Rating Performance or Asserting Status: A Test of a Social Dominance Theory of
Supervisor Bias in Performance Ratings
Jone L. Pearce
The Paul Merage School of Business
University of California, Irvine
Irvine, CA 92697-3125
949-824-6505
[email protected]
Qiumei Jane Xu
College of Business and Management
Northeastern Illinois University
Chicago, IL 60625
773-442-6172
[email protected]
Keywords: social dominance theory, status, diversity, demography
A previous version was presented in the Gender and Diversity in Organizations Division’s 2007
Program at the Academy of Management, Philadelphia, PA. The authors wish to thank Joerg
Dietz, Kristin Behfar, Greg Bigley, and Kenji Klein for comments on earlier drafts.
2
Rating Performance or Asserting Status: A Test of a Social Dominance Theory of
Supervisor Bias in Performance Ratings
ABSTRACT
Supervisory ratings of employees’ performance remain the dominant means of assessing job
performance in organizations despite decades of research documenting their biases. We propose
that a major reason why so little progress in improving the accuracy of these ratings has been
made is the theoretical mis-specification of the reasons for systematic supervisory rating bias.
We extend Social Dominance Theory to explain similarity bias in supervisory ratings and test
this theory against the reigning theory of homophily in a strong inference test. The contradictory
hypotheses are tested against one another in a field study of 358 supervisor-subordinate dyads in
ten organizations. Using hierarchical linear modeling to eliminate explanations at the
organizational and supervisory levels, we find that supervisors’ biases in rating their
subordinates’ contextual and task performance are explained by Social Dominance effects, and
that none of the homophily bias hypotheses are supported. The implications of these results for
addressing bias in supervisory performance ratings, as well as the theories of status in
organizational behavior are discussed.
3
Rating Performance or Asserting Status: A Test of a Social Dominance Theory of
Supervisor Bias in Performance Ratings
Supervisors’ ratings of their subordinates’ job performance are as pervasive as they are
notoriously biased, and this is a problem. Biased appraisals hurt both organizations and
employees. For organizations, inaccurate appraisals undermine organizational performance; if
employee performance is misidentified, flawed information means adjustments cannot be made
and the organization’s own performance is undermined (Banks & Murphy, 1985). Widespread
knowledge of rating bias also impedes attempts to reward for performance (Pearce, 1987).
Inaccurate performance ratings undermine employees’ sense of fairness (Folger, 1992; Fulk,
Brief & Barr, 1985; Kanfer, Sawyer, Earley & Lind, 1987), their trust in their organizations
(Mayer & Davis, 1999), and lead to rewards and promotions going to less able people (Folger,
Konovsky & Cropanzano, 1992). Yet, as important as this problem is, it has defied solution,
despite fifty years of research and experimentation.
We suggest that one reason for this has been a misunderstanding of the reasons for
supervisors’ ratings biases. Heretofore, scholars have assumed that performance rating
inaccuracy results from supervisors’ perceptual errors, or their faulty decision making processes,
or similarity-attraction biases (e.g., Feldman, 1981; Smith, 1996). Yet, these approaches have
been of limited usefulness. Recently Folger et al. (1992) developed an innovative way to work
around supervisory inaccuracy. They proposed that supervisor’s ratings should be treated as
disputes over the allocation of resources and so subject to due-process procedures. Adding dueprocess procedures, such as adequate notice and a fair hearing, do improve employees’
perceptions that the ratings are fairer and more accurate (Taylor, Tracy, Renard, Harrison &
4
Carroll, 1995). This work is an important practical step forward, but did not eliminate reliance on
biased supervisors’ performance ratings, and so a need to have a better understanding of why
they are inaccurate.
Supervisors’ Biased Ratings of Performance
Supervisors’ ratings of their subordinates’ job performance are the primary way
employee job performance is measured for the vast majority of jobs in most organizations
(Landy & Farr, 1980). This reliance on supervisors’ ratings continues despite extensive evidence
of their inaccuracy (Pulakos & Wexley, 1983). Supervisory ratings of performance persist
because, except under rare circumstances, the alternatives are worse. There are few jobs with
comprehensive objective measures of performance, because the employment contract itself is
based on a need for flexibility in the face of uncertainty about future demands (Pearce, 1987).
This is one reason why the rigid use of objective measures of performance is notorious for
creating goal displacement, and for punishing innovation and adaptation to change (Meyer,
1994). Most jobs are too complex, require too much management of the unexpected, and rely on
improvisation to have employee performance judged solely on objective measures of
performance.
Early approaches to the problem of supervisory ratings inaccuracy treated it as a problem
of either supervisors’ flawed perceptions of reality, or of their poor assessments of their
subordinates’ individual differences in performance. Both of these approaches drew on
substantial research in psychology on person perception and psychometrics, but proved unable to
do much to improve supervisors’ ratings accuracy in practice. More systematic measurement
systems and better supervisory training in assessing individual differences based on these
5
theoretical explanations have had only limited effects on ratings accuracy (Jacobs, Kafry &
Zedeck, 1980). Others have tried to improve the accuracy of supervisor’s ratings by
triangulating them with others’ ratings. However, peer ratings are even less accurate than
supervisors’ ratings (Thornton, 1980; Motowidlo, 1982), and can create demoralizing politicking
and retaliation (DeNisi, Randolph & Blencoe, 1983). For most jobs, most of the time, an
employees’ performance needs to be assessed via a retrospective assessments of job effort,
judgment, and performance after the fact (Pearce, 1987). The persistence of supervisory ratings
of performance, however flawed, after so much investment of scholarship and practice in trying
to improve their accuracy strongly suggests that there is no viable alternative to them. Their
inaccuracy remains an important practical problem that can be addressed by better theory.
There is strong evidence that supervisor ratings inaccuracy is not random, but the result
of systematic biases. The best documented is similarity bias: the more similar subordinates are to
their supervisors the higher the subordinates’ performance ratings (Pulakos & Wexley, 1983;
Baltes, Bauer & French, 2007). Research finds that evaluators of all kinds tend to rate those
more similar to themselves more positively than those who are dissimilar (e.g., Golightly,
Huffman & Byrne, 1972; Wexley, Alexander, Greenwalt & Couch, 1980). For example, Graen,
Novak, & Sommerkamp (1982), found that supervisors gave more positive evaluations to more
similar employees. Supervisors rated their subordinates’ contextual and task performance lower
the more demographically different those subordinates are (e. g. (Ferris, Judge, Chachere, &
Liden, 1991; Tsui & O'Reilly, 1989; Tsui, Porter, & Egan, 2002). Similarity bias in supervisors’
ratings of their subordinates’ performance also has widespread effects on organizational behavior.
Wakabayashi, Graen, & Graen (1988) reported that subordinates more similar to their
supervisors received more frequent promotions. Anit’s (2003) research established that
6
demographic differences between supervisors and subordinates were negatively associated with
supervisor’s inclusion of subordinates in participative decision-making. McCain, O'Reilly, and
Pfeffer (1983) and Wagner, Pfeffer, and O Reilly (1984) found that subordinates with
demographically dissimilar supervisors were more likely to leave their jobs. While this
similarity bias in supervisory performance ratings is undisputed, too many scholars have simply
assumed that it is explained by supervisory homophily bias, an assumption that has been resistant
to contradictory evidence.
Homophily Bias
The dominant theoretical explanation for similarity biases in supervisory ratings has been
the similarity-attraction explanation (Byrne, 1971). There is a large body of research in social
and industrial/organizational psychology demonstrating that people tend to be more attracted to,
and prefer to be with, those who are more similar to themselves; another term for this is
homophily bias.1 Byrne (1971) stated that individuals who are similar on demographic
dimensions have had more common life experiences, beliefs and values, and so are more
attracted to the similar members of their work unit. Glaman, Jones and Rozelle (1996) found
that demographically similar co-workers liked one another more and preferred similar others as
coworkers. Homophily bias results from demographic similarities that invoke an attraction
dynamic whereby demographically similar individuals accentuate the positive attributes of each
other and derive positive social identities from identifying with similar others.
This dynamic has been tied to Social Identity and Self-Categorization Theories. It is
proposed that individuals develop positive self-definitions that are tied to their social identities
(Messick & Massie, 1989; Turner, 1987), and classify themselves and others into social
7
categories on the basis of demographic characteristics such as ethnicity, age, gender, education
and tenure in order to build attractive social identities. In this way, others are more attractive to
the individual if they have demographic profiles that are consistent with the categories that the
individual has chosen for self-categorization (Turner, 1987). Those evoking similarity-attraction
or self-categorization theory assume that attraction drives the observed demographic similarity
effects. We shall use the term homophily bias to refer to both explanations.
Homophily bias has been used as a retrospective explanation for a wide range of
observed similarity biases in organizational settings (e.g., Chatman & Sparato, 2005; Ibarra,
1995; Pelled, Eisenhardt & Xin, 1999; Tsui, Egan, & O Reilly, 1992). Focusing on supervisors’
similarity biases in rating their subordinates, Tsui, Xin and Egan (1995) argued that demographic
differences between supervisors and subordinates can undermine supervisor-subordinate
interpersonal attraction, supervisor perceptions of employee performance, and subordinate
perceptions of managerial supportiveness all assumptions that homophily bias is causing the
effect. However, there is growing empirical evidence that homophily bias is not universal, and
can be trumped by concerns for status.
Social Dominance Theory and Bias
Social Dominance Theory proposes that many who are of lower status are not
homophilous, but prefer those of higher status (Sidanius, 1993; Pratto, Sidanius & Levin, 2006).
This effect is used by political scientists and sociologists to explain racial and gender
discrimination, and why those with lower status often accept their disadvantaged positions.
Social Dominance Theory posits that human societies tend to be organized as group-based social
hierarchies with some groups enjoying greater social status and power than other groups (Pratto,
8
Sidanius & Levin, 2006, among others). Those in higher status groups are motivated to
subjugate groups of lower status, while members of lower status groups provide willing support
for their own subjugation.
Although Social Dominance Theory operates at many levels, the empirical research has
been dominated by research on a measure of individual differences called Social Dominance
Orientation (SDO). Umphress, Smith-Crowe, Brief, Dietz & Watkins (2007) applied this
individual differences measure to organizational questions and found that homophily bias was
indeed more common among those in high-status groups, especially among those high in SDO.
Umphress et al.’s (2007) research contradicts the universal homophily bias explanation so
common in organizational demography research, and demonstrated the value of Social
Dominance Theory in organizational settings.
The component of this comprehensive theory that is useful in understanding supervisory
ratings biases is that those in dominant positions will be homophilous (favor in-group members)
while those in subordinate or lower status positions tend not to identify as strongly with those in
the same positions, and will display less homophily (this effect was first proposed by Tajfel &
Turner, 1979; for a recent review see, Sidanius, Pratto, van Laar & Levin, 2004). Hinkle &
Brown (1990) found that lower status individuals did not favor their own ingroup. Levin (1992)
found that some low-status individuals favored the higher status outgroup rather than those in
their own ingroup, and lower status members are less likely to identify with their own ingroup
than are higher status members (Sidanius, Pratto & Rabinowitz, 1994).
According to Social Dominance Theory, status striving and defense, not homophily bias,
explain most of the variance in ingroup identification and intergroup perceptions (Ridgeway,
Boyle, Kuipers & Robinson, 1998; Sidanius, et al., 2004). Homophily does not work uniformly,
9
but is most characteristic of those holding high-status positions who wish to defend their
privileged places.
There is empirical evidence from the organizational literature that a preference for
associating with those with higher status can overwhelm homophily. Tsui, Egan and O’Reilly’s
(1992) found that men responded more negatively than did women to working in groups
dominated by the other gender. Chatman and O’Reilly (2004) intuited that because men
traditionally have occupied a higher status, men would be more sensitive to the status
incongruence of working in women-dominated work groups. They found that men were more
likely to want to transfer out of work groups with greater proportions of women. However,
contrary to their predictions, but consistent with Social Dominance Theory, lower status women
did not favor either women dominated or men dominated work groups. Vecchio (1993) tested a
status incongruence explanation for supervisor-subordinate relational age differences against
similarity-attraction, social competition and loyalty explanations. Unfortunately, his mixed
results did not consistently support any of these explanations. Finally, Perry et al. (1999)
conducted a direct test comparing the homophily and status incongruence explanations for
similarity bias effects. Again, their results were conflicting: while the similarity-attraction
explanations were not supported, neither were most of the status incongruence explanations.
These studies all demonstrated that homophily bias is not universal in organizational
settings, but they do not seem to have successfully unseated the attachment to homophily
explanations in organizational behavior (see Williams & O’Reilly, 1998). We believe there are
three reasons for this failure of contradictory data to shift attachment to a preferred theory. First,
the minor role status has played in theories of organizational behavior means that many scholars
remain unaware of empirically supported theory that better accounts for their empirical data.
10
Second, there is a bias in the field of organizational behavior to build increasingly complex
omnibus theories rather than conduct strong inference tests of competing theoretical arguments.
Third, there has been an absence of a strong competing theory that could better account for their
results (Kuhn, 1970). Social Dominance Theory can provide the theoretically coherent
explanation that is most consistent with the available empirical data that homophily bias is not
universal, and so provide more support in addressing supervisory biases in performance ratings.
Minor Role of Status Striving in Organizational Behavior
Status has had a relatively minor role in organizational behavior, most often treated as a
variable of secondary interest. The concept of status dates back to the earliest days of sociology
(Simmel ([1908] 1950; Weber, [1914] 1978), and has been much studied in that field and in
social psychology (e.g., Blau, 1994; Harvey & Consalvi, 1960). Given this history it isn’t
surprising that there have been numerous definitional debates over the meaning of status and
how it might differ from such ideas as estates, (Weber, 1978), class (Marx, [1894] 1967), power
(Pearce, Ramirez & Branyiczki, 2001); as well as the debate about status as a subjective
evaluation and as an objective structural reality (Wegener, 1992). Building on Pearce et al.’s
(2001) review of the varied definitions of status we adopt their definition of status as one’s
relative respected social standing with reference to a particular social grouping or hierarchy.
Mirroring early interest in status in sociology and social psychology, status was
prominent in early organizational behavior theory; for example, Barnard ([1938] 1968) suggested
that status (what he called prestige) was an important inducement in organizations, and Vroom
(1964) proposed that seeking status is one of the major reasons why people work. Yet since this
early work, studies of the effects of status seeking or defense on motivation, performance and
11
affective responses in organizations have been few. Only recently have scholars in
organizational behavior begun to focus their attention on the role of status and status striving in
organizational behavior. This growing interest is important, because other social scientists have
helped to demonstrate the importance of status in social settings, with Troyer and Younts (1997)
proposing that a primary motivation in social interaction is the avoidance of status loss, and
Waldron (1988) arguing that it has a fundamental biological basis.
There is no question that having high status is a source of many advantages. To cite just
a few examples: Those with higher status are listened to more, receive more deference from
others, and are perceived as having more power (Berger, Zelditch & Anderson, 1966; Sheets &
Braver, 1999). People with higher status have better health and longevity (Marmot, 2004), and
their decisions have more legitimacy (Keashly & Newberry, 1995). Those randomly assigned a
higher status outperform those randomly assigned a lower status (Lovaglia, Lucas, Houser, Thye
& Markovsky, 1998), are able to command higher prices in markets (Ball, Eckel, Grossman &
Zame, 2001), and will lose their business partners abandon when they lose status (Jensen, 2006).
Stuart, Hoang and Hybels (1999) demonstrated that the higher the status of a firm’s affiliates, the
shorter its time to initial public stock offering, and the higher the firm’s valuation once it went
public. D’Aveni (1996) found that the higher the status of employees’ university degrees the
better their mobility opportunities.
Because of its advantages, higher status usually is actively sought in social settings
(Harvey & Consalvi, 1960). Having more money (Weber 1914 1978; Nee, 1996), displaying
status-objects (Sundstrom & Sundstrom, 1986), having more education (Bidwell & Friedkin,
1988), working in more prestigious occupations (Kanekar, Kolsawalla & Nazareth, 1989), and
belonging to elite organizations (D’Aveni & Kesner, 1993) all lead to greater status, and so
12
status attainment is seen as a primary motive for pursuing money, status-objects, education,
certain occupations, and membership in elite voluntary associations. In more organizationfocused research, better task performance can be a means to obtaining higher status
(Shackelford, Wood & Worchel, 1996). What is more, those who lose status at work tend to be
less satisfied, have lower self-esteem, and report more work-related depression (Schlenker &
Gutek, 1987). Elsbach and Kramer (1996) found that when an organization’s status was
denigrated, its members experienced dissonance and acted to emphasize those dimensions on
which their organization had higher rank. Pearce, et al. (2001) suggested that relative status
incongruence was the primary motivator of executives’ corporate strategies in transition
economies. There is overwhelming evidence that high status is actively pursued and vigorously
defended because it provides so many advantages -- power, wealth, dispensations, and longevity,
within organizations or in any other social setting. The importance and centrality of status to
social life is reflected in literature, mass entertainment, and is a pillar supporting the popular
leadership and management self-help industries. Given the power of status striving and defense
in social settings, its central place, and its popular awareness, it is surprising that status does not
have a more prominent role in theories of organizational behavior. There are no handbooks of
research on organizational status, no journal special issues on status in organizations. Status
certainly is mentioned, but usually as one of many possible factors in a list of factors, as, for
example, the single facet of Social Status Satisfaction from Dawis, Lofquist and Weiss’s (1968)
Minnesota Satisfaction Questionnaire.
Quite recently, status is beginning to gain greater attention in the field of organizational
behavior, but at present the work is scattered across subfields such as discrimination (e.g.,
Umphress, et al., 2007), within-group conflict (Phillips & Thomas-Hunt, 2007), workplace
13
harassment (Berdahl, 2007), and organizational learning (Perretti & Negro, 2006), among others.
From across the wide range of organizational behavior topics scholars are increasingly turning to
status to account for empirical puzzles. For example, Pearce et. al (2001) found that those
executives who were willing to undertake large organizational change projects were seeking to
increase a recently degraded status, while those who made no attempts at change occupied highstatus positions. Phillips and Thomas-Hunt (2007) suggested that status distance within task
groups can account for the conflicting findings about the effects of diversity on conflict.
The scattered attention to status in organizational behavior is costly. First, the diversity
of subfields in which status is introduced means that scholars working in these fields focus on
their specific problems, and while they find that status and status striving are useful ways to
think about their problems, they remain unaware of each others’ work and so cannot build on it
and develop our understanding of status in organizations. For example, Van der Vegt,
Bunderson and Oosterhof (2006) deplore their finding that those group members who have the
most expertise receive the most help from their fellow group members, when those with less
expertise needed it more. Those familiar with the status literature, and in particular, that expertise
bestows status and those with more status receive more attention and assistance, would not be
surprised by this finding. Second, the lack of sustained theoretical conversation about the role of
status in organizational behavior means that many empirical phenomena that might be better
explained as status effects are explained in other, less powerful ways. Here we make the case
that status threat, not homophily bias, better explains similarity bias in supervisory ratings.
14
Strong Inference Tests of Competing Theories
A second reason why disconfirming evidence that homophily bias accounts for
supervisor similarity biases is the preference in the field for building omnibus multifaceted
theories, rather than in building and testing theories that seek to identify the most powerful
explanatory constructs. Over forty years ago Platt (1964) argued that science best progressed
through the use of what he called strong inference tests of competing theories. This is Francis
Bacon’s method of inductive inference. It involves proposing plausible alternative hypotheses
and then conducting a crucial test of one against the other. Platt (1964) suggested that theories
which do not exclude anything are not theories at all, and theories that try to predict everything,
predict nothing at all. He argued that the biological and physical scientific specialities he studied
advanced best, not over lifetimes, but through theoretical disagreements and the conduct of headto-head tests of alternative theoretical accounts. Strong inference testing in organizational
behavior is unusual (Notable exceptions include Latham, Erez and Locke, 1988, Vecchio, 1993,
among a few others). Instead, ever more comprehensive theories have become dominant in
organizational behavior.
An omnibus approach to theory development limits theoretical understanding in several
ways. First, if there is no effort to empirically compare different theoretical accounts, we cannot
correct our mistakes and advance knowledge. Some pruning of mis-specified theories is
necessary, and one approach to theoretical pruning is to test one proposed theory directly against
another plausible alternative. In addition, the theoretical relationships among so many variables
in complex models are rarely carefully considered. Too often a plausible single sentence for a
proposed relationship is offered for each bivariate relationship, with no consideration of how
15
these concepts work together as a whole. Are any of them substitutes for one another? How,
exactly do they mutually influence one another? Such a plethora of variables and relationships
overwhelms scholars’ capacities to process theoretical information. Finally, theories that have
survived tests against alternatives are particularly important for a practical field such as
organizational behavior: the consumers of research need to know what to do first. Those wishing
to take action must make trade offs and so need to know what is actually causal, and what can be
left in the background as insufficiently powerful to make it a priority.
This case for more focused theories and strong inference tests of competing claims runs
the risk of being confused with the current disputes about paradigm heterogeneity. Pfeffer’s
(1993) suggestion that knowledge advancement, resources and careers in organizational studies
have been harmed by the absence of comprehensive and cumulative theory building ignited a
storm of controversy (e.g., Perrow, 1994; Cannella & Paetzold, 1994) that continues (e.g., Glick,
Miller, & Cardinal, 2007). Certainly, Platt (1964) writes from a perspective in the biological and
physical sciences, and so would seem to fall within the camp of those who call for more
homogeneity in our theoretical and methodological paradigms. However, this is not the case.
We are not arguing for one best way to develop theories, conduct research, or for one
overarching paradigm for the field. To the contrary, we are arguing for more theories and more
tests of differing theoretical accounts. Strong inference tests surely are not the only way to
advance our understanding of organizational behavior, but they a valuable approach that
currently is under-used in the field. To this end, we offer the following alternative theory of
similarity-bias in supervisory ratings and test it against the homophily bias explanation.
A Social Dominance Theory of Supervisory Ratings Bias
16
Supervisors occupy positions of higher status and dominance than do their subordinates
in most organizations. Greater occupational prestige (Kanekar, Kolsawalla, & Nazareth, 1989)
leads to higher status, as does holding power over others (Levine & Moreland, 1990). Both of
these are associated with higher hierarchical position in most organizational settings. In fact,
Driskell and Salas (1991), among others, have used hierarchical organizational position as the
measure of status in their organizational research. Of course, there are exceptions, particularly
for professional and artistic work, where the most knowledgeable or talented person’s superior
task performance can trump hierarchical position (Pearce, et al., 2001). However, in most
organizational settings, we would expect supervisors to be considered as holding higher status
than their subordinates. It follows, then, that any actions or characteristics, including
demographic ones, serving to compete with and threaten supervisors’ status claims to dominance
relative to their subordinates would be distressing for supervisors. There is substantial evidence
that those who have inconsistent status roles in organizations experience greater stress and strain
(Bacharach, Bamberger & Mundell, 1993). Social Dominance Theory proposes that supervisors,
as members of the dominant group, would be expected to be more subject to homophily bias, and
so any other competing claim for dominance would be threatening to them. Social Dominance
Theory finds that dominant group members seek to maintain their dominant positions, and we
propose that this means they tend to see any subordinate who is a member of a higher competing
status group as providing a challenge to their dominant positions.
There is substantial evidence that demography sends status signals; the status
implications of differing races, ages, genders, and occupations all have been documented. We
can conduct our strong inference test pitting our Social Dominance Theory of Supervisor Ratings
Bias against the homophily one using two distinct demographic characteristics – gender and age.
17
Using two different demographic groups avoids the risk of inadvertently drawing general
theoretical conclusions from relationships that could be particular to the meanings attached to
any one categorical demographic characteristic.
Gender-based Status Incongruence. There is a large body of research finding that men
hold higher status than women (Cohen & Zhou, 1991; Major, McFarlin & Gagnon, 1989;
Thomas-Hunt & Phillips, 2004). When jobs are dominated by women they are perceived to be
less complex (Miller, 1992), and have lower status (Kanekar, Kolsawalla & Nazareth, 1989).
Women are less likely to engage in the higher status actions of interrupting, dominating, and
confidently contributing their expertise than are men (see Thomas-Hunt & Phillips, 2004, for a
review). Further, women are perceived as less competent than men, and must display superior
task competence to overcome this status-enhancing stereotype (Driskell, Olmstead & Salas,
1993). Finally, women are expected to be emotionally sensitive and deferential (the actions of
those with lower status) while managers are expected to be dominating and forceful (the actions
of those with higher status). Thus, it isn’t surprising that many have found incongruence in the
different roles of manager and woman (Byron, 2004; Eagly & Carli, 2007; Kanter, 1977). This
suggests that there will be potential status threat to (lower status) women occupying roles
supervising (higher status) men. When women supervise men the higher status gender role of
their male subordinates threatens their own dominant positions. Social Dominance Theory
research finds that the more insecure a dominant status is the more likely the dominant person is
to show similarity favoritism (Brewer & Kramer, 1985).
For this reason, we expect to find that women who supervise men would tend to rate their
subordinates’ contextual and task performance lower than they would rate their non-status
threatening women subordinates. Status threat, not homophily, would account for gender-based
18
supervisors’ biases in performance ratings. If homophily best explained supervisory bias in
performance ratings, we would expect men supervising women also to rate their women
subordinates’ performances more negatively than that of their male subordinates, since both are
equally dissimilar. We also don’t expect struggles for status dominance when women supervise
women subordinates or men supervise men subordinates, because the dominance of the
supervisor’s position would be uncontested. This allows Social Dominance and homophily bias
theories to be tested against one another.
Hypothesis 1a: Supervisors with gender-based status-threatening subordinates (women
supervising men) will give them poorer contextual performance ratings than they will
non-threatening status congruent subordinates (men supervising women and samegender supervisors and subordinates).
Versus,
Hypothesis 1b: Supervisors with subordinates of dissimilar gender will give them poorer
contextual performance ratings than those supervisors rating subordinates of the same
gender.
Hypothesis 2a: Supervisors with gender-based status-threatening subordinates (women
supervising men) will give them poorer task performance ratings than they will nonthreatening status-congruent subordinates (men supervising women and same-gender
supervisors and subordinates).
Versus,
19
Hypothesis 2b: Supervisors with subordinates of dissimilar gender will give them poorer
task performance ratings than those with supervisors rating subordinates of the same
gender.
Age-based Status Incongruence. The research indicating that slow or stalled career
progression reflects status degradation (Lawrence, 1988) provides an opportunity to compare
Social Dominance Theory and homophily explanations for age-based demographic differences.
While greater age may reflect greater social status in many cultures, in most Western developed
countries we would not expect a direct age effect on relative social status. Rather, the
organizational age literature conducted in the United States indicates that it is holding an ageinappropriate job that leads the incumbent to experience reduced status (Lawrence, 1988). Those
who are older are expected to have more experience and expertise and this enhances and
supports the higher status of the supervisory role (Vecchio, 1993). We propose that having a
subordinate much older than themselves could be viewed as a threat to the dominant status of
supervisors. In most organizational settings, it is normative that subordinates should be younger
than their supervisors (Perry, et al., 1999). As above, Social Dominance Theory research finds
that the more insecure a person with dominant status is, the more likely the person is to show
similarity bias (Brewer & Kramer, 1985).
If the homophily explanation is best, we would expect to find that supervisors with much
younger subordinates would be as negatively biased against them as supervisors with much older
subordinates, since both are equally dissimilar. However, Social Dominance Theory predicts
that only the younger supervisors of noticeably older subordinates would demonstrate homophily
bias in their ratings of their subordinates’ job performance.
20
Hypothesis 3a: Supervisors with age-based status-threatening subordinates (supervisors
noticeably younger than subordinates) will give them poorer contextual performance
ratings than they will their non-status threatening subordinates (subordinates and
supervisors of similar ages or subordinates with older supervisors).
Versus,
Hypothesis 3b: Supervisors with age-dissimilar subordinates (supervisors noticeably
older or younger than subordinates) will give them poorer contextual performance
ratings than they will subordinates of similar ages.
Hypothesis 4a: Supervisors with age-based status-threatening subordinates (supervisors
noticeably younger than subordinates) will give them poorer task performance ratings
than they will to their non-status threatening subordinates (subordinates and supervisor
of similar ages or subordinates with older supervisors).
Versus,
Hypothesis 4b: Supervisors with age-dissimilar subordinates (supervisors noticeably
older or younger than subordinates) will give them poorer task performance ratings than
they will to their subordinates of similar ages.
METHODS
The sample for this study consisted of 358 supervisor-subordinate dyads. These are employees
working in ten companies in the computer manufacturing, electronics, telecommunications, retail
21
industries, and food services located in the United States. Subordinates and supervisors provided
data independently via surveys. Surveys were mailed to the respondents’ work addresses and
were coded to enable the matching of data from the members of the dyad. Respondents were
provided stamped, self-addressed envelopes to return completed surveys. Confidentiality of the
survey was guaranteed. The supervisors had a response rate of 64% and provided data on their
subordinates’ Contextual and Task Performances, and on their own demographic characteristics.
Ninety-seven percent of the supervisors classified themselves as White and reported a mean
organizational tenure of 11.1 years. For each supervisor three subordinates were randomly
selected to receive questionnaires (60% response rate), with employees providing data on their
personal demographics; 87% of the subordinates classified themselves as White and reported a
mean organizational tenure of 8.9 years.
Measures
The independent variables are age and gender, with the supervisor-subordinate dyads
grouped by whether or not they are status-incongruent or status-congruent (called Status
Incongruence) and by whether or not they are relationally dissimilar or relationally similar
(called Dissimilarity). Categorical Gender is measured by a dichotomous variable. A value of
“1” was used to denote men and “2” women. Categorical Age was measured in years. The
means, standard deviations and intercorrelations for all study variables appear in Table 1.
_____________________
Insert Table 1 About Here
_____________________
22
Incongruence and dissimilarity. Status Incongruence-Gender consists of supervisorsubordinate dyads with women supervisors and men subordinates (coded 1, n=10), with all other
dyads categorized as gender status congruent (coded 0, n=348). Supervisor-subordinate dyads
are gender relationally dissimilar if a man supervises a woman or a woman supervises a man
(n=79), all other dyads are considered gender relationally similar (n=279). Dissimilarity-Gender
was coded with a value of 1, and similarity was coded as 0.
For Status Incongruence-Age, there is no theoretical guidance on what age differences
would become noticeably incongruent. We believed Vecchio (1993) and Perry et al.’s (1999)
use of the simple arithmetic differences in ages would classify some dyads which are not
meaningfully different in age (only a year or two) as either dissimilar or status incongruent, and
could account for their non-supportive results. Thus, relying on common sense and the
distribution of the differences as guidelines, three age groups were created. The first group is
composed of supervisors five years younger than their subordinates. The second group contains
supervisors five years older than their subordinates, and the third group has dyads with the
supervisors and subordinates within five years of age. The Status Incongruent-Age group is the
first group, coded 1 (n= 59), with supervisors at least five years younger than their subordinates,
with all other dyads considered age status congruent, coded 0 (n=299). The Dissimilar-Age
dyads are those with supervisors at least five years younger than their subordinates and those
with supervisors at least five years older than their subordinates, coded 1 (n=223). All other
supervisor-subordinate dyads are considered age relationally similar, and are coded 0 (n=125).
Contextual Performance. Subordinate Contextual Performance was measured using nine
items drawn from the existing literature (e.g. Graham, 1986; O’Reilly & Chatman, 1986; Smith
et al., 1983). This scale includes items like “is willing to speak up”, “calls management attention
23
to dysfunctional activities”, “does not keep doubts about a work issue to him/herself”, “
expresses opinions honestly”, “informs management of unproductive policies”, “makes
suggestions to improve organization”, and “makes innovative suggestions to improve
department”. The supervisors stated the extent to which they agreed that their subordinates
exhibited the actions described by the item, on a seven-point agree-disagree Likert scale,
converted to five-point scale here. The Cronbach reliability coefficient for this scale is .93.
Task Performance. Subordinates’ task performance consists of six items asked of
supervisors and used to assess core task performance, originally described in [author identifying
citation removed], where it was found to be sufficiently distinct from subordinate Contextual
Performance to merit separate analyses. Task Performance items focused on quantity, quality
and efficiency of job performance, rated on a seven-point Likert agree-disagree scale, converted
to a five point scale for these analyses. The internal consistency reliability of this scale in this
sample is .96.
Analyses. The hypotheses are at the supervisor-subordinate dyadic level, however these
dyads are nested in two other levels: organizations, and supervisory raters who evaluated the
performance of one, two or three subordinates. Therefore, hierarchical linear models (HLM)
were used in order to be sure that there were controls for any organizational or supervisor effects
in testing these dyadic level hypotheses. The HLM was performed following Garson’s (Garson,
n.d.) and used the SPSS "Mixed models-linear”. Hofmann (1997) suggested that in general,
relatively large sample size are required for HLM models and with fewer groups, we can obtain
sufficient power by having more groups within each organization. Although ten organizations
are included in this study, we have more than three hundred supervisor-subordinate dyads, which
is acceptable for us to do the analysis. As can be seen in Table 1, the supervisors in Organization
24
6 tended to rate their subordinates lower than did supervisors in the other nine organizations,
further evidence that controlling for effects at the organizational level are necessary. For
example, to test Hypothesis 1a and 1b, hierarch linear models were performed with Contextual
Performance as the outcome variable. At the dyad level, Dissimilar-Gender and Status
Incongruence-Gender were used as predictors of Contextual Performance. At the supervisor
level, supervisors’ identification numbers were incorporated as supervisor-level predictors of the
intercept term. At the organizational level, organizations’ identification numbers were used as
predictors of the organization-level intercept terms. Similar procedures were performed for the
other hypothesis tests.
Further, as can be seen in Table 1, younger supervisors tended to give their subordinates
lower performance ratings, and gender is often associated with Contextual Performance.
Therefore, controls for age in the gender tests, and gender in the age tests in all of the HLM
models, to insure that effects of categorical demography are not mistakenly interpreted as
supporting either of the tested theories. Finally, to aid in interpretation we have included all
paired-comparison tests of group means, applying Levene’s tests for inequality of variances to all
tests to control for the group-size skew in these samples.
RESULTS
All of the “a” hypotheses make predictions derived from Social Dominance Theory for similarity
biases in supervisory ratings, while the “b” hypothesis make contradictory predictions based on
homophily bias explanations. Hypothesis 1 concerns gender effects on Contextual Performance
while Hypothesis 2 concerns gender effects on Task Performance. Table 2 presents the results of
25
the hierarchical linear models for gender on both contextual and task performance. Here, and in
the paired comparison tests reported in Table 3, we can see that for Contextual Performance
(Model 1) the prediction for the status threat explanation was supported (Hypothesis 1a) with a
significant negative coefficient for Status Incongruence-Gender but not for Dissimilarity-Gender
(Hypothesis 1b). As shown in Table 2, no significant coefficients were found at the supervisor
or organization levels in either Model 1 or 2. Women supervisors rated their male subordinates
as having significantly lower Contextual Performance than did men rating either men or women
subordinates. In contrast, once the status-threat dyads were removed homophily bias has no
significant effects. Unfortunately, there were no significant gender-based relational demography
effects for subordinates’ Task Performance here (See Table 2), preventing the comparative test
of gender-based status threat vs. homophily explanations for supervisor ratings of Task
Performance (Hypotheses 2a vs. 2b). Nevertheless, for those tests of gender-based performance
ratings biases that were possible, the status-threat hypothesis was supported and the homophily
one was not.
__________________________
Insert Table 2, & 3 About Here
__________________________
Hypothesis 3 addresses the Social Dominance Theory and homophily predictions of agebased effects on Contextual Performance, and Hypothesis 4 makes those predictions for Task
Performance. Table 4 has the results of the hierarchical linear model analyses for the age
differences hypotheses. We can see that the significant negative coefficients support both status
threat explanations for both Contextual Performance (Model 3, Hypothesis 3a) and Task
26
Performance (Model 4, Hypothesis 4a), while none of the homophily coefficients were
significant, once the status-threat dyads were removed (Hypotheses 3b and 4b). Nor were there
any significant coefficients at the supervisory or organizational levels. As can be seen from the
paired comparison tests of group means in Table 5, supervisors rated their subordinates more
than five years older than themselves as significantly lower in Contextual and Task Job
Performance than did supervisors rating subordinates about their own age or who were dissimilar
but younger.
___________________________
Insert Table 4 & 5 About Here
___________________________
DISCUSSION AND CONCLUSION
We developed a Social Dominance Theory of Supervisory Ratings Bias that better accounted for
the existing empirical data on when and why individuals will tend to have similarity biases. It
was tested against the currently dominant universal homophily bias explanation of supervisory
performance ratings by placing predictions from the homophily explanation in head-to-head
strong inference tests against status threat explanations. We found that, for both gender and age,
status threat significantly accounted for the supervisory ratings, while the homophily bias
explanations received no support, once status threatening dyads were removed. With a sample of
358 supervisor-subordinate dyads in ten different organizations the tests were sufficiently
powerful to find support for the non-status-confounded homophily bias effects in supervisory
performance ratings if they existed. These findings provide strong support for the superiority of
27
the Social Dominance Theory-derived status threat explanation for systematic similarity biases in
supervisors’ ratings of their subordinates’ job performance over the current homophily
explanations.
Here we found that supervisors systematically rated those subordinates whose higher
demographic status threatened their own dominant hierarchical status lower in both contextual
and task performance. While those lower ratings could possibly be based on relatively lower real
subordinate performance, it is hard to imagine why men working for women should
systematically perform more poorly than women working for women, or that noticeably older
subordinates should perform more poorly at their jobs than subordinates of a similar age or
younger than their supervisors. Similarly, we did not directly measure either experienced statusthreat or homophily here, but tested for patterns of data that would be predicted from each of
these explanations. Given the difficulty of getting honest self-reports of both status seeking and
homophily bias, this study, as did others in the literature, did not directly seek to obtain these
measures. We did demonstrate that the similarity bias in supervisory ratings that has been
widely assumed apparently had been masking a bias against those whose higher demographic
status threatened the supervisor’s superior status, resulting in a mis-specification of this bias as
one resulting from similarity-attraction or homophily bias. Social Dominance Theory provides a
theoretically coherent account for these, and other, empirical reports of demographic similarity
effects, however, with additional research on status in organizational settings it is possible that
anomalous empirical results will surface, suggesting the need for another theory. However, at
present this study indicates that status-threat, not homophily, best accounts for the welldocumented systematic similarity bias in supervisory performance ratings of their subordinates.
28
This theoretical correction has important practical implications. When this bias was misunderstood as universal homophily bias practical interventions focused on diversity sensitivity,
seeking to increase awareness of homophily bias and so reduce it. This research suggests that
such training would be useless in reducing similarity bias in supervisory ratings: supervisors are
reacting to perceived threats to their dominant status, not to any aversion to someone who is
demographically different. Social Dominance Theory emphasizes that not all differences are
equal. Ignoring the potential status threats and status striving underlying various differences (to
say nothing of behavior) is a prescription for training impotence. This research suggests that
training involving discussions of status striving and status defense, and how they may
consciously or unconsciously color perceptions and actions, while potentially difficult, hold
more promise for effectiveness than abstract admonitions to respect differences.
These tests confirm the claim that status striving and defense are powerful explanatory
variables in organizational behavior. Status defense better accounted for bias in supervisor
ratings of their subordinates’ contextual and task performance than did homophily bias. A focus
on status also is more consistent with previous research showing that homophily bias is not
universal. As Social dominance Theory proposes, status will usually trump homophily, such that
those in dominant status positions tend to devalue the performance of subordinates who display
competing claims to high status. Supervisors exist in social environments, and are as concerned
with maintaining their dominant roles as are people in other social settings. Social Dominance
Theory better accounts for the observed systematic demographic biases in supervisor’s
performance ratings, and provides a more nuanced account of why these biases occur.
Supervisors’ authority is fragile (Sayles, 1989), and they are alert to potential threats to their
dominant status positions. While the effect sizes deriving from demographic threats to their
29
dominance reported here may be small, this may be because demographic threats to status are
relatively weak in these organizations. More direct threats to supervisors’ dominant status
positions, such as insubordination or politicking, would probably be much more threatening.
Social Dominance Theory suggests that supervisors will be sensitive to threats to their dominant
positions and that they will act to protect and enhance their status from any threats to it.
Organizations are rife with status striving for dominance and Social Dominance Theory promises
to be a fruitful source of explanatory power for a variety of organizational behaviors.
Regarding homophily explanations of demography effects, this test demonstrates that
much previous organizational research was conducted in a way that masked the more
theoretically complete and empirically supported reason for these effects – threats to status
dominance. The fact that none of the homophily bias hypotheses were supported, once the status
incongruent effects were removed, combined with the well-established status differences among
different demographic groups, strongly suggest that previous assumptions that homophily is the
best explanation for relational diversity effects more generally appear to be misplaced. Certainly,
the data used in other relational demography studies could be reanalyzed to conduct strong
inference tests of status threat vs. homophily as was done here. In any case, these results indicate
that it is no longer appropriate to assume that relational demographic diversity effects are
primarily driven by homophily bias. Social Dominance Theory provides a credible and powerful
alternative explanation, something that would not have been clear if it was simply added to an
omnibus theory of demographic diversity effects.
This strong inference approach to theory testing, rather than any all-inclusive omnibus
theory, highlighted the more powerful effects of status threat. As such, we hope it illustrated the
potential dangers of all-inclusive theories to the advancement of our knowledge in the field of
30
organizational behavior. Omnibus theories tend to treat all constructs as equally important, and
so may mask false or redundant theoretical claims. By assuming that plausible alternative
hypotheses should to be added to existing theory, rather than pitted against one another, the field
risks stagnating intellectually, and misleading those who wish to put our research to practical use.
Future research in organizational behavior might profitably seek to deconstruct omnibus theories
and subject their multiple relationships to competing strong-inference tests. We might better
advance our understanding of organizational behavior if we more often asked, What is most
important?, rather than, What might explain additional variance?
In conclusion, these results help to demonstrate what has long been apparent to
sociologists and social psychologists –- social status is important to people, and the seeking of
higher status and avoidance of status degradation drives much behavior, in workplaces as in
other social settings. This work supports the growing attention to status in the diverse subspecializations of organizational behavior. Sociologists and social psychologists have long
considered it a central explanatory variable, and the more powerful status threat effects
demonstrated here suggests it could have a prominent place in a wider range of diversity and
organizational behavior theories.
31
FOOTNOTE
1
Ironically, the first systematic examination of homophily effects in the social sciences
(Lazerfeld & Merton, 1954) distinguished between status homophily and value homophily; with
current work in organizational behavior focused only on the effects of value homophily.
32
REFERENCES
Anit, S. 2003 Relationships of participative leadership with relational demography variables: A
multi-level perspective. Journal of Organizational Behavior, 24: 1003-1019.
Bacharach, S. B., Bamberger, P., & Mundell, B. 1993 Status inconsistency in organizations:
From hierarchy to stress. Journal of Organizational Behavior, 14: 21-36.
Ball, S., Eckel, C., Grossman, P. J., & Zame, W. 2001 Status in markets. The Quarterly Journal
of Economics, 116: 161-188.
Barnard, C. I. 1968 The functions of the executive. Cambridge, MA: Harvard University Press.
First published 1938
Baugh, S. G. & Graen, G. B. 1997 Effects of team gender and racial composition on perceptions
of team performance in cross-functional teams. Group and Organization Management, 22: 366383.
Berdahl, J. L. 2007 Harassment based on sex: Protecting social status in the context of gender
hierarchy. Academy of Management Review, 32: 641-658.
Berger, J., Cohen, B. P., & Zelditch, M. 1966 Status characteristics and expectations states. pp.
29-46. In J. Berger, M. Zelditch, & B. Anderson (Eds.) Sociological theories in progress, Vol. I.
Boston: Houghton-Mifflin.
Bidwell, C. E. & Friedkin, N. E. 1988 The sociology of education. In N. J. Smelser (Ed.)
Handbook of Sociology. 449-472 Newbury Park, CA: Sage Publications.
Blau, P. M. 1994 Structural contexts of opportunities. Chicago: University of Chicago Press.
Brewer, M. B. & Kramer, R. M. 1985. The psychology of intergroup attitudes and behavior.
Annual Review of Psychology, 36: 219-243.
Byrne, D.1971 The attraction paradigm. New York: Academic Press.
Byron, K. 2005 Male and female manager's nonverbal ability and others' ratings. Paper
presented at the Annual Meeting of the Society for Industrial/Organizational Psychology, Los
Angeles, CA.
Cannella, A. A. & Paetzold, R. L. 1994 Pfeffer’s barriers to the advance of organizational
science. Academy of Management Review, 19: 331-341.
33
Chatman, J. A. & O'Reilly, C. A. III 2004 Asymmetric reactions to work group sex diversity
among men and women. Academy of Management Journal, 47: 193-208.
Chatman, J. A., Polzer, J. T., Barsade, S. G., & Neale, M. A. 1998 Being different yet feeling
similar: The influence of demographic composition and organizational culture on work processes
and outcomes. Administrative Science Quarterly, 43: 749-780.
Chatman, J. A. & Spataro, S. E. 2005 Using self-categorization theory to understand relational
demography-based variations in people's responsiveness to organizational culture. Academy of
Management Journal, 48: 321-331.
Chattopadhyay, P. 1999 Beyond direct and symmetrical effects: The influence of demographic
dissimilarity on organizational citizenship behavior. Academy of Management Journal, 42: 273.
Chattopadhyay, P., Tluchowska, M. & George, E. 2004 Identifying the ingroup: A closer look at
the influence of demographic dissimilarity on employee social identity. Academy of
Management Review, 29: 180-202.
Cohen, B. P. & Zhou, X. 1991 Status processes in enduring work groups. American Sociological
Review, 56: 179-188.
D'Aveni, R. A. 1996 A multiple-constituency, status-based approach to interorganizational
mobility of faculty and input-output competition among top business schools. Organization
Science, 7: 166-189.
D'Aveni, R. A. Kesner, I. F. 1993 Top managerial prestige, power and tender offer response: A
study of elite social networks and target firm cooperation during takeovers. Organization
Science, 4: 123-151.
Dawis, R. V., Lofquist, L. H., & Weiss, D. J. 1968 A theory of work adjustment [Revision]
Minnesota Studies in Vocational Rehabilitation. No. 23. Minneapolis: University of Minnesota.
DeNisi, A. S., Randolph, W. A., & Blencoe, A. G. 1983 Potential problems with peer ratings.
Academy of Management Journal, 26: 457-464.
Driskell, J. E., Olmstead, B., & Salas, E. 1993 Task cues, dominance cues, and influences in task
groups. Journal of Applied Psychology, 78: 51-60.
Driskell, J. E. & Salas, E. 1991 Group decision making under stress. Journal of Applied
Psychology, 76: 473-478.
Eagly, A. H. & Carli, L. L.2007 Through the labyrinth: The truth about how women become
leaders. Boston: Harvard Business School Press.
34
Elsbach, K. D. & Kramer, R. M. 1996 Members' responses to organizational identity threats:
Encountering and countering the Business Week rankings. Administrative Science Quarterly,
41: 442-476.
Erikson, J., Pugh, W. & Gunderson, E. 1972 Status congruency as a predictor of job satisfaction
and life stress. Journal of Applied Psychology, 56: 523-525.
Feldman, J. M. 1981 Beyond attribution theory: Cognitive processes in performance appraisal,
Journal of Applied Psychology, 66, 127-148.
Ferris, G. R., Judge, T. A., Chachere, J. G., & Liden, R. C. 1991 The age context of
performance-evaluation decisions. Psychology and Aging, 6: 616-622.
Folger, R., Konovsky, M. A., & Cropanzano, R. 1992 A due process metaphor for performance
appraisal. In B. M. Staw & L. L. Cummings (Eds.) Research in Organizational Behavior, 13:
129-177.
Fulk, J., Brief, A. P. & Barr, S. H. 1985 Trust-in-supervisor and perceived fairness and accuracy
of performance evaluations. Journal of Business Research, 13: 299-313.
Garson, G. D. n.d. Title of Topic. Statnotes: Topics in Multivariate Analysis. Retrieved
06/26/2008 from http://www2.chass.ncsu.edu/garson/pa765/statnote.htm.
Glaman, J., Jones, A., & Rozelle, R. 1996 The effects of co-worker similarity on the emergence
of affect in work teams. Group & Organization Management, 21: 192-215.
Glick, W. H., Miller, C. C., & Cardinal, L. B. 2007 Making a life in the field of organization
science. Journal of Organizational Behavior.
Goffman, E. 1959 The presentation of self in everyday life. Garden City, NY: Doubleday.
Golightly, C., Huffman, D. M., & Byrne, D. 1972 Liking and loaning. Journal of Applied
Psychology, 56: 521-523.
Graen, G. B., Novak, M. A., & Sommerkamp, P. 1982 The effects of leader-member exchange
and job design on productivity and satisfaction: Testing a dual attachment model.
Organizational Behavior and Human Decision Processes, 30: 109-131.
Greenberg, J. 1988 Equity and workplace status: A field experiment. Journal of Applied
Psychology, 73: 606-613.
Harvey, O. J. & Consalvi, C.1960 Status conformity to pressure in informal groups. Journal of
Abnormal and Social Psychology, 60: 182-187.
35
Hodge, R. 1962 The status consistency of occupational groups. American Sociological Review,
27: 36-343.
Hofmann, D.A. 1997 An overview of the logic and rationale of hierarchical linear models.
Journal of Management, 23: 723-744
Hyman, M. D. 1966 Determining the effects of status inconsistency. Public Opinion Quarterly,
30: 120-129.
Ibarra, H. 1995. Race, opportunity, and diversity of social circles in managerial networks.
Academy of Management Journal, 38: 673-703.
Jackson, S.E., Brett, J.E. Sessa, V.I., Cooper, D.M. Julin, J.A., & Peyronnin, K. 1991 Some
differences make a difference: Individual dissimilarity and group heterogeneity as correlates of
recruitment, promotions, and turnover. Journal of Applied Psychology, 76: 675-689.
Jacobs, R., Kafry, D. & Zedeck, S. 1980 Expectations of behaviorally anchored rating scales,
Personnel Psychology, 33: 595-640.
Jehn, K., Northcraft, G., & Neale, M. 1999 Why differences make a difference: A field study of
diversity, conflict, and performance in work groups. Administrative Science Quarterly, 44: 741763.
Kanekar, S., Kolsawalla, M. B., & Nazareth, T. 1989 Occupational prestige as a function of
occupant's gender. Journal of Applied Social Psychology, 19: 681-688.
Kanfer, R., Sawyer, J., Earley, P. C., & Lind, E. A. 1987 Fairness and participation in evaluation
procedures: Effects on task attitudes and performance. Social Justice Research, 1: 235-249.
Kanter, R. M. 1977 Men and women of the corporation. New York: Basic Books.
Keashly, L. & Newberry, J. 1995 Preferences for fairness of intervention: Influence of thirdparty control and third-party status and conflict setting. Journal of Social and Personnel
Relationships, 12: 277-293.
Kuhn, T. S. 1970 Structure of scientific revolutions (2nd ed.) Chicago: University of Chicago
Press.
Landy, F. J. & Farr, J. L. 1980 Performance ratings. Psychological Bulletin, 87: 72-107.
Landy, F. J. & Trumbo, D. A. 1980 The psychology of work behavior (rev.ed.) Homewood, IL:
Dorsey.
36
Latham, G. P., Erez, M., & Locke, E. A. 1988 Resolving scientific disputes by the joint design of
crucial experiments by the antagonists: Application to the Erez-Latham dispute regarding
participation in goal setting. Journal of Applied Psychology, 73: 753-772.
Lawrence, B. S. 1988 New wrinkles in the theory of age: demography, norms, and performance
ratings. Academy of Management Journal, 31: 309-337.
Lazerfeld, P. & Merton, R. K. 1954 Friendship as a social process: A substantive and
methodological analysis.. In M. Berger, T. Abel, & C. H. Page (Eds.) Freedom and control in
modern society. 18-66 New York: Van Nostrand.
Leitner, D. W. & Sedlacek, W. E. 1976. Characteristics of successful campus police officers.
Journal of College Student Personnel, 17: 304-308.
Levin, S., Sidanius, J., Rabinowitz, J. L., & Frederico, C. 1998. Ethnic identity, legitimizing
ideologies, and social status: A matter of ideological asymmetry. Political Psychology, 19: 373404.
Levine, J. M. & Moreland, R. L. 1990 Progress in small group research. Annual Review of
Psychology, 41: 585-634.
Lovaglia, M. J., Lucas, J. W., Houser, J. A., Thye, S. R., & Markovsky, B. 1998 Status processes
and mental ability test scores. American Journal of Sociology, 104: 195-228.
Major, B., McFarlin, D. B., & Gagnon, D. 1989Overworked and underpaid: On the nature of
gender differences in personal entitlement. Journal of Personality and Social Psychology, 47:
1399-1412.
Marmot, M. 2004 The status syndrome. New York: Henry Holt.
Marx, K. 1967 Capital (Vol. 3, edited by F. Engels). New York: International Publishers. First
published in 1894.
Mayer, R. C. & Davis, J. H. 1999. The effect of the performance appraisal system on trust for
management: A field quasi-experiment. Journal of Applied Psychology, 84: 123-136.
McCain, B. E., O'Reilly, C. A. I., & Pfeffer, J. 1983 The effects of departmental demography on
turnover: The case of a university. Academy of Management Journal, 26: 626-641.
Messick, D., & Massie, D. 1989 Intergroup relations. Annual Review of Psychology, 40: 45-81.
Meyer, M. W. 1994 Measuring performance in economic organizations. In N. J. Smelser & R.
Swedborg (Eds.) Handbook of economic sociology: 556-580. Princeton, NJ: Princeton
University Press.
37
Miller, J. 1992 Gender and supervision: The legitimation of authority in relationship to task.
Sociological Perspectives, 35: 137-162.
Motowidlo, S. J. 1982 Relationship between self-rated performance and pay satisfaction among
sales representatives. Journal of Applied Psychology, 67, 209-213.
Ng, S. H. 1982 Power and intergroup discrimination. In H. Tajfel (Ed.) Social identity and
intergroup relations (pp. 179-206). Cambridge: Cambridge University Press.
Ng, S. H. 1984 Social psychology and political economy. In H. Tajfel (Ed.) The social
dimension: European developments in social psychology. Vol. 2. (pp. 624-645), Cambridge:
Cambridge University Press.
Pearce, J. L. 1987 Why merit pay doesn't work: Implications from organizational theory. In D.
B. Balkin and L. R. Gomez-Mejia (Ed.), New perspectives in compensation. Englewood Cliffs,
NJ: Prentice Hall, 169-178.
Pearce, J. L., Ramirez, R. R., & Branyiczki, I. 2001 Leadership and the pursuit of status: Effects
of globalization and economic transformation. In W. S. Mobley & M. McCall (Eds.), Advances
in Global Leadership Vol. 2, pp. 153-178.
Pelled, L. H. 1996 Relational demography and perceptions of group conflict and performance: A
field investigation. International Journal of Conflict Management, 7: 230-247.
Pelled, L. H., Eisenhardt, K. M., & Xin, K. R. 1999 Exploring the black box: An analysis of
work group diversity, conflict, and performance. Administrative Science Quarterly, 44: 1-28.
Perry, E. L., Kulik, C. T., & Zhou, J. 1999 A closer look at the effects of subordinate-supervisor
age differences. Journal of Organizational Behavior, 20: 341-357.
Perretti, F. & Negro, G. 2006 Filling empty seats: How status and organizational hierarchies
affect exploration versus exploitation in team design. Academy of Management Journal, 49:
759-777.
Pfeffer, J. 1993 Barriers to the advancement of organizational science: Paradigm development as
a dependent variable. Academy of Management Review, 18: 599-620.
Phillips, K. W. & Thomas-Hunt, M. C. 2007 Garnering the benefits of conflict: The role of
diversity and status distance in groups, pp. 37-55. In K. Behfar and L. Thompson (Eds.) Conflict
in organizational teams: New directions in theory and practice. Evanston, IL: Northwestern
University Press.
Platt, J. R. 1964 Strong inference. Science, 146: 347-353.
38
Pratto, F., Sidanius, J., & Levin, S. 2006 Social dominance theory and the dynamics of
intergroup relations: Taking stock and looking forward. European Review of Social Psychology,
17: 271-320.
Pratto, F., Stallworth, L., Sidanius, J., & Siers, B. 1997 The gender gap in occupational role
attainment: A social dominance approach. Journal of Personality and Social Psychology, 72:
37-53.
Pulakos, E. D. & Wexley, K. N. 1983 The relationship among perceptual similarity, sex, and
performance ratings in manager-subordinate dyads. Academy of Management Journal, 26: 129139.
Ridgeway, C. L., Boyle, E. H., Kuipers, K. J., & Robinson, D. T. 1998 How do status beliefs
develop? The role of resources and interactional experience. Mercian Sociological Review, 63:
331-350.
Riordan, C., & Shore, L. 1997 Demographic diversity and employee attitudes: Examination of
relational demography within work units. Journal of Applied Psychology, 82: 342-358.
Sachdev, I. & Bourhis, R. Y. 1991. Power and status differentials in minority and majority group
relations. European Journal of Social Psychology, 21: 1-24.
Shackelford, S., Wood, W., & Worchel, S. 1996 Behavioral styles and the influence of women in
mixed-sex groups. Social Psychology Quarterly, 59: 284-293.
Sayles, L. R. 1989 Leadership: Managing in real organizations. (2nd ed.) New York: McGraw
Hill.
Schlenker, J. A. & Gutek, B. A. 1987 Effects of role loss on work-related attitudes. Journal of
Applied Psychology, 72: 287-293.
Sheets, V. L. & Braver, S. L.1995 Organizational status and perceived sexual harassment:
Detecting the mediators of a null effect. Personality and Social Psychology Bulletin, 25: 11591171.
Sidanius, J., Pratto, F., Sinclair, S., & van Laar, C. 1996 Mother Teresa meets Genghis Khan:
The dialectics of hierarch-enhancing and hierarchy attenuating career choices. Social Justice
Research, 9: 145-170.
Sidanius, J., Pratto, F., van Laar, C., & Levin, S. 2004 Social Dominance Theory: Its agenda and
method. Political Psychology, 25: 845-879.
Simmel, G. 1950 The secret and the secret society.. In K. H. Wolff (Trans.) The sociology of
Georg Simmel. 305-376 Glencoe, IL: The Free Press. First published 1908.
39
Smith, D. E. 1986 Training programs for performance appraisal: A review. Academy of
Management Review, 11, 22-40.
Stuart, T. E., Hoang, H., & Hybels, R. C. 1999 Interorganizational endorsements and the
performance of entrepreneurial ventures. Administrative Science Quarterly, 44: 315-349.
Sundstrom, E. & Sundstrom, M. G. 1986 Work places: The psychology of the physical
environment in offices and factories. New York: Cambridge University Press.
Taylor, M. S., Tracy, K. B., Renard, M. K., Harrison, J. K., & Carroll, S. J. 1995 Due process in
performance appraisal: A quasi-experiment in procedural justice. Administrative Science
Quarterly, 40: 495-523.
Teahan, J. E. 1975. A longitudinal study of attitude shifts among black and white police officers.
Journal of Social Issues, 31: 47-56.
Thomas-Hunt, M. C., & Phillips, K. W. 2004 When what you know is not enough. Personality
and Social Psychology Bulletin, 30: 1585-1598.
Thornton, G. C. III 1980 Psychometric properties of self-appraisals of job performance.
Personnel Psychology, 33: 263-271.
Troyer, L., & Younts, C. W. 1997 Whose expectations matter? The relative power of first- and
second- order expectations in determining social influence. American Journal of Sociology,
103: 692-732.
Tsui, A. S., Egan, T. D., & O Reilly, C. A. III. 1992 Being different: Relational demography and
organizational attachment. Administrative Science Quarterly, 37: 549-579.
Tsui, A. S. & Gutek, B. A. 1999 Demographic differences in organizations: current research
and future directions. Lanham, Md.: Lexington Books.
Tsui, A. S. & O Reilly, C. A. III 1989 Beyond Simple Demographic Effects: The Importance Of
Relational Demography in Superior-Subordinate Dyads. Academy of Management Journal, 32:
402-423.
Tsui, A. S., Pearce, J. L., Porter, L. W., & Tripoli, A. M. 1997 Alternative approaches to the
employee-organization relationship: Does investment in employees pay off? Academy of
Management Journal, 40: 1089-1121..
Tsui, A. S., Porter, L. W., & Egan, T. D. 2002 When both similarities and dissimilarities matter:
Extending the concept of relational demography. Human Relations, 55: 899-929.
40
Turner, J. 1987 Rediscovering the social group: A self-categorization theory. Oxford:
Blackwell.
Umphress, E. E., Smith-Crowe, K., Brief, A. P., Dietz, J., & Watkins, M. B. 2007 When birds of
a feather flock together and when they do not: Status composition, social dominance orientation,
and organizational attractiveness. Journal of Applied Psychology, 92: 396-409.
Van der Vegt, G. S., Bunderson, J. S., & Oosterhof, A. 2006 Expertness diversity and
interpersonal helping in teams: Why those who need the most help end up getting the least.
Academy of Management Journal, 49: 877-893.
van Laar, C. Sidanius, J., Rabinowitz, J., & Sinclair, S. 1999 The three R’s of academic
achievement” Reading riting, and racism. Personality and Social Psychology Bulletin, 25: 139151.
Vecchio, R. P. 1993 The impact of differences in subordinate and supervisor age on attitudes and
performance. Psychology and Aging, 8: 112-119.
Vecchio, R. P. & Bullis, R. C. 2001 Moderators of the influence of supervisor-subordinate
similarity on subordinate outcomes. Journal of Applied Psychology, 86: 884-896.
Vroom, V. H. 1964 Work and motivation. New York: Wiley.
Wagner, W. G., Pfeffer, J., & O Reilly, C. A. III 1984 Organizational demography and turnover
in top-management groups. Administrative Science Quarterly, 29: 74-92.
Wakabayashi, M., Graen, G. B., & Graen, M. 1988 Japanese management progress: Mobility
into middle management. Journal of Applied Psychology, 73: 217-227.
Waldron, D. A. 1998 Status in organizations: Where evolutionary theory ranks. Managerial and
Decision Economics, 19: 505-520.
Weber, M. 1978 Economy and society: An outline of interpretive sociology (Vol. 2). Berkeley,
CA: University of California Press. First published 1914.
Wegener, B. 1992 Concepts and measurement of prestige. Annual Review of Sociology, 18:
253-280.
Wesolowski, M. A. & Mossholder, K. W. 1997 Relational demography in supervisorsubordinate dyads: Impact on subordinate job satisfaction, burnout, and perceived procedural
justice. Journal of Organizational Behavior, 18: 351-362.
41
Wexley, K. N., Alexander, R. A., Greenwalt, S. P., & Crouch, M. A. 1980 Attitudinal
congruence and similarity a related to interpersonal evaluations in manager-subordinate dyads.
Academy of Management Journal, 23: 320-330.
Wiersema, M. E. & Bird, A. 1993 Organizational demography in Japanese firms: Group
heterogeneity, individual dissimilarity, and top management team turnover. Academy of
Management Journal, 36: 996-1025.
Williams, K. Y. & O'Reilly III, C. A. 1998 Demography and diversity in organizations. In B. M.
Staw & R. I. Sutton (Eds.) Research in Organizational Behavior, 20: 77-140. Stamford, CT:
JAI Press.
Zenger, T. R. & Lawrence, B. S. 1989 Organizational demography: The differential effects of
age and tenure distributions on technical communication. Academy of Management Journal,
32: 353-376.
Table 1
______________________________________________________________________________________________________________________________________________________________________________________________
Means, Standard Deviations, and Intercorrelations among Study Variables
1. Subordinate
Gendera
2. Subordinate Age
3.Supervisor
Gendera
4. Supervisor Age
Mean
s.d.
1
2
1.32
.47
-
39.65
8.77
-.18**
-
1.16
.37
.47**
-.02
3
4
5
6
7
8
9
10
43.50
6.69
.03
.14**
-.11*
-
.03
.17
.06
-.03
.20**
.07
-
6. Org D2
.05
.22
.11*
-.06
.07
.02
-.04
-
7. Org D3
.04
.20
.09
-.16**
-.01
-.07
-.04
-.05
-
8. Org D4
.06
.24
.13*
-.04
.02
.05
-.04
-.06
-.05
-
9. Org D5
.05
.21
.01
.07
.01
-.10
-.04
-.05
-.05
-.06
-
10. Org D6
.15
.35
-.13*
-.16**
-.14**
-.19**
-.07
-.09
-.09
-.10
-.09
-
11. Org D7
.08
.27
.09
.08
.16**
.13*
-.05
-.07
-.06
-.07
-.07
-.12*
12. Org D8
.07
.26
.07
.13*
.15**
.19**
-.05
13. Org D9
.43
.50
-.20**
.14**
-.25**
-.05
-.15**
.03
.17
-.12*
.02
.39**
-.11*
15. DissimilarGenderc
.22
.42
.62**
-.17**
-.05
16. Status
Incongruence-Aged
.16
.37
-.13*
.64**
17. Dissimilar- Agee
12
13
14
15
16
17
18
-
5. Org D1
14. Status
IncongruenceGenderb
11
-
-.06
-.06
-.07
-.06
-.11*
-.08
-.20**
-.18**
-.22**
-.19**
-.36**
-.25**
.07
-.04
-.04
-.04
.12*
-.02
.04
-.05
.03
.09
.10
.10
.05
-.26**
-.03
-.07
-.06
-.05
-.24**
-
.01
.09
-.08
-
-.05
-.03
.01
-.06
.32**
-
.08
-.05
-.07
.00
.16**
.11*
-.11*
-
.62
.49
.03
-.13*
-.02
.22**
-.04
-.01
.13*
.00
-.07
-.07
-.05
.05
.01
.03
.08
.35**
-
18. Contextual
Performance
3.76
.68
-.01
-.01
-. 06
.22**
.03
.05
.04
-.00
-.04
-.14**
.07
.06
.02
-.12*
.02
-.19**
-.09
-
19. Task
Performance
3.83
.70
.08
-.09
-.00
.15**
,03
.04
.02
-.01
-.00
-.11*
-.03
.12*
-.01
-.02
.07
-.19**
-.04
.72**
n = 358 * p < .05 ** p < .01.
a
1 = men, 2 = women;
b
1= men subordinates with women supervisors, 0 = all other supervisor-subordinate pairs
c
1 = men with women supervisors and women with men supervisors, 0 = all other supervisor-subordinate pairs
d
1 = subordinates more than five years older than their supervisors, 0 = all other supervisor-subordinate pairs
e
1 = subordinates more than five years older and more than five years younger than their supervisors, 0 = all other supervisor-subordinate pairs
43
Table 2
______________________________________________________________________
Results of Model Analysis for Gender-based Status Dissimilarity and Incongruence
on Contextual and Task Performance
Model 1
Model 2
Contextual Performance
Task Performance
Variable
Estimate
SE
Estimate
SE
Intercept
Age
Dissimilar-Gender
Status Incongruence-Gender
Variance of organizational level
Variance of supervisor level
3.76
.00
.11
-.60**
.01
.02
.18
.00
.09
.23
.01
.02
4.10
-.01
.11
-.24
.01
.03
.18
.00
.10
.24
.01
.02
** p <.01
43
44
Table 3
_______________________________________________________________
Paired Comparisons for Gender Groups
Mean
Mean
t-test
df
WSupWSub vs. WSupMSub
3.74
3.28
1.90
55
WSupWSub vs. MSupWSub
3.74
3.87
-.98
114
WSupWSub vs. MSupMSub
3.74
3.75
-.06
277
WSupMSub vs. MSupWSub
3.28
3.87
-2.32*
77
WSupMSub vs. MSupMSub
3.28
3.75
-2.23*
240
MSupWSub vs. MSupMSub
3.87
3.75
1.23
299
WSupWSub vs. WSupMSub
3.84
3.74
.35
54
WSupWSub vs. MSupWSub
3.84
3.95
-.77
114
WSupWSub vs. MSupMSub
3.84
3.79
.45
277
WSupMSub vs. MSupWSub
3.74
3.95
-.75
76
WSupMSub vs. MSupMSub
3.74
3.79
-.21
239
MSupWSub vs. MSupMSub
3.95
3.79
1.61
299
Contextual Performance
Task Performance
* p <.05
44
45
Table 4
___________________________________________________________________________
Results of Model Analysis for Age-based Status Dissimilarity and Incongruence on
Contextual and Task Performance
Model 3
Model 4
Contextual Performance
Task Performance
Variable
Estimate
SE
Estimate
SE
Intercept
Gender
Dissimilar-Age
Status Incongruence-Age
Variance of organizational level
Variance of supervisor level
3.79**
.04
-.05
-.31**
.02
.01
.13
.08
.08
.10
.01
.02
3.80
.06
.03
-.36**
.00
.03
.13
.08
.08
.11
.01
.02
** p <.01
45
46
Table 5
_______________________________________________________________
Paired Comparisons for Age Groups
Mean
Mean
t-test
df
Sub>5Yrs vs. Sub<5Yrs
3.48
3.80
-3.65**
221
Sub>5Yrs vs. Similar
Age
3.48
3.84
-3.49**
192
Sub<5 vs. Similar Age
3.80
3.84
-.42
297
Sub>5Yrs vs. Sub<5Yrs
3.53
3.91
-3.52**
221
Sub>5Yrs vs. Similar
Age
3.53
3.86
-3.10**
191
Sub<5Yrs vs. Similar
Age
3.91
3.86
.51
296
Contextual Performance
Task Performance
__________________________________________________________________
** p <.01
46