacles in
the work place. As the distribution of male and
female employees by organizational level suggests,
I ...
women have not yet reached equality within
organizations.
The organizational setting in which the study
was conducted, provided a unique perspective for
collection and interpretation of the data. The
implications of the study do suggest that
organizations can aid their employees in attaining
their full potential. By instituting formal
1 i
policies and procedures, organizations are able to
control a range of employee behaviors. Strict
adherence td the guidelines mitigates individual
86


advantage for high status employees over low status
employees. Reverence to the position rather than
the individual occupying the position is essential.
When organizations legitimate the actions of their
employees, those employees are successful regardless
of individual status characteristics. The inability
of individuals to discriminate in the organizational
setting may eventually lessen informal bias against
women in other situations.
87


APPENDIX A
LETTER OF INTRODUCTION AND CONSENT FORM
My name is Melissa Holm. I am currently an
employee of this firm and a graduate student at the
University of Colorado at Denver. To complete my
degree, I am presently working on an experimental
research project.
The research project will study the effect of
organizational change on the level of conflict
within an organization. Although the company is
aware of this research, this project is not
sponsored by nor in any way affiliated with this
company.
In order to fulfill the requirements of the
project, I need volunteers to assist me in this
research. These volunteers will be asked to read
two short passages and respond to a questionnaire
about each of the situations presented. The entire
project should take no longer than 15-20 minutes. I
would greatly appreciate your participation. In
order to minimize the amount of your personal time
I
required, I will schedule the project for a portion
of a lunch hour.
Your participation in this project will not


expose you to any painful or distasteful experience.
All of the responses from the research are
confidential and anonymous. If you would be willing
to participate, please sign the consent form and
return it to me in the envelope supplied. I will
then contact you by phone with further details
concerning dates and times for participation. I
will answer all questions concerning the purpose of
this study upon completion of your involvement. You
may choose to discontinue participation in this
project at any time.
If you'have any questions concerning this
research, please feel free to contact me on 896-
OS 69. If you have questions concerning your rights
as a subject, you may direct them to the Office of
Research Administration, University of Colorado at
Denver, Boxj123, 80204, telephone 556-2770.
Your participation would be extremely helpful.
If you would be willing to participate, please
complete the form below. Thank you.
89


1
CONSENT FORM
I have been! informed of the research process and I
am willing jto participate.
Name (please print) _______________________________
Sex M F
Age ________;______
Phone Number ______________________________________
Best Time to Call _________________________________
Signature _________________________________________
I
i
90