An Imperfect Match? How Gender and Race Influence Perceptions of Job Applicants by Qualification Levels

Download data and study materials from OSF

Principal investigator:

Katherine Weisshaar

University of North Carolina - Chapel Hill



Sample size: 1814

Field period: 02/27/2020-03/08/2020

While many field experiments have demonstrated that racial and gender discrimination in hiring remains a feature of our modern labor market, there remain unanswered questions about the conditions under which discrimination in hiring does – or does not – occur. We suggest that a key source of variation in discrimination levels may be the qualification of the job applicant, relative to the job’s requirements. In the labor market, job applicants apply to a range of jobs – including those that they are over- or under-qualified for – but this process is obscured in our current understanding of discrimination outcomes. In this survey experiment, we examine how evaluations of job applicants’ may vary according to their perceived gender and race (Black and White women and men), and whether the applicants are qualified or underqualified for a job opening. On a sample of employed adults in the U.S., we empirically test competing theories about whether and how gender and racial stereotypes reduce or change depending on applicants’ qualification level by presenting survey respondents with fictitious job applicants who are indicated to either meet or not meet the job requirements. We find general support for the prediction that stereotyped evaluations by gender and race, particularly for Black men and White women, are amplified when applicants are underqualified, compared to qualified. These findings have theoretical and practical implications for our understanding of discrimination and labor market inequality.

Research Question: How do perceptions of job applicants vary by the applicant’s gender and race, and whether they are qualified or underqualified relative to the job requirements?

We consider three competing hypotheses in response to this research question:
Hypothesis 1: Gender and racial stereotypes for Black men and women and White women, compared to White men, for example along dimensions of commitment, capability, assertiveness, and leadership ability, will be heightened for applicants who are underqualified, and reduced for applicants who are qualified. This prediction is derived from the proposition that evaluators will be more uncertain about an applicant’s ability to perform on the job when an applicant is underqualified, and greater uncertainty will lead to increased reliance on gender and racial stereotypes to “fill in the gaps.”
Hypothesis 2: Gender and racial stereotypes for Black men and women and White women, compared to White men, will be similar regardless of whether job applicants are signaled to be underqualified or qualified. While stereotyped evaluations along dimensions such as commitment, capability, assertiveness, and leadership ability are not expected to differ across applicant qualification, gender and racial differences in hireability will occur among underqualified applicants, and less so among qualified applicants. This prediction is derived from the idea that evaluators’ stereotyping processes are relatively difficult to change according to qualification, but applicant underqualification provides an “excuse” for evaluators to act on these stereotypes and not move forward with hiring.
Hypothesis 3: Underqualification activates a different set of stereotypes than being qualified, and those effects vary across racial and gender groups. Negative stereotypes, for example along dimensions of competence and responsibility, will prevail among underqualified applicants, but applicants who meet the job requirements may experience counteracting stereotypes (for example, assertiveness or leadership ability) which may counteract negative stereotypes. This prediction is derived from literature showing that stereotype content can vary for the same groups across different types of information.

Overall, drawing from theories of intersectional prototypicality, we also expect that Black men and White women will face more negative evaluations than Black women, who may be perceived as similarly assertive or have similar leadership ability compared to White men. We further expect less variation according to gender and race on applicants’ perceived hireability, given the increased susceptibility to social desirability bias for this question and that many respondents will not have had hiring experience.

Experimental Manipulations
Gender/race: Signaled gender (man / woman) and race (Black / White) of job applicant. Qualified or not qualified: Statement on whether the job applicant meets or does not meet the job requirements. The following names were used to signal gender and race: DeAndre Washington, Ebony Washington, Jake Walsh, and Katelyn Walsh.
Outcome variables are all on a scale from 1-7.
Hireability: How hireable do you think the applicant is with his or her current cover letter and resume? 7 means the candidate is extremely hireable, 1 means that the candidate is not at all hireable.
Perceived shortcomings of the applicant’s commitment, capability, leadership ability, responsibility, assertiveness, and fit with other accountants at the firm. These questions are worded: On a scale from 1 (not at all) to 7 (much more), how much does the applicant need more emphasis on his or her commitment?
Summary of Results

For those age 30 and older, we find the following, controlling for whether the respondent identified the intended applicant race:
1. When applicants are described as qualified for the job opening, we find no statistically significant penalties (relative to White men) according to applicants’ signaled gender or race. White women are perceived to be more assertive than White men, and Black men and White women are evaluated higher in perceived fit than White men.
2. When applicants are described as not qualified for the job opening, Black men are evaluated lower than White men on dimensions of capability, commitment, leadership ability, and responsibility. White women are evaluated lower than White men on dimensions of commitment and assertiveness. Black women are not evaluated with statistically significant penalties on any dimension, and Black women are evaluated higher than White men in fit in the unqualified condition.
3. We find no statistically significant differences across experimental treatments in the hireability dependent variable.

These findings offer limited support for Hypothesis 1: that gender and racial stereotypes, particularly against Black men and White women, are more pronounced during evaluations when applicants are unqualified compared to when applicants are qualified.

We consider the support limited because we find that treatment effects vary according to respondents’ age: the above results do not hold for respondents younger than age 30, who exhibit reduced levels of bias against Black job applicants than older respondents. Additionally, while we expected that Black women may face fewer or different evaluation stereotypes than Black men and White women, we did not anticipate that Black women would experience no significant penalties and would be evaluated better than White men on perceived fit when underqualified.

We did not find variation according to gender and race in the hireability variable. We believe this is due to the social desirability of the hiring decision question compared to the evaluation questions, but further research is necessary to address the discrepancy between respondents’ reaction to applicant qualification in the evaluation questions but not in the hiring decision question.