Absence of conflict of interest.
Citation
Highlights
-
The study’s objective was to examine the impact of advertised job compensation schemes and gender differences in the likelihood of applying for a job.
-
The author used a randomized controlled trial design to assess the effectiveness of different forms of advertised job compensation on the likelihood of applying for a job. The analysis compared how likely men and women were to apply for jobs that advertised performance pay versus those that advertised a flat rate of per-hour pay. The author used the email listserv at a large public university to send job advertisements and used a statistical model to compare the number of women versus men submitting a job application in response to a job offering performance pay.
-
The study found a negative and statistically significant relationship between gender and performance-based pay. Compared to men, women were less likely to apply for jobs where the pay was described as performance based. Women were no less likely to apply for jobs where the pay was described as including a non-performance bonus. Whether the work was advertised as in a nonprofit charity setting did not affect this gender difference.
-
The quality of causal evidence presented in this report is low because the author did not ensure that the groups being compared were similar before the intervention. This means we are not confident that the estimated effects are attributable to the difference in advertised compensation schemes for jobs that offer performance-based pay versus flat-rate pay; other factors are likely to have contributed.
Intervention Examined
Advertised Compensation Schemes
Features of the Intervention
The author compared job advertisements sent by email offering three different compensation schemes for short-term student jobs at a large public university. Students in the flat-rate compensation group were offered $24 for a two-hour shift. Students in the tournament group were offered $18 for a two-hour shift and could earn an additional $12 based on their performance. Students in the bonus compensation group were offered $18 for a two-hour shift with a 50 percent chance of receiving an additional $12, which was not tied to performance but was random. In the flat-rate and tournament compensation groups, half of the emailed job advertisements indicated that the work was in a nonprofit charity setting, while the other half did not specify the work setting.
Features of the Study
The author used a randomized controlled trial design to assess the effectiveness of each form of the job advertisement email on the likelihood of recipients applying for the advertised job, focusing on gender differences in the application rate. The author used an email listserv and limited demographic data for 35,462 undergraduate and graduate students from the University of Wisconsin-Madison. Students were randomly assigned to receive an email with one of the three compensation schemes—with some of the students receiving advertisements for work in a nonprofit charity setting—in a job offer for a temporary student employment opportunity. Other information about the advertised jobs did not differ between email versions. About half of the emails went to female students. The author did not provide any additional information on population characteristics.
Findings
The study found a negative and statistically significant relationship between gender and performance-based pay. Compared to men, women were less likely to apply for jobs where the pay was described as performance based. Women were no less likely to apply for jobs where the pay was described as including a non-performance bonus. Whether the work was advertised as in a nonprofit charity setting did not affect this gender difference.
Considerations for Interpreting the Findings
The author had information only on the number of emails that were sent to the student listserv, not the number of students who received or opened the job advertisement emails. Therefore, attrition, or the loss of sample after random assignment, is unknown. Additionally, the author did not provide information on the pre-intervention characteristics of the sample members. These preexisting differences between the groups—and not the advertised compensation scheme—could explain the observed differences in outcomes.