Lack of predictive bias in ability tests and large subgroup diVerences in test
scores produce a projected loss in utility when test scores are used in a less than
optimal manner, as would be the case with banding (Laczo and Sackett 2004 ;
Sackett and Roth 1991 ). Whether these diVerences translate into diVerences in
organizational functioning or not is less clear, as is evidence that the proportion
of members of one subgroup or another in the workforce impacts performance.
The popular view (Doyle 2000 ) is that a well-educated, highly diverse workforce
composed of people working productively and creatively with members of diverse
races, religious backgrounds, and cultural histories is important to maintaining
organizational competitiveness. This view is best represented by an amicus brief
Wled by a large number of Fortune 500 companies in support of the University of
Michigan admissions policies (Grutterv.Bollinger et al. 2003 ). Leonard ( 1990 ) and
Steel and Lovrich ( 1987 ) failed toWnd a relationship between the proportion of
minorities or women in organizations and organizational eYciency. Nonetheless,
aYrmative action policies do improve employment opportunities for minority
groups and women (Kravitz et al. 1997 ), and Holzer and Neumark ( 1996 ) reported
little evidence of substantially weaker job performance among most groups of
minority and female aYrmative action hires. Sacco and Schmitt ( 2005 ), however,
found evidence for a negative relationship between racial diversity and change in
proWtability among 3 , 454 quick service restaurants. The whole question of the
impact of organizational diversity on organizational performance merits further
investigation. Like many other performance phenomena, this relationship is likely
moderated and mediated by the past relational histories and attitudes of the
employees and organizations involved as well as the societal context.
Researchers have also examined employee attitudes toward aYrmative action
policies and the people that beneWt from these policies. Heilman et al. ( 1998 )
reported that aYrmative action programs seem to have negative consequences
for perceptions of employees who are thought to be hired based on group mem-
bership rather than merit. Bell et al. ( 2000 ) reported both negative and positive
reactions to aYrmative action programs among manager and student groups. On
the negative side, these people believed that such programs led to employers hiring
less qualiWed employees, were responsible for reverse discrimination, created the
perception that minorities and women could not succeed on their own, and
required a lot of paperwork and resources. On the positive side, these people felt
that aYrmative action improved the job opportunities of women and minorities,
gave everyone an equal opportunity, and reduced discrimination and conXict
among employees. Using the theory of reasoned action (Fishbein and Ajzen
1975 ), they also found that relatively simple attempts to change attitudes toward
aYrmative action programs caused white attitudes to be more negative in response
to negative information; minority attitudes became more favorable as a function
of positive communications. Thus attitudes became more polarized. Attitudes
and intentions based on the Fishbein–Ajzen formulations were related to overt
308 neal schmitt and brian kim