Lack of diversity in the workforce; black, hispanic, asian (inequality)
sebpp
Reactionstodiversityinrecruitmentadvertising1.pdf
Reactions to Diversity in Recruitment Advertising—Are Differences Black and White?
Derek R. Avery Saint Joseph’s University
The portrayal of racial diversity in corporate recruitment advertisements has become increasingly common. Despite widespread sentiment that ad diversity attracts a broader scope of applicants, empirical researchon this topic is sparse.Consequently, thepresent studymanipulated addiversity at 2hierarchical levels to assess its impact on organizational attractiveness for 273 Black and White university students. In contrast to the predictions of relational demography, White viewers exhibited no effect for ad diversity. Black viewers were attracted by ad diversity but only when it extended to supervisory level positions. More importantly, the effect of race on reactions to ad diversity was contingent on the viewer’s openness to racial diversity (other-group orientation).
As researchers have proliferated the potential benefits of orga- nizational racial diversity (McLeod, Lobel, & Cox, 1996; Richard, 2000; Watson, Kumar, & Michaelsen, 1993; P. Wright, Ferris, Hiller, & Kroll, 1995), the emphasis placed on recruiting minority applicants has risen considerably (Robinson & Dechant, 1997; Thaler-Carter, 2001). As a result, staffing professionals have de- voted substantial resources to enhance the efficacy of minority recruitment efforts. One tactic that has become increasingly com- mon is the diversification of recruitment advertising because many feel that it influences a broader spectrum of people to pursue employment opportunities (Digh, 1999; Doverspike, Taylor, Shultz, & McKay, 2000; Thaler-Carter, 2001). In accordance, researchers are beginning to investigate the validity of this prac- tice, particularly with respect to its impact on prospective Black applicants (Avery, Hernandez, & Hebl, in press; Highhouse, Stierwalt, Bachiochi, Elder, & Fisher, 1999; Perkins, Thomas, & Taylor, 2000). Despite a number of promising findings amid this preliminary
research, several pressing issues have yet to be addressed. For instance, both Avery et al. (in press) and Perkins et al. (2000) demonstrated that racial similarity between the applicant and the employees depicted in the advertisement enhances organizational attractiveness. However, neither study factored in the hierarchical status of the minority employees depicted in the ads. Researchers must consider the vertical integration of minority employees to fully understand the effects of organizational diversity (Cox, 1991). Furthermore, these investigations have treated the members of each racial group as attitudinally homogenous, essentially ig- noring the substantial variance that exists within individual racial groups (Nkomo, 1992; K. M. Thomas, Phillips, & Brown, 1998; K. M. Thomas & Wise, 1999). As Davidson and Friedman (1998) eloquently stated, “Race research that relies solely on comparing
Whites and Blacks on particular dependent variables probably doesn’t do justice to the conceptual complexities of race” (p. 176). Thus, the failure to consider differences within racial-groups rep- resents another limitation of the existing research. By explicating these issues, the present study aims to enhance
the current understanding of the impact of racial diversity in recruitment advertising. Research on relational demography has suggested that increasing the number of Black employees in an advertisement will attract Black applicants while simultaneously deterring White applicants. However, ad diversity is likely to be more critical to Black job seekers given that organizational diver- sity characteristics in general appear to be more salient to Black than to White applicants when evaluating potential employers (K. M. Thomas & Wise, 1999). In addition, the consideration of individuals’ attitudes toward other racial groups (other-group ori- entation) may explain potential within racial group variance in addition to providing a more thorough clarification of any racial differences that are detected. The sections that follow describe the study’s theoretical foundation and present the research hypotheses.
Relational Demography
Anumberof studies haveused relational demography to explain the effects of diversity among organizational members (e.g., Mc- Neilly & Russ, 2000; Riordan & Shore, 1997; Tsui, Egan, & O’Reilly, 1992). In short, relational demography proposes that the outcomes of diversity are contingent on the demographic charac- teristics of a supervisor or group relative to those of the perceiver. High perceptions of similarity tend to elicit favorable responses such as interpersonal attraction, perceptions of procedural fairness, and increased job satisfaction (Tsui&O’Reilly, 1989;Wesolowski & Mossholder, 1997). Despite a relatively broad range of application, researchers have
yet to extend relational demography to the applicant recruitment process.1 Perhaps one reason for this omission is that relational demography is commonly believed to require interpersonal inter-
1 Rynes, Orlitzky, and Bretz (1997) did, however, suggest the plausibil- ity of organizational demography effects in recruitment and provided some preliminary support with respect to differences in age.
I thank Mickey Quiñones, Mikki Hebl, Bob Dipboye, Jennifer George, Murelle Harrison, Sylvia Hysong, Jessica Foster, Dan Turban, Ashley Crump, Scott Tonidandel, and Jerry Kehoe. Correspondence concerning this article should be addressed to Derek R.
Avery, Department of Management, Erivan K. Haub School of Business, Saint Joseph’s University, 5600 City Avenue, Philadelphia, Pennsylvania, 19131-1395. E-mail: [email protected]
Journal of Applied Psychology Copyright 2003 by the American Psychological Association, Inc. 2003, Vol. 88, No. 4, 672–679 0021-9010/03/$12.00 DOI: 10.1037/0021-9010.88.4.672
672
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
action among individuals that is often lacking from the applicant recruitment process. Nonetheless, social interaction need not occur for individuals to express preference for a group (Tsui et al., 1992, p. 551). Hence, relational demography predicts that ad diversity will either increase or decrease applicant attraction on the basis of congruence between the race of the employees portrayed and that of the viewer. This logic explains why matching the race of the recruiter to that of the applicant maximizes organizational attrac- tiveness even when no applicant–recruiter interaction occurs (Young, Place, Rinehart, Jury, & Baits, 1997). Although this prediction of relational demography in the current
context appears straightforward, empirical evidence suggests oth- erwise. Several findings have indicated that racial cues are less salient to White than to Black applicants (Davis & Burnstein, 1981; Avery et al., in press; Mehra, Kilduff, & Brass, 1998; Perkins et al., 2000; K. M. Thomas & Wise, 1999). For instance, Black applicants assigned greater importance to diversity and recruiter characteristics than did their White counterparts (K. M. Thomas & Wise, 1999). In addition, Blacks (but not Whites) demonstrated racial congruence effects when viewing recruitment advertisements (Avery et al., in press; Perkins et al., 2000) and when making friendship choices (Mehra et al., 1998). Such find- ings led these authors to suggest that race is more salient to Blacks because they are constantly faced with the proposition of being in the numerical minority or even a token. Conversely, Whites are rarely, if ever, faced with such a situation. The resulting discrep- ancy in racial salience should produce Black–White differences in the effects of ad diversity. Theprecedingdiscussion suggests thatBlack job seekerswill be
more attracted to organizations that present Black employees in their ads. Conversely, because race is less salient to them, White job seekers shouldbe less likely to be influencedbyaddiversity so long as White employees remain the numerical majority (Perkins et al., 2000; Riordan & Shore, 1997). Nevertheless, it is of prac- tical importance to demonstrate that the potential upside of ad diversity in attracting Blacks does not simultaneously deter White applicants.
Hypothesis 1: Black job seekers will perceive ads depicting racially diverse (Black and White) employees and White managers more favorably than ads depicting only White personnel. White job seekers will be unaffected so long as White employees remain the numerical majority.
The disparity between the proportion of Blacks in an organiza- tion’s overall workforce and the proportion in management pre- sents a further potential complication for the effects of advertise- ment diversity (Adams, 1998; Cox, 1994; Morrison & Von Glinow, 1990). This distributional imbalance led Cox (1991) to conclude that simple organizational aggregates of employee de- mographic distribution do not allow researchers to fully under- stand the effects of organizational diversity. In the current study, the impact of ad diversity on applicants may differ if the diversity runs throughout the organization as opposed to being limited to merely entry-level positions. For this reason, ad diversity is con- ceptualized in terms of racial vertical integration—the relative proportion of racial diversity represented at all hierarchical levels within an organization. The presence of Black employees is undoubtedly important to
Black job seekers (e.g., Leonard, 2001; K. M. Thomas & Wise,
1999). In fact, a recent survey found that nearly half of Black job seekers indicated that the racial composition of the prospective workforce is a factor in their decision-making process (Leonard, 2001). However, the hierarchical status of these Black employees may be even more significant than their presence alone because a key indication of tokenism is when access to high-status positions is restricted to certain groups (S. C. Wright, Taylor, & Moghaddam, 1990). When racial ad diversity extends to upper- level positions, Black applicants should more readily accept that the organization’s commitment to diversity is sincere. Conse- quently, racial ad diversity may only facilitate minority recruit- ment efforts if the diversity is present at more than just the entry level. Managerial diversity should also be important to Black job
seekers for several other reasons. Black employees who report to White supervisors perceive less trust, experience less supervisory support, fewer developmental opportunities, less procedural jus- tice, lower assimilation, more discrimination, and more role am- biguity than those reporting to other Blacks (Jeanquart-Barone, 1993, 1996; Tsui & O’Reilly, 1989). Furthermore, the presence (or absence) of Blacks at the managerial level should convey career progression information to Black applicants (Kanter, 1977). A greater number of Black managers provides more prospective Black mentors for a Black applicant. This is important because Black employees with Black mentors report receiving more career support while being liked more by their mentor (Dreher & Cox, 1996; Ensher & Murphy, 1997; Kalbfleisch & Davies, 1991; D. A. Thomas, 1990). Seeing Black managers in an ad also informs Black applicants regarding potential opportunities for advance- ment because supervisor–subordinate relational demography af- fects minorities’ perceptions of advancement opportunities and promotion rates (Ely, 1994; Riordan & Shore, 1997; James, 2000; Konrad & Pfeffer, 1991; Wentling & Palma-Rivas, 1998). Thus, ads containing Blacks in high-status positions are likely to convey to Black applicants that advancement opportunities will not be hinderedby institutional racism.Similar logic ledDoverspike et al. (2000) to suggest that successful minorities be used in advertising to bolster minority recruitment.
Hypothesis 2: Black job seekers will perceive ads depicting diverse (Black and White) employees and managers more favorably than ads depicting diverse employees and only White managers. White job seekers will be unaffected as long as White employees and managers remain the numerical majority.
Other-Group Orientation With few exceptions (Pelled, Xin, & Weiss, 2001; Vecchio &
Bullis, 2001), research on relational demography has not consid- ered potential moderators. Although racial differences are impor- tant, the variance within often exceeds that between racial groups thereby limiting the predictive validity of race (Cox, Lobel, & McLeod, 1991; Luhtanen & Crocker, 1992; Nkomo, 1992). One factor that is likely to account for somedegreeof thiswithin-group variance is other-group orientation, which pertains to how an individual thinks about and relates to members of other racial or ethnic groups (Phinney, 1992). Those with high other-group ori- entations enjoy interactingwith, and therefore tend to actively seek opportunities to intermingle with, members of other racial or ethnic groups.
673DIVERSITY IN RECRUITMENT ADVERTISING
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
Because it pertains to attitudes toward other racial groups, other-group orientation is likely to predict an individual’s prefer- ences regarding racial workplace composition. This suggests that the effect of ad diversity may depend on the viewer’s other-group orientation. For example, ad diversity should attract White appli- cants with high other-group orientations because they desire con- tact with members of other racial groups. This could explain why in one study White applicants opted for higher levels of racial diversity when evaluating potential employers (K. M. Thomas, 1993). It is also unlikely that all Black job hunters seek racial diversity in a prospective employer. Many Black workers occupy token or even solo status on a daily basis (Jackson, Thoits, & Taylor, 1995). Over time, these individuals probably become ac- climated to such surroundings and may even develop a preference for groups containing few or no minorities (high other-group orientation). Therefore, White job seekers with higher other-group orientations should react to ad diversity in a manner similar to that proposed for Black job seekers in the first two hypotheses. More- over, Black job seekers with higher other-group orientations should be less attracted to diverse than to homogenous (all- majority) ads.
Hypothesis 3: White job seekers with higher other-group orientations will perceive ads depicting racial diversity more favorably than ads depicting no diversity.
Hypothesis 4: Black job seekers with higher other-group orientations will perceive ads depicting no diversity more favorably than ads depicting racial diversity.
Method
Participants
The sample consisted of 144 Black (33% male, 67% female) and 129 White (34%male, 66%female) students solicited frompsychologycourses at three southern universities.2 Announcements were made in psychology classes at each university that the experiment would either fulfill experi- mental requirements or earn extra credit. Two of the universities were public with large enrollments (more than 5,000 students), and the third was a small private institution. The racial composition of the universities also varied. One was very diverse (36% White, 15% Black, 20% Asian–Pacific Islander, 20% Hispanic, 9% “other”), another was moderately diverse (66% White, 7% Black, 13% Asian–Pacific Islander, 10% Hispanic, 4% “other”), and the third was racially homogenous (1 % White, 98% Black, 1%Asian–Pacific Islander). Participationwasvoluntary andeither fulfilled a course requirement or earned extra credit. The average participant was 20.27 years old with 3.36 years of predominantly part-time working experience. More than half (54.6%) of the participants were currently looking for jobs, whereas another 8.1% had recently found one.
Manipulation
On the basis of two of the preeminent models of organizational diversity (Cox, 1991; Kanter, 1977), Web site ads were constructed portraying three types of organizations. The racial composition of the ads was manipulated through three pictures of employees located on each site. The gender ratios were held constant across conditions to control for potentially confounding effects of gender. The first type of organizational ad was labeled uniform. This organization closely resembled Cox’s (1991) description of the mono- lithic organization. It displayed two pictures (one working and one posing) containing five White employees with a caption introducing the employees and informing viewers that these individuals would be their coworkers at
the firm. There was also a picture containing three White managers in an office settingwith a caption introducing themas the individualswhowould be the viewer’s supervisors. Although all-White ads are probably less common than in previous
years, they appear to still be used in personnel recruitment. In fact, a recent examination of the recruitment Web sites of Fortune magazine’s 100 best companies indicated that, whereas 58 companies displayed pictures of people on their sites, only 38 of those firms varied the racial composition (Cober, Brown, Blumental, & Levy, 2001). It should also be noted that the ads were pilot tested using graduate students to ensure that (a) there were no significant differences in attractiveness among the employees presented in the ads and (b) that the race of each employee was clear to the viewer. The second type of organization, designed to resemble Cox’s (1991)
plural organization, was labeled skewed. In this ad, two of the White coworkers from the uniform ad were replaced with Black coworkers in the photos, but the same three White managers were shown. Although the proportion of Black employees slightly exceeds that suggested by Kanter (1977) for skewed groups, it conforms to her broader description in that, “the numerically dominant types control the group” (p. 208). The last type of organization (balanced) used the same coworker photos as the skewed ad but replaced one of the White managers with a Black one. This ad parallels Cox’s multicultural organization. No organization was portrayed as having Blacks at the managerial level without Black representation at the entry-level because the likelihood of encountering managerial diversity increases as a function of the organization’s entry-level diversity (Konrad & Pfeffer, 1991). This design adheres to the recent suggestion that future studies investigate the effects of vertical integration of minorities (Gilbert & Ones, 1998, p. 700).
Measures
For each of the following measures, respondents indicated their level of agreement with statements on a 7-point Likert-type scale. Higher scores on the measures indicate stronger agreement. Organizational attractiveness. Perceptions of organizational attrac-
tiveness were assessed using four items from prior research (Perkins et al., 2000; Sinar & Highhouse, 2000). A sample item is: “This company is attractive as a place for employment.” This scale proved to be internally consistent (coefficient ! ! .82). Other-group orientation. Phinney’s (1992) multigroup ethnic identity
measure (MEIM) was used to assess participants’ other-group orientation (coefficient ! ! .77). This inventory, which allows for simultaneous assessment of other-group orientation across various racial groups, has proven to be reliable (Phinney, 1992). A sample item is: “I like meeting and getting to know people from ethnic groups other than my own.” Demographic control variables. To control for possible confounding
effects, the following demographic variables were collected: economic status, gender, age, and work experience.3,4 These variables are commonly included as covariates in race-related research to disentangle their effects from those of race, thereby allowing for more accurate attribution of potential findings. A 5-point ordinal scale was used to assess participants’ economic status.
2 To avoid alerting the participants of the racial nature of the study, non-Black minorities were allowed to participate; however, their data were not analyzed for the purposes of this experiment. 3 One reviewer raised the question of potential interactive effects of race
and gender on the dependent variable; however, including gender as an independent variable produced no statistically significant effects. 4 Salary, which was manipulated to address questions not pertaining to
those discussed here, was also included as a covariate.
674 AVERY
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
Procedure The data were collected by computer. Initially, the participants logged
on to the experiment Web site’s opening page where they were instructed that they would be asked to evaluate the recruitment Web site of a growing consulting firm. They were then assigned to one of the three experimental conditions by a random number generator. This redirected them to one of three Internet Web sites designed to resemble a typical organizational recruitment Web site (Cober et al., 2001). Each included a brief company description of a fictitious up-and-coming organizational consulting firm (Essex), followed by a “virtual site visit.” The three sites were identical except for the photo manipulation in the virtual site visit. After reading the organizational description and finishing the virtual site visit, each partici- pant completed all of the measures and submitted his or her responses to the experimenter by anonymous e-mail.
Results
Similar toWilliamsandBauer’s (1994) test of compatibility, the participant’s college of origin was examined as an independent variable to test for differences across the threedata collection sites. No significant effects were detected after controlling for economic status and race, and thus the data were combined. The resulting means, standarddeviations, andcorrelations are located inTable1. Unexpectedly, both organizational attractiveness and other-group orientation were correlated to the race of the participant, with Black participants indicating significantly higher organizational attractiveness and lower other-group orientations than their White counterparts. The respondents’ level of agreement with a statement included
at the end of the survey (“Black workers are employed at all organizational levels in this company”) was used to evaluate the effectiveness of the ad diversity manipulation. A one-way analysis of variance (ANOVA) indicated that the effect of ad diversity was statistically significant, F(2, 270) ! 31.29, p " .01, "2 ! .19. Moreover, the means of each condition significantly differed from one another (M ! 2.23, 3.28, and 3.72, ps " .05, respectively). Although none of the conditions elicited strong agreement with the item (the scale midpoint is 4), the pattern is consistent with what was expected. These results also accurately reflect that the per- centage of Black employees never exceeded the numerical minority. At the conclusion of the measures, the participants were also
asked about the purpose of the study in an open-ended question. Although most simply regurgitated what they were told at the onset of the experiment (“To get my opinion about Essex and its Web site”), a small number of the participants surmised that the
study’s true purpose pertained to ethnicity in the workplace. How- ever, the pattern of these individuals’ responses did not differ from that of the unsuspecting respondents, suggesting that these indi- viduals were not uniquely sensitized to the manipulation. In addi- tion, to ensure that there were no differences between active and inactive job seekers, an analysis of covariance (ANCOVA) was performed, including a categorical job-seeking status variable with race and ad diversity as independent variables. Job-seeking status failed to exhibit any statistically significant effects on attractive- ness (main or interactive), indicating that active and inactive job seekers responded similarly.
Statistical Analyses
The research hypotheses were examined using hierarchical moderated multiple regression (MMR). To assess the impact of ad diversity, two coding variables were created. The first (restricted diversity) usedcontrast coding to compare theuniformand skewed companies while simultaneously minimizing the impact of the balanced organization (Cohen & Cohen, 1983). The second vari- able (unrestricted diversity) used contrast coding to compare the balanced and skewed organizations (Cohen & Cohen, 1983). This coding strategy addresses differences between (a) organizations depicting Black representation at only the entry level and racially homogenous organizations and (b) organizations depicting Black employees at all hierarchical levels and those with diversity re- stricted to the entry level. The control variables were entered in Step 1 (see Table 2). In Step 2, other-group orientation, viewer race, and the two diversity variables were added. In Step 3, the composite two-way interactions between the variables were entered. Finally, the two, three-way interactions were entered (Step 4). Hypothesis 1 proposed that the presence of Blacks at the entry
level (restricted ad diversity) would enhance the attraction of Black viewers while having no impact on White viewers. The Restricted Diversity # Viewer Race interaction approached con- ventional levels but was not statistically significant (# ! $.35, p ! .10). However, the observed trend was counter to what was predicted because White viewers exhibited a slight negative reac- tion to restricted diversity, whereas Black viewers exhibited no effect (see Table 3). Consequently, Hypothesis 1 was not supported. The second hypothesis proposed that Black, but not White, job
seekers would view unrestricted ad diversity (Blacks in positions of power) as more attractive than restricted diversity. The Unre-
Table 1 Means, Standard Deviations, and Correlations for All Variables
Variable M SD 1 2 3 4 5 6 7
1. White (race) 0.47 0.50 2. Female 0.66 0.47 .01 3. Age 20.27 2.32 $.19** $.02 4. Work experience 3.36 2.42 $.06 $.13* .46** 5. Economic status 3.44 1.47 .53** $.06 $.14* .01 6. Organizational attractiveness 18.50 4.67 $.30** $.02 .03 $.03 $.18** 7. Other-group orientation 5.78 0.92 .28** .01 .01 .03 .12 .11
Note. N ! 273. The variables “White” and “Female” are dummy variables wherein Black and male ! 0 and White and female ! 1. * p " .05. ** p " .01.
675DIVERSITY IN RECRUITMENT ADVERTISING
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
stricted Diversity # Viewer Race interaction was statistically significant (# ! $.43, p " .05), indicating that Black viewers were more attracted to organizations depicting Blacks in both higher and lower status positions than to those wherein Black representation was limited to the entry level. Conversely, their White counterparts’ perceptions were unaffected by the presence of Black managers (see Table 3). Thus, Hypothesis 2 was supported. Hypotheses 3 and 4 proposed three-way interactions between
other-grouporientation, addiversity, andviewer race.Specifically, positive effects of ad diversity were proposed for White viewers with higher other-group orientations (Hypothesis 3). Negative effects of ad diversity were proposed for Black viewers with higher other-group orientations (Hypothesis 4). The Viewer Race # Restricted Diversity # Other-Group Orientation interac- tion was statistically significant (# ! .92, p " .01), indicating that the effect of restricted ad diversity differed depending on the race
and other-group orientation of the viewer. As hypothesized, White viewers with higher other-group orientations exhibited positive effects of restricted ad diversity. In addition, Black viewers with higher other-group orientations exhibited the hypothesized nega- tive effect of restricted ad diversity. White viewers with low other-group orientations exhibited a similar effect. A graphic representation of the tests of Hypotheses 3 and 4 is
presented inFigure1.Tograph interactions involvingacontinuous variable, Cohen and Cohen (1983) advocated dichotomizing the variable by using respondents who fall 1 SD above and below the mean on the variable to constitute the high and low categories, respectively. However, to ensure adequate racial representation in the high and low other-group orientation groups, each racial group was dichotomized individually in this manner. Figure 1 appears to indicate an interaction between viewer race,
unrestricted ad diversity, and other-group orientation. This inter- action, however, although statistically significant (# ! .46, p " .05), did not follow the predicted pattern. Black viewers with
Table 3 The Interactive Effects of Viewer Race and Advertisement Diversity (Proportional Representation) on Organizational Attractiveness
Proportional representation Black viewers White viewers
Uniform M 19.18 17.61 SD 4.33 5.07 n 51 44
Skewed M 18.94 16.45 SD 4.18 4.55 n 47 42
Balanced M 21.46 16.98 SD 3.46 4.77 n 46 43
Table 2 The Effect of Virtual Diversity, Viewer Race, and Other-Group Orientation on Organizational Attractiveness
Variable Step 1 # Step 2 # Step 3 # Step 4 #
Gender .01 ($.11, .13) .01 ($.10, .12) .03 ($.09, .14) .02 ($.09, .13) Age .03 ($.11, .17) $.01 ($.15, .13) $.03 ($.17, .11) $.01 ($.14, .12) Economic status $.17** ($.29, $.05) .00 ($.13, .14) $.00 ($.14, .13) $.02 ($.16, .11) Work experience $.02 ($.16, .11) $.03 ($.16, .10) $.02 ($.14, .11) $.03 ($.15, .10) Experience type $.06 ($.18, .07) $.05 ($.17, .07) $.05 ($.17, .07) $.07 ($.18, .05) Viewer race (race) $.36** ($.50, $.22) $.36** ($.50, $.22) $.33** ($.47, $.20) Restricted diversity (RD) .01 ($.12, .14) .34 ($.08, .76) .31 ($.10, .72) Unrestricted diversity (UD) .10 ($.03, .23) .51* (.09, .92) .57** (.16, .98) Other-group orientation (OGO) .20** (.08, .31) .46** (.10, .82) .56** (.21, .92) Race # RD $.35 ($.77, .07) $.42* ($.83, $.01) Race # UD $.43* ($.85, $.04) $.55** ($.97, $.13) Race # OGO $.28 ($.64, .08) $.40* ($.75, $.05) RD # OGO .01 ($.13, .16) $.86** ($1.3, $.44) UD # OGO .02 ($.12, .16) $.40 ($.83, .02) Race # RD # OGO .92** (.50, 1.35) Race # UD # OGO .46* (.03, .89) %R2 .05** .11** .03 .06** R2 .05 .16 .19 .24
Note. N ! 273. For the race variable, Black ! 0, White ! 1. The numbers in the parentheses are the lower and upper bounds, respectively, for the 95% confidence intervals. * p " .05. ** p " .01.
Figure 1. The interactive effects of viewer race, advertisement diversity (proportional representation), and other-group orientation on organiza- tional attractiveness. Lower other-group orientation is indicated by a thin line, higher other-group orientation is indicated by a bold line.
676 AVERY
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
higher other-group orientations exhibited a stronger positive effect of unrestricted diversity than Black viewers with lower other- group orientations. White viewers’ perceptions of organizational attractiveness were unaffected by the presence of high status Blacks irrespective of their other-group orientation. Hence, Hy- potheses 3 and 4 were partially supported.
Discussion
The present study makes a significant contribution to the recent literature assessing the impact of advertisement diversity. Al- though support for relational demography was mixed, the results clearly demonstrate that there are both between and within-race differences in the effects of ad diversity. It is also apparent that the racial vertical integration or hierarchical proportional representa- tion of minority employees has a significant bearing on how organizations portraying ad diversity are perceived by job seekers. The inclusion of other-group orientation helped to further ex-
plicate the racial differences in the responses to ad diversity. An examination of Table 3 erroneously suggests that ad diversity has no impact on potential White applicants and that only unrestricted ad diversity affects prospective Black applicants. In actuality, restricted ad diversity appears to attract some White applicants while deterring others. Furthermore, consistent with relational demography, Blacks with lower other-group orientations exhibited the hypothesized positive effect of restricted ad diversity (Hypoth- esis 1). However, this effect was offset by the proposed negative effect for Blacks with higher other-group orientations (Hypoth- esis 4). As expected, Blacks with higher other-group orientations opted
for organizations portraying no diversity to those portraying only restricted diversity. These individuals appeared to perceive such organizations as presenting the best opportunity for maximizing interaction with other groups. For many traditionally racially ho- mogenous organizations, attracting diverse applicants remains a major challenge (Doverspike et al., 2000; K. M. Thomas & Wise, 1999). This finding further demonstrates the importance of con- sidering moderators when examining relational demography ef- fects (Pelled et al., 2001; Vecchio & Bullis, 2001). It also provides a potential measure of encouragement for racially homogenous organizations that seek to diversify. Another potential explanation for this finding is that Black
applicants may associate some advantages with occupying solo status. Being a racial minority in the numerical minority increases one’s visibility (Kanter, 1977),whichneednot alwaysbenegative. For instance, a recent study reported that solo Blacks received higher ratings and were selected to be leader more often than White group members (Craig & Rand, 1998). These ratings in- cluded leadership ability, percentage of task contribution, warmth, and friendliness. Thus, to the extent that Black job seekers antic- ipate being regarded in this manner, they may even opt for solo status with the hopes of facilitating their advancement within the company. However, cynicism on the part of some Black participants may
be an equally plausible explanation. Perhaps restricted racial di- versity implied to Black participants that the organization’s com- mitment to diversity was insincere and that the Black employees in this condition were merely tokens. This is consistent with the notion that diversity in the upper echelons of organizations con- veys more symbolic meaning than diversity at lower hierarchical
levels (Cox, 1991; Ely, 1994; Milliken & Martins, 1996). In addition, Blacks’ relative disenchantment with organizations dis- playing restricted diversity may serve them well. Groups with higher status minorities tend to experience less conflict than those wherein minorities have no power (Tolbert, Andrews, & Simons, 1995).
Limitations and Directions for Future Research
As with any study, caution should be exercised when drawing conclusions from the results. The sample represents one potential limitation. For instance, it should be noted that the participants were obtained from universities located in the southern region of the United States wherein the proportion of Black residents is higher than in some other regions. This may have impacted the race-related attitudes of the participants originating from the re- gion. The use of undergraduates may also represent a potential limitation because it is uncertain how well the results generalize to applicants whose work experience significantly exceeds that of the current sample. However, the results coincide with a recent survey wherein 44% of Black applicants reported that the racial compo- sition of prospective firms influences their job seeking process (Leonard, 2001). Moreover, college recruitment has been identi- fied as one of the best and most practical ways to diversify organizations (Adams, 1998; Digh, 1999; Doverspike et al., 2000). Therefore, understanding how ad diversity impacts the perceptions of college students is important in its own right. The study’s experimental stimuli also pose potential limitations.
The company described in the manipulation was a consulting firm. Although consulting is applicable across a broad range of aca- demic disciplines, it is possible that this type of work dissuaded some of the participants. In addition, although practical limitations restrict the number of conditions that can be examined in any one experiment, the potential combinations of ad diversity are limit- less. Last, the stimuli portray employees at only two hierarchical levels and therefore do not allow for the consideration of potential racial disparity between first-line supervisor and higher level ex- ecutive positions.5 This is important because many minority ap- plicants may look for evidence of such a glass ceiling to surmise their ownadvancementopportunitieswithin the company (Riordan & Shore, 1997). Subsequent studies should further explore this issue. Diversity extends beyond the Black and White focus of the
present study. For instance, other types of racial diversity may elicit a different pattern of responses. However, minority job seekers appear to be more concerned with whether the employees depicted in job ads are fellow minorities than with the specific race of the individual(s) (Avery et al., in press). This suggests that the results of the current study may generalize well to other types of racial diversity. Yet, race is but one form of visible diversity. Future research could focus on determining the effects of portray- ing other forms (e.g., gender, age, physical ability, culture) on viewers’ reactions. Furthermore, it is important to consider how ad diversity fits into the larger context of personnel recruitment. In the present study, several important factors were not allowed to vary in an attempt to maximize experimental control. Thus, subsequent examinations should explore the relative impact of ad diversity on
5 Thanks to an anonymous reviewer for raising this issue.
677DIVERSITY IN RECRUITMENT ADVERTISING
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
organizational attractiveness when factors such as pay, schedule flexibility, industry, and so forth are variable. The two unanticipated main effects of race (other-group orien-
tation and organizational attractiveness) may also stimulate future study. The effect of race on other-group orientation was likely a product of the participants’ prior experiences because more posi- tive and less negative intergroup contact tends to produce higher other-group orientations (M. O. Wright & Littleford, 2002). Given that Blacks are more likely than Whites to have experienced discrimination (Crow, Fok, & Hartman, 1998), it is not surprising that their other-group orientations were lower. Conversely, inter- preting the correlation between race and attraction is considerably more speculative. Williams and Bauer (1994) reported a similar effect of race in their examination of the impact of diversity management policies on applicant attraction. They concluded that there must have been some aspect of their recruitment brochure, “that was constant across both experimental conditions that caused racialminoritygroupmembers to evaluate the companydifferently than White participants” (p. 303). Although this explanation could also apply to thepresent study, it is alsopossible thatminorities are more receptive in general to recruitment literature.
Conclusions
The findings have significant implications for race research. Although researchers have often considered variance among ma- jority group members, they have tended to treat minority groups as attitudinally homogenous (Cox & Nkomo, 1990; Nkomo, 1992; K. M. Thomas & Wise, 1999). The effect of other-group orienta- tion further demonstrates that there is significant variance within racial minority groups (Cox et al., 1991; Luhtanen & Crocker, 1992). Moreover, similar interactive effects of ethnic identity have been reported in product advertising and marketing research (Green, 1999; Simpson, Snuggs, Christiansen, & Simples, 2000), suggesting that all racial relational demography effects might be contingent on ethnic identification and other-group orientation. Despite considerable variance in the effects of ad diversity, it
appears that companies canbenefit from its use if they are strategic in their approach. To attract Blacks (and potentially other minor- ities), organizations should convey that employee racial diversity is unrestricted and extends beyond merely entry-level positions. Furthermore, it appears that some White applicants may be at- tracted by ad diversity. However, companies should be aware that advertisement diversity is not a recruitment panacea because it does not entice all Black applicants and may even curtail the interest of some White applicants. In sum, differences in reactions to diversity in recruitment advertising are more complex than merely Black and White.
References
Adams, M. (1998, August). Building a rainbow, one stripe at a time. HR Magazine, 72–79.
Avery, D. R., Hernandez, M., & Hebl, M. R. (in press). Who’s watching the race? Racial salience in recruitment advertising. Journal of Applied Social Psychology.
Cober, R. T., Brown, D. J., Blumental, A. J., & Levy, P. E. (2001, April). What do the “best companies” do? A qualitative analysis of Internet recruiting practices. Paper presented at the 16th Annual Conference of the Society for Industrial & Organizational Psychology, San Diego, CA.
Cohen, J., & Cohen, P. (1983). Applied multiple regression/correlation analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum.
Cox, T., Jr. (1991). The multicultural organization. Academy of Manage- ment Executive, 5(2), 34–47.
Cox, T., Jr. (1994). Cultural diversity in organizations: Theory, research, & practice. San Francisco: Berrett-Koehler.
Cox, T., Jr., Lobel, S. A., & McLeod, P. (1991). Effects of ethnic group cultural differences on cooperative and competitive behavior on a group task. Academy of Management Journal, 34, 827–847.
Cox, T., Jr., & Nkomo, S. M. (1990). Invisible men and women: A status report on race as a variable in organization behavior research. Journal of Organizational Behavior, 11, 419–431.
Craig, K. M., & Rand, K. A. (1998). The perceptually “privileged” group member: Consequences of solo status for African Americans and Whites in task groups. Small Group Research, 29, 339–358.
Crow, S. M., Fok, L. Y., & Hartman, S. J. (1998). Who is at greater risk of work-related discrimination: Women, Blacks, or homosexuals? Em- ployee Responsibilities and Rights Journal, 11, 15–26.
Davidson, M., & Friedman, R. A. (1998). When excuses don’t work: The persistent injustice effect among Black managers. Administrative Sci- ence Quarterly, 43, 154–183.
Davis, L., & Burnstein, E. (1981). Preference for racial composition of groups. Journal of Psychology, 109, 293–301.
Digh, P. (1999, October). Getting people into the pool: Diversity recruit- ment that works. HR Magazine, 94–98.
Doverspike, D., Taylor, M. A., Shultz, K. S., & McKay, P. F. (2000). Responding to the challenge of a changing workforce: Recruiting non- traditional demographic groups. Public Personnel Management, 29, 445–457.
Dreher,G.F.,&Cox,T.H. (1996).Race, gender, andopportunity:A study of compensation attainment and the establishment of mentoring relation- ships. Journal of Applied Psychology, 81, 297–308.
Ely, R. J. (1994). The effects of organizational demographics and social identity on relationships among professional women. Administrative Science Quarterly, 39, 203–238.
Ensher, E. A., & Murphy, S. E. (1997). Effects of race, gender, perceived similarity, and contact on mentor relationships. Journal of Vocational Behavior, 50, 460–481.
Gilbert, J. A., & Ones, D. S. (1998). Role of informal integration in career advancement: Investigations in plural and multicultural organizations and implications for diversity valuation. Sex Roles, 39, 685–704.
Green, C. L. (1999). Ethnic evaluations of advertising: Interaction effects of strength of ethnic identification, media placement, and degree of racial composition. Journal of Advertising, 28, 49–64.
Highhouse, S., Stierwalt, S. L., Bachiochi, P., Elder, A. E., & Fisher, G. (1999). Effects of advertised human resource management practices on attraction of African American applicants. Personnel Psychology, 52, 425–442.
Jackson, P. B., Thoits, P. A., & Taylor, H. F. (1995). Composition of the workplace and psychological well-being: The effects of tokenism on America’s Black elite. Social Forces, 74, 543–557.
James, E. H. (2000). Race-related differences in promotions and support: Underlying effects of human and social capital. Organization Science, 11, 493–508.
Jeanquart-Barone, S. (1993). Trust differences between supervisors and subordinates: Examining the role of race and gender. Sex Roles, 29, 1–11.
Jeanquart-Barone, S. (1996). Implications of racial diversity in the supervisor–subordinate relationship. Journal of Applied Social Psychol- ogy, 26, 935–944.
Kalbfleisch, P. J., & Davies, A. B. (1991). Minorities and mentoring: Managing the multicultural institution. Communication Education, 40, 266–271.
Kanter, R. M. (1977). Men and women of the corporation. New York: Basic Books.
678 AVERY
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
Konrad, A. M., & Pfeffer, J. (1991). Understanding the hiring of women and minorities in educational institutions. Sociology of Education, 64, 141–157.
Leonard, B. (2001, April). Diverse workforce tends to attract more female and minority job applicants. HR Magazine, 27.
Luhtanen, R., & Crocker, J. (1992). A collective self-esteem scale: Self- evaluation of one’s social identity. Personality & Social Psychology Bulletin, 3, 302–318.
McLeod, P. L., Lobel, S. A., & Cox, T. H. (1996). Ethnic diversity and creativity in small groups. Small Group Research, 27, 248–264.
McNeilly,K.M.,&Russ, F.A. (2000).Does relational demographymatter in a personal selling context? Journal of Personal Selling and Sales Management, 20, 279–288.
Mehra, A., Kilduff, M., & Brass, D. J. (1998). At the margins: A distinc- tiveness approach to the social identity and social networks of under- represented groups. Academy of Management Journal, 41, 441–452.
Milliken, F. J., & Martins, L. L. (1996). Searching for common threads: Understanding the multiple effects of diversity in organizational groups. Academy of Management Review, 21, 402–433.
Morrison, A. M., & Von Glinow, M. A. (1990). Women and minorities in management. American Psychologist, 45, 200–208.
Nkomo, S. M. (1992). The emperor has no clothes: Rewriting “race in organizations.” Academy of Management Review, 17, 487–513.
Pelled, L. H., Xin, K. R., & Weiss, A. M. (2001). No es como mi: Relational demography and conflict in a Mexican production facility. Journal of Occupational and Organizational Psychology, 74, 63–84.
Perkins, L. A., Thomas, K. M., & Taylor, G. A. (2000). Advertising and recruitment: Marketing to minorities. Psychology & Marketing, 17, 235–255.
Phinney, J. S. (1992). The multigroup ethnic identity measure: A new scale for use with diverse groups. Journal of Adolescent Research, 7, 156– 176.
Richard, O. C. (2000). Racial diversity, business strategy, and firm per- formance:A resource-basedview.AcademyofManagement Journal, 43, 164–177.
Riordan, C. M., & Shore, L. M. (1997). Demographic diversity and employee attitudes: An empirical examination of relational demography within work units. Journal of Applied Psychology, 82, 342–358.
Robinson,G.,&Dechant,K. (1997).Buildingabusiness case for diversity. Academy of Management Executive, 11(3), 21–31.
Rynes, S. L., Orlitzky, M. O., & Bretz, R. D. (1997). Experienced hiring versus college recruiting: Practices and emerging trends. Personnel Psychology, 50, 309–339.
Simpson, E. M., Snuggs, T., Christiansen, T., & Simples, K. E. (2000). Race, homophily, and purchase intentions and the Black consumer. Psychology & Marketing, 17, 877–889.
Sinar, E. F., & Highhouse, S. (2000). Organizational attraction measures: Construct delineation and theoretical framework. Paper presented at the 15th Annual Conference of the Society for Industrial & Organizational Psychology, New Orleans, LA.
Thaler-Carter, R. E. (2001, June). Diversify your recruitment advertising. HR Magazine, 92–100.
Thomas, D. A. (1990). The impact of race on managers’ experiences of developmental relationships (mentoring and sponsorship): An intra- organizational study. Journal of Organizational Behavior, 11, 479–492.
Thomas, K. M. (1993). Recruiting workforce 2000: A policy-capturing study. Unpublished doctoral dissertation, Pennsylvania State University, University Park.
Thomas, K. M., Phillips, L. D., & Brown, S. (1998). Redefining race in the workplace: Insights from ethnic identity theory. Journal of Black Psy- chology, 24, 76–92.
Thomas, K. M., & Wise, P. G. (1999). Organizational attractiveness and individual differences: Are diverse applicants attracted by different factors? Journal of Business & Psychology, 13, 375–390.
Tolbert, P. S., Andrews, A. O., & Simons, T. (1995). The effects of group proportions on group dynamics. In S. E. Jackson & M. N. Ruderman (Eds.), Diversity in work teams: Research paradigms for a changing workplace (pp. 131–159). Washington, DC: American Psychological Association.
Tsui, A. S., Egan, T. D., & O’Reilly, C. A. (1992). Being different: Relational demography and organizational attachment. Administrative Science Quarterly, 37, 547–579.
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.
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.
Watson, W. E., Kumar, K., & Michaelsen, L. K. (1993). Cultural diversi- ty’s impact on interaction process and performance: Comparing homog- enous and diverse task groups. Academy of Management Journal, 36, 590–602.
Wentling, R. M., & Palma-Rivas, N. (1998). Current status and future trends of diversity initiatives in the workplace: Diversity experts’ per- spective. Human Resource Development Quarterly, 9, 235–253.
Wesolowski, M. A., & Mossholder, K. W. (1997). Relational demography in supervisor–subordinate dyads: Impact on subordinate job satisfaction, burnout, and perceived procedural justice. Journal of Organizational Behavior, 18, 351–362.
Williams, M. L., & Bauer, T. N. (1994). The effects of a managing diversity policy on organizational attractiveness. Group & Organization Management, 19, 295–308.
Wright, M. O., & Littleford, L. N. (2002). Experiences and beliefs as predictors of ethnic identity and intergroup relations. Journal of Multi- cultural Counseling and Development, 30, 2–20.
Wright, P., Ferris, S. P., Hiller, J. S., & Kroll, M. (1995). Competitiveness through management of diversity: Effects on stock price valuation. Academy of Management Journal, 38, 272–287.
Wright, S. C., Taylor, D. M., & Moghaddam, F. M. (1990). Responding to membership in a disadvantaged group: From acceptance to collective protest. Journal of Personality and Social Psychology, 58, 994–1003.
Young, I. P., Place,A.W.,Rinehart, J. S., Jury, J.C.,&Baits,D.F. (1997). Teacher recruitment: A test of the similarity–attraction hypothesis for race and sex. Educational Administration Quarterly, 33, 86–106.
Received May 14, 2002 Revision received November 18, 2002
Accepted December 4, 2002 !
679DIVERSITY IN RECRUITMENT ADVERTISING
T hi
s do
cu m
en t i
s co
py ri
gh te
d by
th e
A m
er ic
an P
sy ch
ol og
ic al
A ss
oc ia
tio n
or o
ne o
f i ts
a lli
ed p
ub lis
he rs
. T
hi s
ar tic
le is
in te
nd ed
s ol
el y
fo r t
he p
er so
na l u
se o
f t he
in di
vi du
al u
se r a
nd is
n ot
to b
e di
ss em
in at
ed b
ro ad
ly .
