Page 395 - Discrimination at Work The Psychological and Organizational Bases

HEILMAN AND HAYNES
362
differing levels of ambiguity associated with available performance infor­
mation.
In the first study, the level of ambiguity was varied by the degree of
precision of the performance information provided (Heilman, Block, &
Stathatos, 1997). Managers from a large Northeastern insurance company
participated in this study concerning "personnel selection, placement, and
advancement processes" during the course of company-sponsored train­
ing sessions. Fifty-one percent of the managers were men and 49% were
women, and 94% were 25 to 54 years old. The methodology employed in
this study was similar to the initial laboratory study described in this chap­
ter, the first study in Heilman et al. (1992). However, in this investigation
only one job was used, that of a computer programmer. It was chosen to
represent a job that was male sex-typed, but not extremely so, so that the
effects of affirmative action would not be obscured by stereotype-based
incompetence inferences. As in the earlier study, participants were given
packets containing a job description and an employment application con­
taining information about an employee's education and work experience.
However, in this study, participants also were given a six-month job activity
summary for the employee, purportedly written by the employee's super­
visor. Following these materials was a questionnaire asking for reactions
to the employee on several measures including perceived competence.
Once again the hiree was either male or female and, when female, ei­
ther associated with affirmative action or not. This was accomplished in
exactly the same manner as in the Heilman et al. (1992) study. Performance
information was manipulated via the supervisor's response to a question
immediately following the description of the employee's activities, "check
the category which best describes this employee's performance in the past
six months." The actual response and the response format were together
used to vary the ambiguity of the performance information. In the success
conditions, the employee was always rated in the highest category, but the
range of categories differed. In clear success conditions, there were five
category ratings: top 5%, top 10%, top 25%, top 50%, or bottom 50%. In
ambiguous success conditions, there were only two categories: top 50% or
bottom 50%. In addition, there was a condition with no information about
success, in which no rating scale was presented in the materials, and a fail­
ure condition in which the employee had been placed by his supervisor in
the bottom 50% in terms of performance.
Results revealed that, as in earlier studies, with no information about
success affirmative action women were rated as less competent than both
the women not associated with affirmative action and the men. This pat­
tern of data persisted when information was provided about success that
was ambiguous, i.e., rated to be in the top 50% rather than the bottom