THE EFFECTS OF DISABILITY AND GENDER
ON PERCEPTIONS OF POWER AND PERFORMANCE EXPECTATIONS
B.S., Colorado State University, 1992
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Master of Arts
This thesis for the Master of Arts
has been approved
Nett, Barbara (M.A., Psychology)
The Effects of Disability and Gender on Perceptions of Power and Performance
Thesis directed by Professor Herman Aguinis
This exploratory study was undertaken to examine the impact of disability and
gender on perceptions of power and performance expectations. Two hundred
seventy-two participants read a hypothetical resume for the position of instructor of
psychology and responded to scales measuring perceptions of power and teaching
effectiveness. Drawing from research involving other minority groups, it was
hypothesized that disability would negatively influence perceptions of power and
teaching effectiveness. Results of two 5X2 MANOVAs revealed no significant
main effects for disability and gender on the measures of power or teaching
effectiveness. The results are encouraging for individuals with disabilities in
academic environments because disability did not detrimentally influence any of the
dependent variables included in the study. Suggestions for future research on power
and its relationship to disability are provided for both academic and broader
This thesis accurately represents the content of the candidates thesis. I recommend
I wish to express my sincere thanks to Dr. Herman Aguinis for his excellent advising
and continual support and enthusiasm throughout this project. I also wish to
acknowledge the contributions of Dr. Kurt Kraiger and Dr. Laura Goodwin whose
interest and participation was greatly appreciated.
1. INTRODUCTION .................................................... 1
Power as a Determinant of Success ........................... 5
Power Defined ............................................ 5
Credibility as a Base of Power ........................ 7
Minority Groups and Misperceptions of Power .................... 8
Extending Minority Group Research to the Present Study ........ 10
Hierarchy of Acceptance for Disability ................. 13
Gender and Disability ................................. 14
Disability and Performance Expectations ....................... 16
2. METHOD ......................................................... 19
Overview of Design .......................................... 19
Participants ................................................. 19
Procedure ................................................... 19
Dependent Measures ............................................ 21
Power Bases ............................................. 21
Teaching Effectiveness .................................. 22
3. RESULTS .................................................... 23
Manipulation Checks .......................................... 23
Psychometric Properties of the Scales .................... 24
Main Study .................................................. 27
Effects of Disability and Gender on Perceived Power .... 27
Effects of Disability and Gender on Performance
Expectations ........................................... 32
Effects of Disability and Gender on Hiring and Salary
Recommendations ....................................... 35
Post Hoc Analyses ...................................... 37
4. DISCUSSION ................................................. 38
Contributions ................................................ 43
Limitations .................................................. 45
Suggestions for Future Research .............................. 47
Sample Stimulus Materials .................................... 51
REFERENCES .......................................................... 59
Figure 3.1 Mean Ratings on Each of the Power Bases by Disability ........... 30
Table 3.1 Power Scale Reliabilities and Intercorrelations ........... 25
Table 3.2 SEEQ Scale Reliabilities and Intercorrelations ............ 26
Table 3.3 Means and Standard Deviations on Each of the Power
Bases by Disability and Gender ................................ 29
Table 3.4 Means and Standard Deviations on SEEQ Factors of
Teaching Effectiveness by Disability and Gender ............... 34
Table 3.5 Means and Standard Deviations on Hiring and Salary
Recommendations ............................................... 36
Table 4.1 Correlations Between Social Distance Scores and Mean
Scores on Measures of Power, SEEQ, and Hiring and Salary
Recommendations ............................................... 42
Despite legislative and social changes, persons with disabilities remain
underrepresented and underutilized in the workforce: Of the-15.6 million Americans
with disabilities of working age (16-64) in 1992, more than 65% were unemployed
(US Bureau of the Census, 1993). In contrast, nearly 80% of working-age persons
without disabilities participate in the workforce. This imbalance in workforce
participation seems even more striking considering that, when polled, more than 80%
of unemployed persons with disabilities expressed a desire to work, even at the cost
of reduced disability benefits (Gutman, 1993).
Although some portion of persons with disabilities are unable to work, disability
remains associated with low rates of workforce participation and with lower earnings
for those persons with disabilities who are employed, despite their educational level
(Asch, 1984). While the passage of the Americans With Disabilities Act (ADA) of
1990 has provided considerable impetus to persons with disabilities in their struggle
for equal participation in employment, progress toward equality is slow. The 32,993
ADA-related cases filed with the U.S. Equal Employment Opportunity Commission
in 1994 (Bias in the Workplace, 1995) alone is testament to the point that inequity
remains pervasive in the collective American workplace.
Although the underlying reasons for these employment problems are multi-
faceted and complex, it has been suggested time and again that one source of the
problem stems from societys negative attitudes, stereotypes and perceptions
associated with disability. Attitudes refer to enduring learned predispositions to
behave in a consistent way toward an object, in this case, toward individuals with
disabilities. A stereotype is a cognitive structure containing the perceivers
knowledge and beliefs about-a social group and its members (Hamilton, Sherman, &
Ruvolo, 1990). Stereotyping, then, involves forming impressions and trait
descriptions of particular groups and assigning these traits to a particular individual
once his/her membership in that group becomes known (Arvey & Faley, 1988). A
perceptual bias is a tendency to interpret, perceive, remember or explain an
individuals actions in ways consistent with expectations (Jussim, 1989).
Negative attitudes and stereotypes held by employers influence the employment
of persons with disabilities with respect to hiring, type of position, compensation, and
advancement opportunities (Braddock & Bachelder, 1994). The unfounded
assumptions of employers about persons with disabilities in terms of productivity,
absenteeism, turnover, and cost of accommodation and insurance defeats opportunity
for many persons with disabilities.
Stone and Colella (1996) proposed a model that illustrates the treatment of
persons with disabilities in organizations is a function of person characteristics,
environmental factors, and organizational characteristics and that the relationships
among these factors are mediated by observers cognitions and affective states. Stone
and Colella (1996) further pointed out that most research on disability-related issues
has failed to focus on the processes underlying the attitudinal and perceptual biases
associated with disabilities.
As an advocate of disability research intended to ultimately improve the societal
conditions for persons with disabilities, Asch (1984) proposed an approach to the
study of disability which alleges that, ,such research should seek to analyze the
psychological, environmental, and social concomitants of disability in the situations
studied and to ponder whether differences in power and status may explain some of
the findings. This statement recognizes that researchers studying disability must
remain attentive to extraneous variables which may confound their results and
interpretations. Only through well-constructed research designs may researchers be
able to conclude that the presence of differences may be attributed solely to disability
and contextual factors, and do not stem from power and status differences between
the nondisabled persons and the persons with disabilities. However, before the
unequivocal results sought by Asch can be attained, it must be determined that
persons with disabilities are not perceived as having diminished power as a function
of the disability. If perceptions of power of persons with disabilities are influenced
by the disability itself, interpretations may still be confounded by power and status
Consequently, the current research project was undertaken to examine the
possibility of this perceptual bias and to explore whether disability has an effect on
perceptions of power made of individuals with disabilities. Although the concept of
social power has received considerable attention in both the social and organizational
psychology literature, until now, the topic of power and how it relates to disability has
been disregarded. This research represents the first empirical attempt to examine
perceptions of power made of persons with disabilities.
Because the current research was restricted to a student sample, the focus became
that of the academic employment arena in order to lend an element of realism to the
study and to attempt to breach the gap of generalizability to the organizational
context. Although there are inherent limitations to generalizability when utilizing
student samples, the academic context may be a very telling environment in which to
explore these perceptions because the possibility of discrimination among college
instructors with disabilities seems highly likely given that, as of 1991, only 0.9% full-
time college faculty members reported having a physical disability (American
Association for the Advancement of Science, 1991).
Power as a Determinant of Success
Power, as discussed here, may be defined as an individuals capacity to influence
another. Bases of social power stem from the characteristics, abilities, and resources
an individual possesses which are instrumental in his/her ability to influence another.
The classic framework developed by French and Raven (1959) identified five
potential bases of power: Reward, coercive, legitimate, referent, and expert. Reward
power refers to an individuals perceived ability to provide rewards, whereas coercive
power is based on the individuals perceived ability to deliver punishments.
Legitimate power is based on the perception that an individual has the legitimate
authority to make certain requests and demands of another (perhaps by virtue of their
supervisory role). An individual is said to possess referent power if another desires to
be associated with him/her. Finally, if an individual is perceived as having the ability
to provide another with some type of knowledge, he/she is believed to possess expert
While the importance of power in organizations has long been discussed in the
social and organizational psychology literature, power is now being recognized as
instrumental for success in the academic setting. For instance, it is fairly well
established that the use of referent and expert power enhances student motivation and
learning (self-report of course grade expected), while the use-of coercive power, and
to some extent, legitimate power, may serve to hinder motivation and learning
(Richmond, 1990; Richmond & McCroskey, 1984). Romer and Whipple (1991)
describe the functional role of power and its importance in student-professor
collaboration. They contend that there is a delicate relationship between power and
authority in student-professor collaborations that, when managed effectively, provides
the student with a liberating educational experience. Aguinis, Nesler, Quigley, Lee,
and Tedeschi (1996) provided empirical support for the importance of power in
student-professor collaborations. Aguinis et al. (1996) examined the role of power in
the graduate student-faculty supervisor relationship. They found that professors high
referent, expert, and reward power contributed to positive ratings of graduate
students perceptions of the quality of their professor-student relationship. However,
the perceived use of coercive power by professors had a detrimental effect on
graduate students perceptions of the quality of the same relationship. The authors
noted that this power relationship is important because, as mentors, graduate faculty
are in the position to substantially influence the students graduate experience and
Credibility As a Base of Power. .
Recognizing that there may exist more bases of power than those proposed by
French and Raven (1959), Nesler, Aguinis, Quigley, and Tedeschi (1993) proposed
that credibility may, in fact, be another possible base of power. These researchers
characterize persons with high credibility as those who are consistently both honest
and accurate in their communications and, conversely, persons with low credibility
are those who are not truthful and do not match words and actions. These researchers
asserted that, if credibility is a base of power, there should be a direct relationship
between credibility and perceived power. Results from their study supported this
assertion and provide evidence that credibility may operate as a base of power, much
as those defined by French and Raven (1959). Given this finding, credibility is
included in the current research as an additional power base of interest.
Credibility, like power, is enjoying increased popularity in the educational
psychology literature (Beatty & Zahn, 1990; Frymier & Thompson, 1992). Aguinis
et al. (1996) included measures of graduate faculty credibility in their examination of
power in the graduate student-faculty supervisor relationship. Regressing credibility
ratings on scores for the five power bases, they reported that the main effects of the
five power bases accounted for 61% of the variance in credibility ratings. Graduate
supervisors ratings of credibility were positively associated with students
perceptions of their expert power and negatively associated with perceptions of their
coercive power. Beatty and Zahn (1.990) investigated the impact of instructor
credibility on student course ratings in communications courses. They reported that
instructor credibility was positively associated with overall ratings of level of
satisfaction with the course and instructor, students intentions to recommend the
course and instructor to peers, and intentions to take additional courses from the
instructor. Similarly, Frymier and Thompson (1992) found that students perceptions
of instructor credibility are positively associated with students motivation to study.
While the topic of power and disability has not yet been examined, research on
power has focused on other minority groups. An understanding of this related
research may provide a preliminary framework for deriving hypotheses about the
perceived power of persons with disabilities.
Minority Groups and Misperceptions of Power
It is widely recognized that although women and minorities are achieving equal
representation in the workforce, they are not advancing as far or at the same rates as
their white male counterparts. As described in a report issued by the U.S. Department
of Labor, there is a level beyond which few minorities and women had either
advanced or been recruited, and minorities tended to be found at lower levels of
management than women (Report on the Glass Ceiling Initiative, 1991). It has been
suggested that one reason for this disparity in organizational representation at higher
levels may be perceived differences in power. That is, minority-group membership
may serve to influence an individuals perceived power..
Ragins and Sundstrom (1989) asserted that perceptions are critical to the
development of power. They further point out that these perceptions are highly
susceptible to the influence of stereotypes and attributions which may distort this
process of developing power. For instance, Ragins (1989) reasoned that the perceived
power of women may be distorted through gender-based stereotypes which imply that
power is stereotypically male.
In the case of women and minorities, stereotypes and attributions about their
abilities contribute to what is commonly referred to as the glass ceiling effect which
limits their opportunities for promotion or advancement. However, these same types
of distorted perceptions may have more dramatic implications for persons with
disabilities. That is, instead of persons with disabilities seeking to surpass barriers to
advancement, they are most often seeking to destroy barriers which limit their access
to employment opportunities in the first place. It may be that perceptual biases
associated with disability diminish the perceived capacity of persons with disabilities
to function effectively; impeding their advancement and wrongfully limiting their
accomplishments. Preconceptions of persons with disabilities as powerless
individuals may render them unable to achieve the same organizational status and
benefits as others, similar to the dilemmas experienced by women and minorities.
Extending Minority Group Research to the Present Study
Drawing from the previous literature on power involving other, more traditional,
minority groups allows for the development of analogous hypotheses regarding
individuals with disabilities.
Social psychological research suggests stereotypes are more likely evoked in the
absence of individuating information about a person (Hamilton, Sherman, & Ruvolo,
1990). Because the number of college faculty with disabilities is very insignificant, it
is highly likely that students have never before encountered instructors with
disabilities. For this reason, they will not have had the opportunity to develop a
cognitive framework about instructors with disabilities. Therefore, there is a greater
likelihood of a reliance on stereotypes to form perceptions than there would have
been had they encountered instructors with disabilities which may have discontinued
any preexisting stereotypes.
Stereotypes and attributions associated with disability may serve to distort the
perceived power of persons with disabilities, much as the perceived power of women
and minorities is also influenced by stereotypes. Therefore, it is necessary to examine
empirically whether individuals with disabilities are perceived as powerless, and
whether these perceptions serve to limit their advancement and success in the
workforce. This question is particularly important in light of the fact that persons
with disabilities are even more likely to experience hiring and wage discrimination
due to uncertainties about productivity than are members of other minority groups
Common stereotypes about persons with disabilities include notions that they are
incompetent (Fine & Asch, 1988), dependent or helpless (Fowler & Wadsworth,
1991), and inferior (Wright, 1983). Given that these stereotypes are somewhat
synonymous with powerlessness, as logic dictates, these stereotypes would detract
from the perceived power of persons with disabilities. Thus:
Hypothesis 1: Individuals with disabilities will be perceived as having
less power and credibility than individuals without disabilities.
Given the exploratory nature of the study, individual hypotheses for each power
base were not developed due to an absence of previous research on the topic.
However, because the power bases emerge from various characteristics, abilities, and
resources associated with the individual it is reasonable to infer that these power
bases will be differentially influenced by disability. For instance, because referent
power arises from an individuals desire to be associated with another, an individual
who holds negative perceptions of disability is likely to have a lesser desire to be
associated with persons with disabilities. Studies examining attitudes toward
disability through the use of social.distance measures suggest that people generally do
not desire to be closely associated with persons with disabilities. For instance, Olkin
and Howson (1994) demonstrated that more favorable attitudes are expressed toward
persons with disabilities when levels of intimacy are more distant. This may be
interpreted as meaning that people have less desire to be associated with individuals
with disabilities as the level of intimacy increases. Thus:
Hypothesis la: Individuals with disabilities will be perceived as having
less referent power than individuals without disabilities.
Similarly, because expert power is based on an individuals perceived knowledge
and expertise, stereotypes which imply that individuals with disabilities are
incompetent may be translated into diminished perceptions of an individuals expert
power. In other words, common stereotypes which ascribe to persons with
disabilities the traits of incompetence and inferiority may lead others to hold negative
perceptions of the individuals ability to provide them with what could be considered
expert knowledge. Consequently:
Hypothesis lb: Individuals with disabilities will be perceived as having
less expert power than individuals without disabilities.
Hierarchy of Acceptance for -Disability
It is well documented that attitudes toward individuals with disabilities are not
homogeneous across the wide range of disabilities. Research has clearly determined
that different disabilities seem to evoke different stereotypes and attitudes. Most
researchers agree that attitudes toward individuals with disabilities vary as a function
of several factors, some of which include the visibility of the disability (Stone et al.,
1992;. Yuker, 1994), the degree to which the disability is associated with either
uncertainty or the inability to predict a persons behavior (Stone et al., 1992), and
attributions about the cause or origin of the disability (Esses & Beaufoy, 1994; Yuker,
Moreover, researchers have recognized that there is a relatively stable hierarchy
of acceptance for disability types (Olkin & Howson, 1994; Tringo, 1970). This
research indicates that the most accepting or favorable attitudes are demonstrated
toward physical disability (e.g., quadriplegia), then sensory impairments (e.g.,
deafness, blindness), following by psychogenic or mental disabilities such as
mental retardation, alcoholism, and mental illness. Research examining attitudes
toward persons with AIDS has revealed that the attitudes are predominantly negative
(Esses & Beaufoy, 1994), presumably because of the attributions surrounding the
origin of AIDS (e.g., homosexuality, IV drug use, promiscuous behavior) (Pulliam,
Based on these findings, it cannot be expected that ratings of power will be
uniformly affected by the stimulus of disability; but rather, that the ratings will be
differentially influenced by the various stereotypes, preconceptions, and attributions
associated with different disabilities. It is reasonable to infer that perceptions (a
cognitive component of attitudes) will also vary with the nature of the disability. In
other words, it is reasonable to hypothesize that perceptions of power will be more
biased toward those specific disabilities which attitudes are known to be more
Hypothesis 2: Individuals with AIDS, alcoholism, and chronic depression
will be perceived as having less power than individuals with quadriplegia,
who, in turn, will be perceived as having less power than individuals without
Gender and Disability
Recently, it has been noted that women with disabilities are sometimes perceived
as having a double disability (i.e., that of being a woman and having a disability)
(Hanna & Rogovsky, 1991). There appears to be some empirical support for this
undesirable claim. Females with disabilities who are employed full-time earn only
65% as much as men with disabilities employed full-time (Bowe, 1992). Women
with disabilities encounter wage discrimination based on both gender and impairment
(Johnson & Lambrinos, 1985).
Gender, race, and social class interact to determine womens and mens positions
within organizations (Harlan & Berheide, 1994). Many types of occupations are
gender stereotyped; identifying men with certain positions and women with others.
For instance, successful middle managers are perceived to possess traits most often
associated with males (Schein, 1973). These gender-based role expectations
influence perceptions about the type of person appropriate for particular positions.
Because perceptions are swayed by gender-based stereotypes (Ragins &
Sundstrom, 1989), it is reasonable to presume that these perceptions may also be
subject to influence by stereotypes associated with other minority groups, such as
persons with disabilities. Gender-based stereotypes may have a more pronounced
effect on women with disabilities, producing this double disability.
In other words, females with disabilities may be perceived as having relatively
less power than males with disabilities. Thus:
Hypothesis 3: Females with disabilities will be rated as having less
power and credibility than both the individuals without disabilities
and the males with disabilities.
Disability and Performance Expectations
Similar to stereotypes, expectancies about individuals also tend to bias the way
they are perceived by others (Stone, Stone, & Dipboye, 1992). Expectations are
important because expectancies are anticipatory beliefs based on stereotypes which
are elicited upon assigning an individual to membership in a category. There is
considerable empirical support for the effects of expectancies on behavior in social
situations (Jussim, 1989; Rosenthal & Jacobsen, 1968; Snyder, Tanke, & Berscheid,
1977; Zanna & Cooper, 1974).
Personnel decisions are often an expression of informal expectations about the
gender, race, and class of the person believed to be most appropriate for the particular
position (Harlan & Berheide, 1994). Invalid expectancies made of individuals with
disabilities may have an indirect impact on personnel decisions affecting them.
Stereotypic expectancies regarding the abilities of individuals with disabilities may
contribute to lower evaluations of their anticipated performance, and consequently,
leave them at a preconceived disadvantage with regard to employment decisions. The
inherent uncertainty in predicting the effect of impairment on productivity and the
misinformed notion that employees with disabilities are less productive than other
workers often leads to discrimination in the employment context. Therefore, for
persons with disabilities, erroneous expectancies regarding their abilities artificially
and unduly circumscribes their role expectations and opportunities.
. In the academic context, an important component in the performance evaluation
of college faculty is the examination of student course evaluations. Marsh (1984)
determined that student evaluations of courses are more a function of the instructor
than the course itself. Consequently, negative evaluations made of instructors with
disabilities may perpetuate discrimination of the instructor with regard to tenure,
promotion, salary, and job assignment.
If, consistent with stereotypes, disability has an adverse effect on perceptions of
power and credibility, it is necessary to determine whether these perceptions promote
negative expectations of individuals with disabilities. Finally:
Hypothesis 4: Individuals with disabilities will receive lower ratings
of expected performance than the nondisabled individuals.
In sum, this study examined the effect of disability and gender on perceptions of
power, credibility, and performance expectations. This study represents a departure
from previous relevant research in at least two ways. First, this study examined
perceived power of individuals with disabilities, a topic unexplored in disability
research until now. Exploring the unfamiliar relationship between power and
disability may provide information about attitudinal and perceptual biases associated
with disability and about power and status differences which may be unknowingly
confounding some disability research.
Second, while previous research on power has examined gender and racial
differences in perceived power, this research has failed to include persons with
disabilities as a minority group of interest. Such integration of disability into
mainstream research is critical to advancing theory and understanding of the
Overview of Design
The research design consisted of a 5 X 2 between-groups factorial experiment,
with disability and gender serving as the independent variables. Dependent variables
were perceptions of power and expectations of teaching effectiveness.
Undergraduate psychology students were invited to participate in the study in
exchange for credit toward the course research participation requirements. 272
subjects participated in the study. Subjects were predominantly white (63.6%).
Eigthy-four participants (30.9%) were male and 177 (65.1%) were female. The mean
age was 23.09 years (s.d. = 6.11); eleven subjects did not reveal their age.
Participants were provided with a resume depicting a hypothetical candidate for a
vacant teaching position within the psychology department. Subjects were informed
that the psychology department was interested in gathering students impressions of
the candidates probable classroom performance (See Appendix).
The resume described the candidate as being near completion of a doctoral
degree in clinical.psychology, and as having modest teaching, research and clinical
experience. The attempt was made to present the candidate in average terms; not
having extensive credentials or experience so as to not unduly diminish or heighten
the students perceptions of their qualifications. In other words, the Candidates
qualifications were not intended to influence the outcome of the study; it was
expected that their qualifications would be viewed neutrally.
Subjects were randomly assigned to one of ten experimental conditions depicting
the candidate as having one of five disabilities: HIV/AIDS, alcoholism, chronic
depression, quadriplegia, or no disability; and as being either male or female.
Questionnaires were completed anonymously and participants were encouraged
to provide honest responses.
Subject information concerning experience with persons with disabilities was
collected through the inclusion of two social distance questions. Subjects were asked
to indicate who the closest person they know is with a both general disability and with
the specific disability in their particular treatment condition. Responses were made
on an 8-point scale (ranging from 1 = No One to 8 = Myself). The modal response to
both questions was 1.00, indicating subjects, overall, had minimal experience with
persons with disabilities. 17.3% of participants reported knowing no one with a
general disability while 41.5% of participants indicated they knew no one with the
specific disability in question. Participants reported closer levels of familiarity with
persons with general disabilities (mean = 3.53; sd = 2.03) than with persons having
the specific disability (mean = 2.62; sd = 2.08) in question.
After reviewing the resume, participants were asked to respond to a modified
version of Hinkin and Schriesheims (1989) power scales (See Appendix, Questions
1-26). Participants evaluated the candidate on six bases of power. This scale was
designed to assess perceptions of the instructors five bases of power (referent, expert,
legitimate, reward, and coercive) delineated by French and Raven (1959). In
addition, this modified scale included items developed by Nesler et al. (1993)
intended to assess the perceived credibility of the instructor. All ratings were
provided on a 9-point Likert-type scale (1 = Strongly Disagree to 9 = Strongly Agree).
Modified versions of this scale have been used very successfully in studies
involving both undergraduate (Aguinis & Adams, 1995; Nesler et al., 1993) and
graduate students (Aguinis et al., 1996).
Participants evaluated the candidates probable teaching performance using a
modified version of Marshs Student Evaluations of Teaching Quality (SEEQ)
(Marsh, 1987). The SEEQ is a 35-item scale designed to measure nine distinct factors
of teaching effectiveness (learning, enthusiasm, organization, group interaction,
individual rapport, breadth, examinations, assignments, and overall) (See Appendix,
Questions 27-57). The SEEQ has been widely used and researched. Between 1976
and 1990, nearly 1,000,000 SEEQ forms were administered in approximately 50,000
courses (Marsh, 1991).
For purposes of the present study, the SEEQ was modified to reflect student
opinions of probable teaching performance, rather than true evaluations of actual
performance. In addition, questions pertaining to student and course characteristics
were not included, resulting in a 31 -item scale.
In addition, the original 5-point Likert-type scale used for the SEEQ was changed
to a 9-point scale in an attempt to increase the variability of responses and to remain
consistent with the scale format used in the power bases questionnaire.
Finally, participants were asked to provide responses to questions about their
endorsement of the candidate for the hypothetical teaching position. For instance,
participants were asked to indicate how strongly they felt the candidate should be
offered the position, and what salary level the candidate merited.
Prior to the main study, a pilot study was conducted to verify the effectiveness of
the disability and gender manipulations. Thirty-four participants were presented with
a resume depicting the candidate as either male or female and as having either
HIV/AIDS, alcoholism, chronic depression, quadriplegia, or no disability. In addition
to the power and SEEQ scales, subjects were asked to respond to items to determine
whether their stimulus resume was that of a male/female and whether the candidate
had a disability. When asked, Does this candidate have a disability: (Yes/No/Not
Sure), all of the subjects in the Chronic Depression or Quadriplegia conditions
responded Yes. However, the majority of subjects in the HIV/AIDS and
Alcoholism conditions (62.5% and 66.6%, respectively) responded No or Not
Sure to the same question. To compensate for this poor correct response rate in
these two conditions, a second and more specific question concerning disability was
added to the survey (i.e., This candidate has HIV/AIDS or alcoholism: Yes/No).
Because HIV/AIDS and Alcoholism are not considered, by many, to be true or
typical disabilities, the additional questions were added to the survey to determine
whether the disability manipulation was truly not detected by participants or whether
they simply didnt consider these conditions disabilities.
The addition of the more precise questions improved the rate of correct responses
to over 95% and confirmed the effectiveness of the manipulation. This indicated that
although participants were perceiving the manipulation, there was uncertainty or
misunderstanding on their part about whether these conditions may be labeled
Psychometric Properties of the Scales
The scales included in this study were chosen on the basis of their demonstrated
reliability and validity (cf. Hinkin & Schriesheim, 1989; Marsh, 1982).
Scale reliabilities and intercorrelations for the measures of power are provided in
Table 3.1. The significant correlations are consistent with previous findings (Hinkin
& Schriesheim, 1989, 1990; Nesler, et al., 1993; Ragins, 1989; Ragins & Sundstrom,
1990) which demonstrate a strong pattern of intercorrerlation among the power bases.
: Table 3.1. Scale Reliabilities and Intercorrelations
Coercive Crediblity Expert Legitimate Referent Reward
Credibility .33 (.86)
Expert .23 .38 (.79)
  
Legitimate .32 .51 .62 (.85)
   
Referent .30 .51 .52 .69 (.80)
    
Reward .27 .36 .63 .65 .61 (.72)
     
Note: All correlations are significant at the pc.Ol level, two-tailed.
Scale reliabilities (Cronbachs alpha) are represented by the numbers in parentheses and appear on the diagonal. Sample sizes
appear in brackets.
Results from the extensive research on the SEEQ have determined it to be a
highly reliable and valid instrument for assessing teaching effectiveness at the
university level (Marsh, 1982). The SEEQ demonstrates remarkable reliability.
Marsh (1982) reported that typical item reliabilities average about 0.90 when based
on 25 or more student responses; internal consistency reliaibilty estimates range from
0.88 to 0.97 for each of the different factors; and the average correlation between end-
of-term and retrospective ratings is 0.83, indicating long-term stability over a several-
year period. The SEEQ has also been validated against faculty self-evaluations and
student learning as-measured.by objective examinations (Marsh, 1982). The median
validity coefficient between student and faculty ratings on the same factors was 0.49.
SEEQ scale reliabilities and intercorrelations are provided in Table 3.2.
Table 3.2. SEEQ Scale Reliabilities and Intercorrelations
Assignments Breadth Enthus Exams Group Interaction Individ. Rapport Learning Organ zatn Overall
Breadth .66 (.82)
Enthusiasm .63 .68 (.80)
  
Exams .77 .74 .65 (.75)
   
Group .68 .74 .76 .70 (.86)
Interaction     
Individual .70 .82 .75 .76 .80 (.84)
Rapport   257]   
Learning .58 .63 .75 .66 .69 .72 (.79)
      260]
Organization .60 .74 .80 .72 .76 .78 .65 (.82)
       
Overall .57 .51 .61 .61 .57 .57 .53 .58 (.86)
        
Note: All correlations are significant at the p<.01 level, two-tailed.
Scale reliabilities (Cronbachs alpha) are represented by the numbers in parentheses and appear on the diagonal. Sample sizes
appear in brackets.
Two separate 5X2 (disability x gender) multivariate analysis of variance
(MANOVA) procedures were performed. The first using measures of the six power
bases as the dependent variables, and the second using the nine factors of teaching
effectiveness as the dependent variables.
Effects of Disability and Gender on Perceived Power
A 5 X 2 MANOVA was performed to examine the effects of gender and
disability on perceived power. Prior to analysis, the data were examined for outliers
and missing data and the assumptions of the MANOVA model were examined.
Forty-five cases were excluded from the analysis because of missing data. An
additional six cases were excluded from the analyses because they were determined to
be outliers, having standardized residual values greater than 3.0 on the dependent
variables (SPSS/PC+ Base System Users Guide, Version 5.0, p. 308). No violations
of the linearity, normality, or homogeneity of variance-covariance assumptions were
detected. Although there was some discrepancy in sample sizes across cells, the
sensitive Boxs M test for homogeneity of dispersion matrices yielded F(189, 41130)
= 1.12,p > .05, confirming the homogeneity of variance-covariance matrices, and
validating the use of the MANOVA procedure.
The results of this analysis revealed that the combined power bases were not
affected by disability, Wilks Lambda = .87, F(4,102) = 1.21, p > .05, r|2 = .03;
gender, Wilks Lambda = .97, F(l, 102) = .91, p > .05, r|2 = .03; or their interaction,
Wilks Lambda = .90, F(4,102) = .96, p > .05, r|2= .03. The statistical power
analyses for the MANOVA procedure for disability, gender, and their interaction
yielded power estimates of .90, .36, and .80, respectively.
The main focus of this study, represented by HI, was to determine whether
candidates with disabilities would be perceived as having less power and credibility
than the nondisabled candidates. This hypothesis was not confirmed. The test for the
presence of a multivariate main effect for disability (as reported above) was not
significant, providing insufficient evidence to substantiate the presence of a direct
relationship between disability and perceived power. The mean ratings on each of the
power bases, separated by disability and gender, are presented in Table 3.3; a graphic
depiction of these mean ratings appears in Figure 3.1.
Table 3.3. Means and Standard Deviations on Each of the Power Bases by Disability and Gender.
COERCIVE mean (s.d.) CREDIBILITY mean (s.d.) EXPERT mean (s.d.) LEGITIMATE mean (s.d.)
MALE 5.80 (0.97) 6.64 (1.07) 7.29 (0.95) 6.38 (1.11)
FEMALE 6.07 (0.95) 6.50 (1.36) 6.88 (1.17) 6.32 (104)
MALE 6.38 (0.84) 6.60 (0.98) 7.07 (0.77) 6.63 (0:78)
FEMALE CHRONIC DEPRESSION 6.10 (0.91) 6.63 (0.97) 6.90 (0.96) 6.55 (0.95)
MALE 5.99 (0.99) 5.70 (0.95) 6.83 (1.00) 6.21 (0.99)
FEMALE 6.15 (1.02) 6.62 (1.12) 7.15.(0.76) 6.49 (0.85)
MALE 5.96 (1.02) 5.86 (0.97) 6.91 (1.27) 6.10 (1.34)
FEMALE 5.88 (1.01) 6.35 (0.90) 6.94 (1.11) 6.36 (0.92)
MALE 6.04 (1.15) 6.14 (0.88) 7.00 (1.15) 6.31 (1.38)
FEMALE 5.76 (1.25) 5.89 (1.17) 6.74 (1.01) 6.05 (1.24)
mean (s.d.) mean (s.d.)
MALE 6.06 (1.31) 6.05 (1.22)
FEMALE 6.09 (1.26) 6.22 (1.28)
MALE 6.28 (1.10) 6.32 (1.05)
FEMALE 6.28 (0.86) 6.24 (0.72)
MALE 5.93 (0.99) 6.24 (0.86)
FEMALE 6.00 (0.98) 6.14 (1.21)
MALE 5.89 (1.08) 6.28 (1.30)
FEMALE 6.12 (1.09) 6.30 (0.99)
MALE 5.80 (0.98) 6.23 (1.51)
FEMALE 5.33 (1.12) 5.67 (1.17)
-s Chronic Depression
- No Disability
Figure 3.1. Mean ratings on each of the power bases by disability
Individual hypotheses concerning disability and power were made regarding
referent and expert power. HI a proposed that disability would negatively influence
ratings of referent power. This hypothesis was examined through univariate follow-
up tests to the previously mentioned 5X2 MANOVA. This hypothesis was not
confirmed. Although, results of follow-up univariate tests revealed a significant
effect of disability on ratings of referent power, F(4, 211) = 2.59, p < .05, r| = .05,
the results were not in the predicted direction. Instead, the higher ratings of referent
power were assigned to the candidates with disabilities. The highest ratings of
referent power were assigned to candidates in the alcoholic condition, followed by
HIV/AIDS, quadriplegia, chronic depression, and lowest ratings were assigned to the
nondisabled candidate. (Again, female candidates received higher ratings of referent
power than male candidates.) This would seem to indicate that participants held a
stronger desire to be associated with the candidates with disabilities compared to the
Similar to HI a, Hlb proposed that disability would negatively influence ratings
of expert power and was examined through univariate follow-up tests to the
previously mentioned 5X2 MANOVA. This hypothesis was also not confirmed,
F(4, 211) = .27, p > .05, r|2 = .01. Again, the nondisabled candidate received the
lowest ratings. The highest mean ratings for expert power were assigned to the
candidates in the HIV/AIDS condition, followed by chronic depression, alcoholism,
and quadriplegia. This would indicate that subjects viewed the disabled candidates as
possessing more knowledge and expertise than the nondisabled candidate.
The univariate follow-up analyses for the remaining power bases indicated a
significant effect of disability only on ratings of credibility, F(4, 211) = 3.31, p < .05,
r| = .06. No significant univariate effects of disability were found for ratings of
coercive, F(4, 211) = .87, p > .05, r|2 = .02; legitimate, F(4,211) = .99, p > .05, r|2 =
.02; or reward power, F(4, 211) = .71, p > .05, t|2 = .01.
Based on the hierarchy of preference for disability, H2 proposed that the
candidates perceived power would vary with disability type. More specifically, this
hypothesis predicted that the nonvisible.and mental disabilities would diminish
perceptions of power more so than quadriplegia. A one-way ANOVA was performed
to examine mean differences across the power bases. To provide evidence for this
hypothesis, mean ratings on the power bases would have to be higher in the
quadriplegia condition, with lower ratings in each of the other disability conditions.
Candidates in the quadriplegia condition did not have the highest mean ratings on any
of the power bases.
As stated previously, gender did not have a significant multivariate effect on the
combined power bases, failing to provide support for the third hypothesis. H3
proposed that a gender effect would be detected, with female candidates being rated
as having less power and credibility than both the candidates without disabilities and
the male candidates with disabilities. Instead, females were rated higher on every
power base except Coercive (see Table 3.3), and, here, the difference was negligible.
Effects of Disability and Gender on Performance Expectations
Before proceeding with the second 5 X 2 MANOVA, the original data set was
again examined for outliers and assumptions of the MANOVA model. Thirty-six
cases were excluded from this MANOVA because of missing data. An additional
eight cases were excluded from the analyses because they were determined to be
outliers, having standardized residual values greater than 3.0 on the dependent
Here, the homogeneity of variance-covariance assumption may have been
violated, posing a threat to the robustness of the MANOVA procedure. As a
preliminary check for robustness of the procedure, the within-cell variances of the
dependent variables were examined. In no case did this ratio exceed 20:1, which
would further suggest that the assumption of homogeneity of variance-covariance had
been violated (Tabachnick & Fidell, 1989). In fact, this ratio only slightly exceeded
3:1. However, Boxs M test produced F(405, 40802) = 1.29, p < .01, indicating the
robustness of this procedure may have been compromised.
The final hypothesis, H4, represented the other focus of this study and proposed
that candidates with disabilities would receive lower ratings of expected teaching
performance than the nondisabled candidates. Multivariate tests of significance for
the effect of disability on performance expectations failed to confirm this hypothesis,
Wilks Lambda = .82, F(4, 104) = 1.21, p > .05, r\2 = .05. Univariate follow-up tests
revealed a significant effect for disability on only one of the eight scales of teaching
effectiveness, Enthusiasm, F(4, 218) = 2.82, p < .05, r\ = .05. The mean ratings for
each of the measures of teaching effectiveness, separated by disability and gender, are
provided in Table 3.4.
Table 3.4. Means and Standard Deviations on SEEQ Measures of Teaching Effectiveness by Disability and Gender
ASSGNMTS BREADTH ENTHUSIASM EXAMS GROUP INT. RAPPORT LEARNING
mean (s.d.) mean (s.d.) mean (s.d.) mean (s.d.) mean (s.d.) mean (s.d.) mean (s.d.)
MALE 6.65 (1.46) 6.34 (1.17) 6.49 (1.46) 6.37 (1.33) 6.40 (1.14) 6.60 (1.29) 6.43 (1.32)
FEMALE 6.22 (1.22) 6.24 (0.99) 6.25 (1.20) 6.03 (1.06) 6.19 (1.26) 6.20 (1.14) 6.22 (1.08)
MALE 6.53 (1.05) 5.91 (0.94) 6.25 (0.84) 6.04 (1.27) 6.57 (0.99) 6.11 (0.92) 6.20 (1.10)
FEMALE 6.30 (1.09) 6.59 (1.09) 6.39 (0.97) 6.35 (1.02) 6.59 (0.95) 6.73 (1.12) 6.88 (0.97)
MALE 6.00 (0.98) 5.75 (0.91) 5.60 (1.05) 5.91 (0.74) 5.82 (1.00) 5.79 (1.02) 5.87 (0.87)
FEMALE 6.67 (1.05) 6.52 (1.05) 6.09 (0.98) 6.44 (1.05) 6.61 (1.25) 6.40 (1.03) 6.48 (1.02)
MALE 6.36 (1.38) 6.15 (1.24) 6.07 (1.03) 6.25 (1.29) 6.16 (1.26) 6.36 (1.38) 6.21 (1.00)
FEMALE 6.75 (1.29) 6.26 (1.02) 6.60 (1.10) 6.25 (1.12) 6.70 (1.33) 6.71 (1.20) 6.53 (1.21)
MALE 6.25 (1.33) 6.26 (1.15) 5.94 (1.13) 6.42 (1.42) 6.30 (1.25) 6.31 (1.32) 6.10 (1.05)
FEMALE 6.27 (1.09) 6.08 (1.04) 5.88 (1.00) 5.97 (1.14) 6.08 (1.16) 6.10 (1.36) 6.19 (1.03)
mean (s.d.) mean (s.d.)
MALE 6.19 (1.06) 6.10 (1.56)
FEMALE 6.11 (1.01) 5.90 (1.19)
MALE 5.99 (0.96) 6.29 (097)
FEMALE 6.41 (1.10) 5.88 (1.08)
MALE 5.74 (0.79) 5.26 (0.91)
FEMALE 6.14 (0.99) 6.44 (1.01)
MALE 6.07 (1.17) 5.62 (1.27)
FEMALE 6.51 (1.06) 6.21 (1.19)
MALE 6.19 (1.16) 5.63 (1.50)
FEMALE 5.96 (0.98) 5.50 (1.06)
Although no formal hypothesis was made regarding the disability X gender
interaction on the measures of teaching effectiveness, multivariate tests of
significance for disability X gender were significant, Wilks Lamda = .83, F(36,
788.70) = 1.16, p < .05, r| = .06 Univariate follow-up analyses revealed a significant
effect for the interaction of disability X gender on the variable Overall, F(4, 218) =
3.50, p < .05, q = .06. Here, higher mean scores were assigned to the males in the
HIV/AIDS, alcoholic, and no disability conditions. In contrast, higher mean scores
were assigned to the females in the chronic depression and quadriplegia conditions.
Effects of Disability and Gender on Hiring and Salary Recommendations
Separate ANOVAs were performed to examine the effect of disability and gender
on hiring and salary recommendations made by participants. The means for hiring
and salary recommendations, separated by disability and gender, appear in Table 3.5.
Table 3.5. .Means and Standard Deviations for Hiring and Salary Recommendations
HIRING mean (s.d.) SALARY mean (s.d.)
MALE 6.28 (1.43) 5.24 (2.05)
FEMALE 6.41 (1.34) 4.80 (2.15)
MALE 6.46 (1.74) 5.25 (1.98)
FEMALE 6.32 (1.06) 4.82 (1.93)
MALE 6.19 (1.17) 4.65 (1.85)
FEMALE 6.46 (1.17) 5.12 (2.25)
MALE 6.26 (1.06) 5.37 (1.92)
FEMALE 6.60 (1.30) 5.24 (2.52)
MALE 5.88 (1.26) 5.12 (2.01)
FEMALE 6.04 (1.22) 4.93 (2.06)
No formal hypotheses for hiring and salary recommendations were made prior to the
study. Neither ANOVA produced significant results for the effects of disability or
Although not statistically significant, it is interesting to note that male candidates
received higher mean ratings than females on salary recommendations. By disability,
highest salary recommendations were assigned to candidates with quadriplegia,
followed by alcoholism, control condition, HIV/AIDS, and chronic depression.
Post Hoc Analyses
Post hoc analyses were conducted to compare differences between participants
with and without experience with individuals with disabilities. Two extreme groups
were identified: participants who reportedly knew no one with a disability (n = 47)
and participants who reported having a spouse/significant other or immediate family
member with a disability, or, themselves reported having a disability (n = 33). T-tests
were conducted to compare differences between groups on mean ratings of perceived
power across all disability conditions. The results were statistically nonsignificant.
And, finally, t-test comparisons were made to examine differences between groups on
percieved power in each of the disability conditions individually. Here, sample sizes
in the experience group became extremely small, precluding any reliable inferences
from being made from the results.
Jordan, Blake and Aguinis (1996) proposed that personnel specialists fears
regarding the legal complexity of the ADA, the potential for litigation offered under
its provisions, and its impact on personnel selection testing practices are exaggerated.
They suggested that such fears are unfounded if solid professional personnel practices
are followed. While solid personnel practices will go far to avoid litigation, mere
compliance with the ADA will do little to ensure the full integration of persons with
disabilities in the mainstream workforce. For individuals with disabilities, full
integration will be achieved when they are considered an integral part of all
activities, events, organizational functions, levels and life (Hill & Hill, 1994).
Legislation alone cannot produce this level of integration. Consequently, now more
than ever there is the need for research which identifies the attitudinal and perceptual
biases hindering the full workforce participation of individuals with disabilities.
This study represents the first attempt to explore possible biases involving
perceptions of power made of persons with disabilities. Gender and racial stereotypes
are thought to diminish the perceived power of women and minorities, contributing to
their difficulty achieving equal representation at higher organizational levels. Noting
that individuals with disabilities continue to experience employment discrimination,
this study set out to examine whether stereotypes associated with disability would
similarly diminish the perceived power of persons with disabilities.
The results failed to provide support for the hypotheses that perceptions of power
and expectancies would be detrimentally influenced by the presence of a disability.
While there were no significant main multivariate effects for disability or gender on
any of the dependent variables, the data nonetheless revealed some interesting trends.
The means on all of the power and SEEQ variables are higher in the disability
conditions than the control (no disability) condition. Even where univariate follow-
up analyses revealed significant effects, the means were not in the expected direction.
There are at least three possible explanations for these findings.
The first explanation is that no stereotype or bias exists specifically in reference
to power for persons with disabilities. If there were such a stereotype, one would
expect to observe differences on the power bases between the candidates with
disabilities and the nondisabled candidate. The manipulation checks confirmed that
participants in the disability conditions were, in fact, perceiving the manipulation.
However, because the means across all conditions were highly similar, it may be that
a disability stimulus produces no effect on perceptions of power. The extremely
small effect size estimates for the effect of disability on the measures of power
certainly suggest this may be the case. This explanation is further supported by the
fact that the estimated power of the analyses was sufficient to have revealed any
differences among treatment conditions (e.g., estimated power of .90 for disability).
A second possible explanation for the findings is that the information provided in
the resumes about the candidates qualifications was sufficient to dispel any existing
stereotypes or misperceptions associated with disability. It is possible that
participants were influenced more by the achievements of the candidates than by the
disability stimulus. Among the power bases, expert power had the highest mean
rating. This may indicate that although the resumes were designed to portray the
candidates in average terms, to undergraduate students, achieving a doctorate degree
may seem exceptional. Therefore, this individuating information may have been
sufficient to focus the participants attention on the candidates apparent abilities.
A third and highly plausible possibility for the findings is that the results are the
product of contamination by social desirability. It is well established that one of the
most potent determinants of attitudes toward disability is familiarity or contact with
persons with disabilities (Yuker, 1994). It is presumed that positive contact with
persons with disabilities (i.e., contact that disconfirms stereotypical notions) leads to
attitudes that are more favorable. On social distance questions, participants indicated
an average level of familiarity with a person with a generic disability that fell
somewhere between neighbor and close friend] however, the modal response
indicated 17.3% of participants knew no one with a generic disability. Participants
indicated an average level of familiarity with a person with the specific disabilities in
question that fell somewhere between co-worker/classmate and neighbor; but again,
that most respondents (41.5%) knew no one with the specific disability in their
particular treatment condition.
Examination of the correlations between social distance questions and means on
the dependent variables, as presented in Table 4.1, reveals that many of the
correlations are negative or essentially zero, suggesting participants experience with
individuals with disabilities has no (positive) effect on their perceptions. This is
entirely contradictory to social psychological research which suggests that exactly the
opposite is true. This may indicate that participants responses were contaminated by
social desirability or other type of bias.
Similarly, the means for referent power were higher in all of the disability
conditions than the control condition. This would suggest that participants have a
strong desire to be closely associated with the candidates with disabilites, which is
sharply contradictory to disability research involving social distance measures.
- Table 4.1. Correlations Between Social Distance Scores and Means Scores on Measures of Power,
Teaching Effectiveness and Hiring and Salary Recommendations
DEPENDENT MEASURE SOCIAL DISTANCE 1 SOCIAL DISTANCE 2
Coercive -.002 .034
Credibility .010 .032
Expert -.029 -.001
Legitimate .061 .045
Referent .044 .034
Reward -.066 -.070
Assignments -.045 -.039
Breadth -.100 -.083
Enthusiasm -.038 -.092
Exams .004 -.045
Group Interaction .002 -.069
Learning -.119 -.081
Organization -.021 -.115
Rapport -.018 -.136
Overall -.064 -.156*
Hiring -.080 -.084
Salary -.102 -.071
* Statistically significant at p < .05, two-tailed.
Female candidates received significantly higher ratings of credibility than male
candidates. Previous research suggests credibility, much like power, is a gender-
based trait most commonly associated with males (Fandt & Stevens, 1991). This
would imply that women are at an increased disadvantage for establishing themselves
as powerful, credible professionals. However, Fandt and Stevens (1991) also noted
that exposure to women in positions of responsibility may counter stereotypical
attitudes regarding womens ability to perform in gender-atypical roles. It is possible
any possible gender bias on the part of the participants was lessened by previous
experiences with female instructors.
Interestingly, although the results were not significant, females with disabilities
received stronger endorsements than males with regard to hiring but not salary
recommendations. This finding is consistent with recent trends which suggest that
women may be achieving equity in workforce numbers, but still lag behind men in
These findings are encouraging with regard to the employment of individuals
with disabilities in an academic context. The disability manipulation was effective
but did not have a detrimental effect on the dependent variables included in this study.
That is, undergraduate students did not view the candidates with disabilities as having
less power or as being a less effective instructor than the nondisabled candidate.
Further investigation into the topic of power and disability in other contexts with
other subject pools is needed before it can be concluded with any certainty that
disability-related stereotypes regarding power do not exist.
The results of this study provide cautionary evidence about the views of
individuals with disabilities in academic professions. According to the undergraduate
students involved in this study, disability does not affect the perceived power of
college instructors with disabilities. This may suggests that to the extent the success
of college instructors is influenced by their perceived power, individuals with
disabilities are not at an increased disadvantage. The results warrant additional
research on the topic of power and disability in other professional employment
settings. These results also provide interesting information about the perceived power
of individuals with disabilities as a minority group. Research examining the
similiarities and differences in the perceptions of different minority groups may
provide additional information about the processes underlying the perceptual and
attitudinal biases associated with each.
This exploratory study had two major contributions to the scientific literature.
One, it extended the research on social power by including an as of yet overlooked
relationship, that of power and disability. This study attempted to generalize findings
of power involving minority groups to another, although less traditional, minority
Second, the hypotheses examined in this study were drawn, in part, from similar
research on power involving other minority groups. In an effort to extend disability
research, some have proposed that the study of disability can and must be examined
within the context of a minority-group framework (Fine & Asch, 1988; Yuker, 1994).
Yuker (1994) is critical of much of the past research on disability because it seldom
refers to similar research in social psychology such as attitudes toward African-
Americans, women, and other minority groups. Yuker stressed that these types of
research are interrelated and each set of researchers has much to learn from the others.
Additionally, much of disability research has been criticized for taking the
experience of disability as equivalent while ignoring other variables such as degree of
impairment, gender, race, and environmental or contextual variables. While this
study included gender and several different disabilities as individual difference
variables, it may have been more interesting to manipulate environmental or
contextual variables to explore their interactions with disability, perceived power, and
The most obvious limitation of this study is the use of the student sample for
exploring an organizational question. Ultimately, the best methodology for exploring
this research question would be in a naturally occurring setting, with true instructors
with disabilities. Unfortunately, the negligible population of instructors with
disabilities almost necessitates the use of a contrived design. The use of the paper-
people design almost forces participants to rely on stereotypes when forming
impressions because of the limited information it provides (Stone & Colella, 1996).
However, this study attempted to defeat those limitations by framing the study in a
manner intended to be meaningful to the students. This, in essence, defined students
as the population of interest; lessening some concerns about the external validity of
the study (Gordon, Slade, & Schmitt, 1987). And, because the primary intent of this
study was to explore the underlying psychological processes associated with
perceptions of disability, and not to answer questions about actual organizational
decision-making processes, as such, a student sample is adequate (Greenberg, 1987).
Stone and Colella (1996) urged more realistic operationalizations of the construct
of disability be employed. In hindsight, this procedure could have been improved
upon. Perhaps disguising the disability manipulation in a hypothetical response to an
interview question about teaching style or the like would have appeared more
realistic, in addition to providing participants with more meaningful information on
which to base their responses. A more realistic operationalization such as this would
serve to enhance the authenticity of the information.
Disability research is increasingly criticized for using disability as an
independent or manipulated variable (Fine & Asch, 1988; Kerr & Bodman, 1994).
Disability researchers argue that this research simply documents the characteristics or
qualities of persons with disabilities which are considered deficient when compared to
nondisabled persons. Kerr and Bodman (1994) argued that researchers are well
advised to abandon this type of convenient research and begin focusing on
understanding the interaction between the person and the environment and isolating
conditions under which undesirable behaviors (e.g., discrimination, prejudice) may be
changed. In defense of this research project, it may be argued that before problems
may be ameliorated, they must be identified, described, and otherwise fully
understood. This exploratory study was an attempt at information-gathering toward
Suggestions for Future Research
Research utilizing student samples will inevitably continue. However, future
exploration into the issues contained in the study using student samples would greatly
benefit from some methodological improvements. First, the contextual circumstances
of the study may be improved upon by employing an attribute-ambiguity design to
better disguise the study motives. Such a design allows participants to mask hidden
motives in ambuiguity provided by the research design (Snyder, Kleck, Strenta, &
Mentzer, 1979). Thus, well-controlled research designs may uncover true motives or
reactions associated with disability that researchers suspect are present, but are
otherwise masquerading behind socially acceptable response preferences (Snyder, et
Second, captilizing on the use of student samples by pre-selecting participants
with pertinent background characteristics or relevant training or experience may also
lead to more relevant findings (Gordon, Slade, & Schmitt, 1986). Concerns about
external validity are central to the criticisms of student-sample research. However,
the external validity of student-sample research may be improved upon by relying on
trained experimental subjects, conducting extensive debriefing to reveal participant
perceptions, understandings, and meanings associated with the research situation, and
employing subjects with demographic and interest profiles similar to the nonstudent
population to whom researchers wish to generalize (Gordon, et al., 1986). For
instance, future student research may select subjects based on such characteristics as
experience with individuals with disabilities, knowledge of various disabilities, and
perceptions of various disabilities within a specific context.
Third, more realistic operationalizations of disability would also strengthen
future research endeavors. Paper-people manipulations presume that disability is
central to the individuals self-concept and ignore other personal and environmental
factors Abandoning paper-people manipulations acknowledges that disability need
not be the most salient characteristic of an individual with a disability. In the near
future, virtual reality may provide innovative opportunities for researchers (Pierce &
Aguinis, in press) and serve to enliven otherwise flat stimulus materials.
Accepting Kerr and Bodmans (1994) methodological criticisms, future research
should also manipulate true independent variables instead of individual difference
variables and generate research which challenges conventional stereotypes rather than
that which accepts or confirms those stereotypes. For instance, by manipulating
levels of power, and comparing high versus low power conditions, researchers may
explore the processes underlying the perceptions associated with disabilities.
Along the same lines, future research examining the effects of disability on
power should explore whether any differences stem from personal versus position
power. Personal power is power stemming from the attributes of the person; while
position power stems from attributes of the situation (Bass, 1960). Efforts to
understand how the characteristics of the individual and the situation interact to
determine their power may provide more information about how the various bases of
power may be affected by disability.
Ideally, this type of research question is best examined in an organizational
setting with real incumbents and personnel decision-makers in order to assess the true
relationship between disability and power as it occurs naturally in broader
organizational settings. One interesting project may be to conduct longitudinal
studies of persons with disabilities to see whether perceptions of power and
credibility change over time with career advancement and achievements.
More broadly, and perhaps most worthy, is that research which seeks to identify
organizational characteristics and strategies which promote integration and full
utilization of the talents of individuals with disabilities. Instead of identifying
characteristics of persons with disabilities which may be changed or managed
effectively to produce positive outcomes, more valuable effort may be that aimed at
identifying how organizational contexts can be corrected to successfully include the
full participation of individuals with disabilities at all levels.
Sample Stimulus Materials
The Psychology Department is in the process of hiring a clinical psychologist
to fill a vacant teaching position. The Psychology Department strives to recruit and
hire well-qualified individuals who can provide students will valuable learning
experiences. As psychology students, it is likely that you will have interactions with
the hired candidate through your courses or advising sessions. Therefore, your
impressions and opinions of the candidate are very important. For that reason, we
are interested in gathering feedback from the students about each of the
The information we are asking you to provide is intended to assess the
candidates probable classroom performance as an instructor from a students point of
view. Your opinions are an important source of feedback. Often, faculty members
disregard students opinions when making hiring decision. Ultimately, it is the
students who suffer from incorrect hiring decisions. Therefore, your participation
will be very helpful in giving us the students impressions of the candidates teaching
Each candidate was asked to provide a regular vitae, an abbreviated vitae, and
other supporting materials as part of their application. The regular vitae, which are
typically ten or more pages long, and other materials are reviewed by the psychology
faculty. The candidates were asked to make their abbreviated vitae no longer than 2
Enclosed, you will find an abbreviated vitae submitted by one of the
candidates and a brief questionnaire. We ask that you review the vitae and, based on
your impressions, provide responses to approximately 75 questions. You may feel
that some of the questions are similar, or even repetitive. However, each question
provides useful information. Therefore, it is very important that you provide a
response to each item.
Please read the ENTIRE vitae VERY CAREFULLY. After doing so, please
provide a response to each of the questions asked. Your responses are very
important, so PLEASE be thoughtful and read the materials very carefully.
You will NOT be asked to include your name or other identifying information
in your responses. Your identity will remain anonymous and individual responses
will remain completely confidential. It is important that you be AS HONEST AS
POSSIBLE in the responses you provide. You cannot be penalized for your honest
Thank you in advance for your cooperation.
CHRISTOPHER L. RANDALL
Teaching and research position in psychology for a university that is reputable
for both basic and applied research.
Ph.D. Candidate, Clinical Psychology
Department of Psychology, University of Wisconsin
Ph.D. expected May, 1996
Master of Science, Clinical Psychology
Department of Psychology, University of Wisconsin
Bachelor of Science, Psychology
Department of Psychology, Columbia University
Intern, Madison County Hospital
December 1994 December 1995
Duties included monitoring behavior of patients, conducting group exercises,
administering personality measures, and observing individual counseling
Graduate Teaching Assistant, University of Wisconsin
Department of Pychology
August 1994 May 1995
Responsible for teaching undergraduate introductory psychology courses.
Duties included preparing and presenting lectures, writing and grading exams
and term papers, and holding office hours and review sessions.
Graduate Research Assistant, University of Wisconsin
Department of Psychology
May 1995 present
Assisted with research projects investigating abnormal thought processes as
they occur in response to sleep deprivation.
Author of two articles currently in press in professional psychology journals.
Co-author of three papers presented at professional psychology conferences.
Student Member, American Psychological Association
Student Member, Society of Clinical Psychologists
Member, National Coalition for Persons With HIV/AIDS
As a person living with HIV, I am committed to the advancement of basic and
applied research investigating the psychological issues related to HIV/AIDS.
The following 60 items are intended to assess your opinions of the candidates probable classroom
performance. For some of the items, you may feel that you do not have enough information on which
to base your opinion. However, please make your best guess based on the available information.
Also, you may feel that some of the items are similar or repetitive. However, each item provides
useful information. Therefore, please provide a response to every item. Please CIRCLE the number
that most accurately represents your opinion on each of the statements.
Strongly Disagree Disagree THIS INSTRUCTOR Neither Agree Strongly Agree
1. could make me feel personally accepted. 1 2 3 4 5 6 7 8 9
2. tells the truth. 1 2 3 4 5 6 7 8 9
3. could make my coursework difficult for me. 1 2 3 4 5 6 7 8 9
4. could make the course unpleasant. 1 2 3 4 5 6 7 8 9
5. could give me undesirable class assignments. 6. could provide a strong personal reference to future 1 2 3 4 5 6 7 8 9
employers on my behalf. 7. could share with me her/his considerable experience 1 2 3 4 5 6 7 8 9
and/or training. 1 2 3 4 5 6 7 8 9
8. could influence my getting a good grade in the course. 1 2 3 4 5 6 7 8 9
9. could provide me with needed knowledge. 1 2 3 4 5 6 7 8 9
10. is a person who keeps his/her word. I 2 3 4 5 6 7 8 9
11. could make me feel that I have commitments to meet. 12. could give me the feeling that I have responsibilities 1 2 3 4 5 6 7 8 9
to fulfill. 1 2 3 4 5 6 7 8 9
13. could make me feel valued. 1 2 3 4 5 6 7 8 9
14. could make being in class distasteful. 1 2 3 4 5 6 7 8 9
15. could make me feel like (s)he approves of me. 1 2 3 4 5 6 7 8 9
16. could give me good course-related suggestions. 1 2 3 4 5 6 7 8 9
17. could give me extra help on assignments. 1 2 3 4 5 6 7 8 9
18. is accurate in her/his statements. 1 2 3 4 5 6 7 8 9
19. could write a strong letter of recommendation on my behalf. 1 2 3 4 5 6 7 8 9
20. could provide me with sound advice related to the course. 1 2 3 4 5 6 7 8 9
21. could make me recognize that I have tasks to accomplish. 1 2 3 4 5 6 7 8 9
22. does what (s)he says (s)he will do. 1 2 3 4 5 6 7 8 9
23. could make me feel important. 1 2 3 4 5 6 7 8 9
24. makes accurate statements of fact. 25. could make me feel like I should satisfy my course 1 2 3 4 5 6 7 8 9
requirements. 1 2 3 4 5 6 7 8 9
26. is honest. 1 2 3 4 5 6 7 8 9
Neither Agree Strongly
27. I will learn something that I consider valuable l 2 3 4 5 6 7 8 9
28. My interest in the subject will increase as a consequence of this course 1 2 3 4 5 6 7 8 9
29. Required readings/text will be valuable 1 2 3 4 5 6 7 8 9
30. Students will be invited to share ideas and knowledge I 2 3 4 5 6 7 8 9
31. I will leam and understand the subject materials in this course I 2 3 4 5 6 7 8 9
32. Instructor will have a genuine interest in individual students 1 2 3 4 5 6 7 8 9
33. Instructor will be dynamic and energetic in conducting the course 1 2 3 4 5 6 7 8 9
34. Instructor will present points of view other than his/her own when appropriate I 2 3 4 5 6 7 8 9
35. Instructor will enhance presentations with the use of humor 1 2 3 4 5 6 7 8 9
36. Instructors explanations will be clear 1 2 3 4 5 6 7 8 9
37. Instructor will adequately discuss current developments in the field I 2 3 4 5 6 7 8 9
38. Proposed objectives will agree with those actually taught 1 2 3 4 5 6 7 8 9
39. Students will be encouraged to participate in class discussions 1 2 3 4 5 6 7 8 9
40. Course materials will be well-prepared and carefully explained 1 2 3 4 5 6 7 8 9
41. Instructor will make students feel welcome in seeking help/advice outside of class 1 2 3 4 5 6 7 8 9
42. Students will be encouraged to ask questions and will be given meaningful answers 1 2 3 4 5 6 7 8 9
43. Instructor will be enthusiastic about teaching the course 1 -2 3 4 S 6 7 8 9
44. Readings/homework, etc. Will contribute to appreciation and understanding of subject 1 2 3 4 5 6 7 8 9
45. Students will be encouraged to express own ideas and/or question the instructor 1 2 3 4 5 6 7 8 9
46. Methods of evaluating student work will be fair and appropriate 1 2 3 4 5 6 7 8 9
47. Instructor will be friendly to individuals 1 2 3 4 5 6 7 8 9
49. Instructor will give lectures that facilitate taking notes I 2 3 4 5 6 7 8 9
48. Instructor will be adequately accessible to students during office hours or after class I 2 3 4 5 6 7 8 9
50. Instructor will contrast the implications of various theories 1 2 3 4 5 6 7 8 9
51. Instructors style of presentation will hold my interest during class 1 2 3 4 5 6 7 8 9
Strongly Disagree Neither
52. Instructor will present the background or origin of
ideas/concepts developed in class 1 2 3 4 5 6 7 8 9
53. Feedback on examinations/graded materials will be valuable 1 2 3 4 5 6 7 8 9
54. Exams/graded materials will test course content emphasize 1 2 3 4 5 6 7 8 9
55. The course will be intellectually challenging and stimulating 1 2 3 4 5 6 7 8 9
Very Poor Moderrate Good Very
Poor (Average) Good
56. Compared to other college courses I have taken, this course
will be...? 1 2 3 4 5 6 7 8 9
57. Compared to other college instructors I have had, this instructor
will be...? 1 2 3 4 5 6 7 8 9
58. If this candidate were invited to the Psychology Department to present a short lecture in
support of his/her interest in obtaining the position, I would:
59. My recommendation for hiring this person is:
60. Consider that the average starting salary for this position varies between $24,000-
$38,000. If this candidate were hired, what salary would you recommend
(check only one):
______less than $24,000
______ $24,000 $25,999
______ $26,000 $27,999
______ $28,000 $29,999
______ $32,000 $33,999
______ $34,000 $35,999
______ $36,000 $37,999
Please provide the following personal data. We are interested in the group responses to these
questions. We are not interested in individual responses. Your anonymity will be protected.
What is your gender? _____Male _________ Female
What is your age? _____
Status: ____ Freshman __________ Sophomore __________Junior ________ Senior
Ethnicity: Asian American Native American
African American White Hispanic
Other (please specify: )
The closest person I know with a disability is (check only one):
Co Worker or Classmate
Next Door Neighbor
Immediate Family Member
The closest person I know with HTV/AIDS is:
______ CoWorker or Classmate
______Next Door Neighbor
______ Close Friend
______ Immediate Family Member
______ Spouse/Significant Other
THANK YOU FOR YOUR PARTICIPATION!!
Aguinis, H., & Adams, S. R. (1995). Social-role vs. structural models of gender and
influence use in organizations: A strong inference approach. Manuscript
submitted for publication.
Aguinis, H., Nesler, M. S., Quigley, B. M., Lee, S., & Tedeschi, J. T. (1996). Power
bases of faculty supervisors and educational outcomes for graduate students.
Journal of Higher Education, 67, 267-297.
Aguinis, H., Nesler, M. S., Quigley, B. M., & Tedeschi, J. T. (1994). Perceptions of
power: A cognitive perspective. Social Behavior and Personality, 22, 377-384.
Allen, J. G. (1993). Complying with the ADA: A small business guide to hiring and
employing the disabled. New York, NY: Wiley.
American Association for the Advancement of Science (1991). Investing in human
potential: Science and engineering at the crossroads.
Americans With Disabilities Act of 1990. 1990. P.L. 101-336.
Arvey, R. D., & Faley, R. (1988). Fairness in selecting employees. Menlo Park,
Asch, A. (1984). The experience of disability: A challenge for psychology.
American Psychologist, 39,529-536.
Bass, B. M. (1960). Leadership, psychology, and organizational behavior. New
Beatty, M., J., & Zahn, C. J. (1990). Are student ratings of communication
instructors due to easy grading practices? An analysis of teacher credibility
and student-reported performance levels. Communication Education, 39, 275-
Bias in the workplace. (1995, Sept. 3). The Coloradoan, p. El.
Bowe, F. (1992). Adults with disabilities: A portrait. Washington, DC: Presidents
Committee on Employment of People with Disabilities.
Braddock, D., & Bachelder, L. (1994). The glass ceiling and persons with
disabilities. Washington, DC: Glass Ceiling Commission, U. S. Department
Carver, C. S., Glass, D. C., & Katz, I. (1978). Favorable evaluations of blacks and
the handicapped: Positive prejudice, unconscious denial, or social desirability.
Journal of Applied Social Psychology, 8, 97-106.
Esses, V. M., & Beaufoy, S. L. (1994). Determinants of attitudes toward people with
disabilities. Journal of Social Behavior and Personality, 9, 43-64.
Fandt, P. M., & Stevens, G. E. (1991). Evaluation bias in the business classroom:
Evidence relating to the effects of previous experience. Journal of Psychology,
Fine, M., & Asch, A. (1988). Disability beyond stigma: Social interaction,
discrimination, and activism. Journal of Social Issues, 44, 3-21.
Fowler, C. A., & Wadsworth, J. S. (1991). Individualism and equality: Critical
values in North American culture and the impact on disability. Journal of
Applied Rehabilitation Counseling, 22, 19-23.
French, J. R. P., & Raven, B. H. (1959). The bases of social power. In D.
Cartwright (Ed.), Studies in social power (pp. 150-167). Ann Arbor, MI:
Institute for Social Research.
Frymier, A. B., & Thompson, C. A. (1992). Perceived affinity-seeking in relation to
perceived teacher credibility. Communication Education, 41, 388-399.
Gordon, M. E., Slade, L. A., & Schmitt, N. (1986). The science of the sophomore
revisited: From conjecture to emipiricism. Academy of Management Review, 11,
Gutman, A. (1993). EEO law and personnel practices. Newbury Park, CA: Sage.
Hamilton, D. L., Sherman, S. J., & Ruvolo, C. M. (1990). Stereotype-based
expectancies: Effects on information processing and social behavior. Journal of
Social Issues, 46, 35-60.
Hanna, W. J., & Rogovsky, B. (1991). Women with disabilities: Two handicaps
plus. Disability, Handicap & Society, 6, 49-63.
Harlan, S. L., & Berheide, C. W. (1994). Barriers to workplace advancement
experienced by women in low-paying occupations. Washington, DC: Glass
Ceiling Commission, U. S. Department of Labor.
Hill, F. S., & Hill, E. L. (1994). The ADA: Going beyond the law. Academy of
Management Executive, 8, 17-32.
Hinkin, T. R., & Schriesheim, C. A. (1989). Development and application of new
scales to measure the French and Raven (1959) bases of social power. Journal
of Applied Psychology, 74, 561-567.
Johnson, W. G. (1986). The Rehabilitation Act and discrimination against
handicapped workers: Does the cure fit the disease? In M. Berkowitz & M.
A. Hill, Disability and the labor market (pp. 242-261). Ithica, NY: ILR
Press, Cornell University.
Johnson, W. G., & Lambrinos, J. (1985). Wage discrimination against handicapped
men and women. Journal of Human Resources, 20, 264-277.
Jordan, G. A., Blake, A. D., & Aguinis, H. (1996, April). Implications of the
Americans with Disabilities Act for personnel selection. Paper presented at the
meeting of the Society for Industrial and Organizational Psychology, San Diego,
Jussim, L. (1989). Teacher expectations: Self-fulfilling prophecies, perceptual
biases, and accuracy. Journal of Personality and Social Psychology, 57, 469-
Kerr, N., & Bodman, D. A. (1994). Disability research methods: An argument for
the use of Galilean modes of thought in disability research. Journal of Social
Behavior and Personality, 9, 99-122.
Louis Harris & Associates, Inc. (1986). The ICD survey of disabled Americans:
Bringing disabled Americans into the mainstream. New York: Author.
Marsh, H. W. (1987). Students evaluations of university teaching: Research
findings, methodological issues, and directions for future research [Entire
issue]. International Journal of Educational Research, 11(3).
Marsh, H. W. (1982). SEEQ: A reliable, valid, and useful instrument for collecting
students evaluations of university teaching. British Journal of Educational
Psychology, 52, 77-95.
Nesler, M. S., Aguinis, H., Quigley, B. M., & Tedeschi, J. T. (1993). The effect of
credibility on perceived power. Journal of Applied Social Psychology, 23,
Olkin, R., & Howson, L. J. (1994). Attitudes toward images of physical disability.
Journal of Social Behavior and Personality, 9, 81-96.
Pierce, C. A., & Aguinis, H. (in press). The use of virtual reality in organizational
behavior research. Journal of Organizational Behavior.
Pulliam, R. M. (1993). Reactions to AIDS patients as a function of attributions about
controllability and promiscuity. Social Behavior and Personality, 21, 297- 302.
Ragins, B. R. (1989). Power and gender congruency effects in evaluations of male
and female managers. Journal of Management, 15, 65-76.
Ragins, B. R., & Sundstrom, E. (1989). Gender and power in organizations: A
longitudinal perspective. Psychological Bulletin, 105, 51-88.
Richmond, V. P. (1990). Communication in the classroom: Power and motivation.
Communication Education, 39, 181-195.
Richmond, V. P., & McCroskey, J. C. (1984). Power in the classroom II: Power
and learning. Communication Education, 33,125-136.
Romer, K. T., & Whipple, W. R. (1991). Collaboration across the power line.
College Teaching, 39, 66-70.
Rosenthal, R., & Jacobsen, L. (1968). Pygmalion in the classroom: Teacher
expectation and student intellectual development. New York: Holt, Rinehart
Schein, V. E. (1973). The relationship between sex role stereotypes and requisite
management characteristics. Journal of Applied Psychology, 57, 95-100.
Snyder, M. L., Kleck, R. E., Strenta, A., & Mentzer, S. J. (1979). Avoidance of the
handicapped: An attributional ambiguity analysis. Journal of Personality and
Social Psychology, 37,2297-2306.
Snyder, M., Tanke, E. D., & Berscheid, E. (1977). Social perception and
interpersonal behavior: On the self-fulfilling nature of social stereotypes.
Journal of Personality and Social Psychology, 35, 656-666.
SPSS/PC+Base System Users Guide (Version 5.0). (1992). Chicago: SPSS, Inc.
Stone, D. L., & Colella, A. (1996). A model of factors affecting the treatment of
disabled individuals in organizations. Academy of Management Review, 21,
Stone, E. F., Stone, D. L., & Dipboye, R. L. (1992). Stigmas in organizations: Race,
handicaps, and physical unattractiveness. In K. Kelley (Ed.), Issues, theory, and
research in Industrial/Organizational psychology (pp. 385-457). Amsterdam:
Elseveir Science Publishers.
Tabachnick, B. G., & Fidell, L. S. (1996). Using multivariate statistics (3rd ed.).
New York: Harper Collins.
Tringo, J. L. (1970). The hiearchy of preference toward disability groups. Journal of
Special Education, 4, 295-306.
U. S. Bureau of the Census. (1993). Poverty in the United States: 1992. Current
Population Reports, Consumer Income, Series P60-185, 91-96.
U. S. Department of Labor (1991). A report on the glass ceiling initiative.
Washington, DC: U. S. Government Printing Office.
Word, C. O., Zanna, M. P., & Cooper, J. (1974). The nonverbal mediation of self-
fulfilling prophecies in interracial interaction. Journal of Experimental Social
Psychology, 10, 109-120.
Yuker, H. E. (1994). Variables that influence attitudes toward people with
disabilities: Conclusions from the data. Journal of Social Behavior and
Personality, 9, 3-22.