Out of role? Out of luck: The influence of race and leadership status on performance appraisals

Leadership Status on Performance Appraisals

With the historic passage of the Civil Rights Act of 1964, many racial minority groups hoped that the days of blatant discriminatory employment practices against them would come to an end. On the surface, these groups have made significant progress in their fight for occupational equality. For example, women and minorities represent 85% of the net increase in today's workforce population (Johnston & Packer, 1987). Despite this unprecedented growth, however, many Black workers continue to be dramatically underrepresented in leadership and managerial positions around the nation. For example, of the 434 total members of the U.S. House of Representatives, only 37 are Black. Even more egregious is the fact that there is not a single Black senator among the 100 members of the U.S. Senate (This Nation, 2001). Blacks are also underrepresented in leadership positions in the armed forces (Dovidio & Gaertner, 1996). Within the corporate world, there is also a dearth of Black leadership, as Black workers only comprise 4.7% of America's managerial workforce (DiTomaso & Thompson, 1988).

The advancement and assimilation of Black leaders into the predominantly White organizational hierarchy is often seen as one of the greatest challenges facing corporate America today (Aldefer, Tucker, Morgan, & Drasgow, 1983). Because the prototype of leaders is still that of a White man (Runkle & Ayman, 1997), many Black employees are either consciously or unconsciously relegated to subordinate positions with little possibility for upward mobility or promotion (Morrison & Von Gilinow, 1990). Even when Black employees do ascend into the upper echelons of management, they are often faced with more obstacles to success. For instance, research consistently reveals that Black employees are consistently given lower ratings on job evaluations than are White employees (Greenhaus, Parasuraman, & Wormley, 1993; Kraiger & Ford, 1985), especially on subjective measures of performance (Ford, Kraiger, & Schechtman, 1986). Rather than reflecting any innate inabilities, this finding speaks more to the fact that Black employees (particularly those in leadership positions) are at a systematic disadvantage relative to their White counterparts.

Obstacles Facing Black Leaders

Pettigrew and Martin (1987) outlined reasons why these Black leaders might not perform as well as equally qualified White professionals. For example, if a Black employee's managers, co-workers, and subordinates harbor "the soft bigotry of low expectations," then it subsequently is difficult for Blacks to overcome these negative racial stereotypes. Because many Whites believe that Blacks in leadership positions are hired to fill affirmative action quotas, they subsequently presume that these Black employees are incompetent and unqualified (Heilman, Block, & Lucas, 1992). As such, Black leaders often are not assigned premium job opportunities, are not given discretion on the job, are not included in informal job networks that will help them learn best practices, and are not respected by their subordinates (Greenhaus, Parasuraman, & Wormley, 1993; Richards & Jaffee, 1972). If this pattern continues, then a self-fulfilling prophecy will develop, and what was originally a perceived incompetence will eventually become an actual incompetence if Black managers are not allowed to develop new knowledge, skills, and abilities (KSAs) that would allow them to grow in their job responsibilities.

Furthermore, these negative effects can also be engendered when Black employees themselves suspect that they are hired simply on the basis of a particularistic characteristic (e.g., race) and not a universalistic characteristic (e.g., merit). If leaders believe that they were hired to "fill a quota," they are likely to devalue their own worth and self-efficacy (Brown, Charnsangavej, Keough, Newman, & Rentfrow, 2000; Heilman, Simon, & Pepper, 1987). Unfortunately, these low expectations that Black workers might internalize can also create a self-fulfilling prophecy. People who perceive that they are hired because of particularistic characteristics (even if that is not the actual reality) are likely to characterize themselves as having poor leadership skills and are subsequently less likely to be interested in persisting as a leader (Heilman et al., 1987).

Another reason why qualified Blacks might not perform up to their capability is because of their solo and "token" status within the organizational hierarchy (Sackett, DuBois, & Noe, 1991). Because Black solos are subject to more negative stereotyping and are viewed less positively than White majority group members (Crocker & McGraw, 1984; Taylor, Fiske, Etcoff, & Ruderman, 1978), Blacks in managerial positions are not likely to be associated with leadership (Craig & Feasel, 1998; Craig & Rand, 1998). Furthermore, because there are so few Blacks in organizational leadership positions, many Black managers are likely to feel token stress (Jackson, Thoits, & Taylor, 1995). This occurs when Blacks in traditionally underrepresented positions (i.e., upper management) are called upon to speak on behalf of their entire race. As such, having to act as a spokesperson for an entire group becomes an extreme burden, as it requires solos to consistently be cognizant of the effects of their words and actions (Cioffi, 1994; Saenz, 1991).

Similarly, Black leaders might also feel a sense of stereotype threat in their jobs (Steele, 1999). That is, if Black managers are aware of the stereotypes surrounding their group (i.e., that they aren't "leadership material" and instead should be in subordinate positions; Smith, 1990), they may act in ways that defy the stereotypes. However, in doing so they often feel so much pressure and stress trying to dispel the stereotype that they end up inadvertently confirming it. For example, if Black managers are so worried about making mistakes on the job that they constantly check and recheck their work and question their decisions, they may actually be perceived as an inefficient and ineffective leaders. Intuitively, it might seem that highly qualified Blacks might be immune to stereotype threat, as the stereotypes would not apply to them. However, it is often the most skilled and motivated Blacks (and, consequently, the ones most qualified for leadership positions) who are most negatively affected by stereotype threat because they are the group most concerned with wanting to dispel the negative stereotype (Steele, 1999).

Even when Black leaders do succeed or fail based on their merit (and not external causes listed previously), the implications are different for them than for White leaders. For instance, a mistake made by a Black employee is often attributed to internal causes (e.g., lack of effort or ability) whereas a White employee's mistake is attributed to external causes (e.g., bad luck or a difficult task; Orpen, 1981; Yarkin, Town, & Wallston, 1982); however, the opposite pattern is found for attributions of successful job performance (Greenhaus & Parasuraman, 1993). These two different patterns are consistent with the general finding that schema-inconsistent behavior is often attributed to external factors, whereas schema-consistent behavior is attributed to internal ones (Jackson, Sullivan, & Hodge, 1993). Research also has shown that evaluators tend to disproportionately gather stereotype-consistent information (e.g., Black leaders' successes) and discount or forget stereotype-inconsistent information (e.g., Black leaders' failures; Snyder, 1981).

Aversive Racism and Black Leaders

As all of the above research shows, many social and organizational factors make it extremely hard for even the most well-qualified and ambitious Black leaders to have successes in leadership positions. Perhaps the most difficult obstacle for Black managers to overcome is aversive racism (Dovidio & Gaertner, 1996; Gaertner & Dovidio, 1977, 1986). Dovidio and Gaertner's theory posits that although most people today are not openly racist, subtle and insidious forms of prejudice will emerge if people can justify their negative affect towards Blacks based on factors other than race (Gaertner & Dovidio, 1986). In being able to rationalize their socially undesirable feelings about Blacks, Whites are able to maintain their nonprejudiced self-image by attributing their negative actions to other nonracial reasons.

Because it is threatening to White hegemony for Blacks to have influence and authority, many Whites are either consciously or unconsciously motivated to find ways to perpetuate the existing imbalance of power by reacting negatively toward Blacks in high status positions (Dovidio & Gaertner, 1981; Knight, Giuliano, & Sanchez-Ross, 2001). Although organizational members may be uncomfortable having Black managers, they also know that it is both illegal and "politically incorrect" to discriminate against them without sufficient reason. Indeed, some research shows that if White participants cannot find a satisfactory justification to validate their prejudice, they will actually act more positively towards Blacks than Whites (Linville & Jones, 1980; McConahay, 1983; Rogers & Prentice-Dunn, 1981).

s such, because they do not want to nullify their nonprejudiced self-image, White evaluators may use any mistakes that Black managers make on the job (even if the failures are due to any of the external factors previously outlined) to justify discriminatory feelings and actions towards them (Gaertner & Dovidio, 1977). Furthermore, they may attribute Black managers' failures to internal causes and evaluate them negatively, while reasoning that the more prototypical White managers' mistakes are externally driven and evaluate them positively. As such, the confirmatory standards for Black leaders subsequently would be higher than those for other groups (Biernat & Kobrynowicz, 1997).

Although aversive racism has been extensively studied, no research to date has explored how or if mistakes can be used as sufficient justifications of negative evaluations of both Black and White subordinates and leaders. Therefore, the purpose of the present study was to determine how the race, status, and job-related mistake of employees affects evaluators' ratings of them. Based on aversive racism theory it was predicted that a large mistake would be more pernicious for Black managers than for any other group. That is, it was expected that making a small mistake would not be a sufficient enough reason for participants to justify their negative feelings toward Black leaders. However, it was predicted that making a large mistake on the job would give participants enough justification to evaluate Black managers more unfavorably than Black subordinates, White managers, and White subordinates. Furthermore, although it was expected that Black managers would be evaluated less favorably than Black subordinates, based on schema-consistency theory and prototypes, it was expected that White managers would be rated more favorably than White subordinates, thus confirming Whites in their stereotypically high-status positions.

Method

Participants

Data were collected from 156 White undergraduate students at a small, private university in the Southwest. Students received partial course credit for their participation.

Design and Procedure

A 2(Race: Black or White) X 2(Status: Subordinate or Manager) X 2(Mistake: Small or Large) between-subjects design was used to explore the effect of a male target employee's status, race, and degree of mistake on participants' evaluations of him (only male targets were used in this study to reduce design complexity). After agreeing to complete the questionnaire, which was introduced as "a case study of performance appraisals," participants read a summary of information about a hypothetical employee at their organization and were then asked to evaluate the employee on a number of dimensions. Participants then were debriefed, thanked for their participation, and dismissed.

Each summary was followed by ten questions designed to assess participants' reactions to the target employee. Specifically, participants were asked to rate on 7-point scales ranging from 1(Unacceptable) to 7(Exceptional) the performance of the target on several dimensions, including his work quality, interpersonal skills, conceptual skills, administrative skills, stress tolerance, motivation to achieve, leadership abilities, potential for leadership abilities. Definitions for each of the dimensions were provided to participants. Participants also were asked to rate on a 7-point scale with endpoints at 1(Definitely not) and 7(Absolutely;) the extent to which the target deserved a promotion. These items were obtained from actual performance evaluation scoring sheets in order to increase mundane realism. Participants then completed a manipulation check to determine the salience of the manipulated variables. Specifically, they were asked to record the target employee's race, job title, and recent performance record in the organization.

Materials

All participants were asked to imagine that they had been middle level managers at a motorcycle manufacturing company for the past five years and that their primary responsibilities involved managing lower-level subordinate workers and reporting to higher-level managers. Participants were then given job descriptions for these two job titles; specifically, they were told that lower-level workers worked with cross-trained teams to construct motorcycles from start to finish and that higher-level managers consolidated production reports and supervised middle level managers.

Next, participants were given information about a target employee, "Mark Turner" who was either their lower-level subordinate or their higher-level manager, depending on the condition. In the summary, participants were given ambivalent information about him (e.g., that he was accurate but aloof) to enhance the possibility that participants could seek out schema-consistent evidence in their decision processes. Additionally, within the context of other information given in the summary, Mark was described as being either Black or White. Participants were also told that during a meeting the previous week, one of his subordinates made a comment that Mark found to be insulting to him, although it was not intended to be. In the small mistake condition, Mark cut the meeting short because of the insult, but then later agreed that the remark was innocuous. In the large mistake condition, Mark also cut the meeting short and did not give his subordinate a vital piece of information that the subordinate needed to perform his job. All information summaries were identical, except that the race, status, and degree of mistake varied according to the experimental condition.

Results

The data were analyzed using a 2(Race: Black or White) X 2(Status: Subordinate or Manager) X 2(Mistake: Small or Large) between-subjects Multivariate Analysis of Variance (MANOVA). Contrary to predictions, the three-way interaction between race, status, and degree of mistake was not obtained, F(9, 137) = .46, 2 = .90, [eta]^sup 2^ = .03. However, as expected a two-way interaction between race and status did emerge, F(9, 137) = 2.20, p = .026, [eta]^sup 2^.13.1

To further explore the race x status interaction, the data were analyzed using a between-subjects Analysis of Variance (ANOVA) with a sequential Bonferroni correction to control the Type I error rate. Consistent with predictions, a two-way interaction was revealed for the target employee's motivation to achieve, F(1,145) = 7.62, p = .007, [eta]^sup 2^ = .05. As seen in Figure 1, participants rated White managers (M = 3.67, S.D. = 1.06) as being more achievement oriented than White subordinates (M = 3.20, S.D. = 1.08; t(74) = -1.92, p = .029), and they evaluated Black managers (M = 3.37, S.D. = .85) as being less achievement oriented than Black subordinates (M = 3.86, S.D. = 1.07), t(76) = 2.23, p = .015). A similar two-way interaction was obtained for the target's stress tolerance, as illustrated in Figure 2, F(1, 145) = 12.40, p = .001, [eta]^sup 2^ = .08. Specifically, although participants did not believe that White managers (M = 3.85, S.D. = 1.22) were better able to managers (M = 3.38, .S.D. = .87) as being less able to effectively handle stress than Black

Finally, a two-way interaction revealed that consistent with predictions, race and status influenced participants' perceptions of the quality of the target employee's work, as shown in Figure 3, F(1, 144) = 5.52, p = .02. Although Black managers (M = 4.91, S. D. = .97) were handle pressure than White subordinates (M = 3.72, S.D. = 1.06; t(75) = -.50, p = .62), they did evaluate Black

subordinates (M = 4.42, S.D. = .94; t(76) = 5.02, p = .0001).

evaluated as producing work similar in quality to that of Black subordinates (M = 5.18, S.D. = 1.06; t(76) = 1.16, p = .13), White managers (M = 5.26; S.D. = .91) were perceived as producing higher quality of work than White subordinates (M = 4.81; S.D. = 1.05; t(75) = -2.03, p = .02).

Discussion

The results of the present study confirm that, regardless of the degree of mistake made by the employee, participants viewed Black leaders and White subordinates more negatively than White leaders and Black subordinates. This pattern is consistent with previous research that shows that people who violate their stereotypical social roles are viewed more negatively than those who conform to their proscribed societal roles (Rosenthal & Jacobson, 1968; Knight & Giuliano, 2001). Thus, it appears that participants were either consciously or unconsciously motivated to perpetuate the status quo's imbalance of power between Whites and Blacks. Moreover, the present study revealed that, consistent with the outgroup polarization effect, participants viewed "good outsiders [as] better and bad outsiders [as] worse than comparable insiders" (Fiske & Taylor, 1991, p. 135).

Our study also revealed that making any kind of mistake, large or small, was enough for participants to be able to justify negative reactions toward a target employee. This finding has especially troubling implications for Black managers and leaders in today's workforce, given the many obstacles (e.g., low expectations, token stress, stereotype threat, external attributions for success) that they must repeatedly overcome to succeed in upper-level management positions. Practically speaking then, these results imply several points that organizations should be cognizant of in managing Black leaders. Primarily, they should implement clear and unambiguous performance criteria to ensure that their appraisal systems are not inadvertently perpetuating discrimination (Cascio & Bernardin, 1981). Furthermore, organizations should also train evaluators to be aware of the unique burdens that Black and other minority leaders must surmount and to be aware of the potential cognitive and affective biases that may be present in working with and evaluating an employee of a different race.

Theoretically speaking, this study also has important implications. For example, it confirms the social dominance orientation paradigm, as the White participants in this study seemed to want to maintain the imbalanced distribution of power in society by rewarding Black employees in lower status positions and White employees in higher status positions. Although these "rewards" were small in a statistical sense, many scholars (e.g., Martell, Lane, & Emrich, 1996) argue that it is exactly these small, seemingly insignificant instances of discrimination that result in "molehills becoming mountains" (Valian, 1998). In fact, the apparently trivial privileges that in-group members receive in time develop into the accumulation of advantage for majority group members and a relative disadvantage for all other groups. Furthermore, the results of this study also support the shifting standards theory, or the idea that people have lower minimum standards but higher confirmatory standards for minority group members (Biernat & Kobrynowicz, 1997). Given the current findings, it seems possible that status might have a moderating effect on shifting standards, as the minimum standards for racial minorities in subordinate positions might be especially low, whereas the ability standards for minorities in leadership positions might be especially high.

An important limitation of this study is that it involved evaluations by untrained undergraduates of "paper people" and not actual employees (Gorman, Clover, & Doherty, 1978). This type of methodology often relies on the cognitive-based "stereotype-fit model of discrimination" to the exclusion of behavioral, affective, and social determinants of discrimination (Dipboye, 1985). Although we cannot draw conclusions about the frequency with which aversive racism affects job appraisals, our results do indicate that it can happen (Berkowitz & Donnerstein, 1982). Despite the inherent limitations of "paper people" research, it is important to use a laboratory-based experimental methodology (with its strengths of determining causality and controlling irrelevant variables) to complement field research (with its strength of external validity). Through this process of triangulation and examining the same phenomenon from different perspectives, we can hopefully begin to be able to truly understand prejudice and its effects on organizations and the individuals working within them.

Another limitation of the present study is that the largely White sample precluded the use of examining participant race as potential moderating factor. Evaluators generally give higher ratings to members of their own race (Craig & Rand, 1998; Hamner, Kim, Baird, & Bigoness; 1974), and some research suggests that it is even acceptable for Blacks but not Whites to exhibit this in-group favoritism (Judd & Park, 1988). As such, future studies should investigate the extent to which Black and White evaluators use varying degrees of mistakes to justify lower ratings of out-group and in-group members. The current research only addressed the evaluation of male leaders: further research should also determine the extent to which the aversive racism affects Black women. Although some past research suggests that Black female workers face "double jeopardy" of racism and sexism (Sanchez-Hucles, 1997), other research has found that they actually might be evaluated positively because of their dual membership in two protected groups (Haley, 2001). As such, additional research should examine whether the pattern of results found in the present study is further influenced by either the race of the evaluator or the gender of the Black target employee.

In a more applied tone, future research should also examine ways to reduce aversive racism in the workplace and remove barriers to success for Black leaders and leaders from other underrepresented groups. For example, some have posited that leader-member exchange theory (LMX) is a viable framework through which to understand and successfully manage race relations within organizations (Scandura & Lankau, 1996). By addressing issues such as respect, trust, and obligation between racially diverse dyads, organizations can hopefully minimize the problems and maximize the advantages inherent in these types of pairs. Furthermore, more studies should be conducted to determine the extent to which different diversity training methods are effective in reducing backlash against minority employees, especially those in leadership positions. As America's workforce is becoming less homogenous (Johnston & Packer, 1987), an organization's ability to successfully utilize managers from different racial groups is not just beneficial to minority group members-it is essential to the well being of the entire organization. Through effective and diligent management, organizations can capitalize on the unique characteristics of diverse leaders to make a heterogeneous workforce the competitive advantage it can and should be.