Gender-Fair Framing of Job Titles

How to cite

Gorny, Paul M.; Nieken, Petra. 2021. "Gender-Fair Framing of Job Titles." AEA RCT Registry. AEARCTR-0008060. https://doi.org/10.1257/rct.8060-4.0

Download: BibTeX · RIS — for reference managers without a browser translator. Zotero and most others pick up this page's citation metadata directly.

Paul M. Gorny* , Petra Nieken

* Primary investigator

Registry ID
AEARCTR-0008060
DOI
https://doi.org/10.1257/rct.8060-4.0
First registered
Last updated
Status
Completed
Trial dates
2021-09-06 – 2022-11-15
Countries
Germany ()
Registry record
http://www.socialscienceregistry.org/trials/8060

Abstract

We study the effects of gender-fair framing of job titles in online job ads in cooperation with a large online job platform in Germany. We use a randomized controlled trial (RCT) design. We are particularly interested in how the usage of the gender-fair variant affects the female to male applicant ratio as well as the overall number of applications. The goals of our study are (i) to investigate whether the female applicant ratio can be increased by deviating from the “generic masculinum” which is typically used in German, (ii) study whether any changes in overall applicant numbers are due to a specific sex of the applicants, and (iii) investigate whether and how these effects differ across industries and jobs.

We focus on three broad categories of jobs as used by the online job platform: “IT & Development”, “Business and Management” and “Marketing and Sales”. For the jobs contained within these categories, we can use secondary data to see whether these jobs empirically more often are occupied by men or women. We only use full-time positions for jobs not already advertised using gender-fair titles by the hiring company within each of the above-mentioned categories.

Our study encompasses two treatments per each of the three broad industry categories: In the baseline, we record the number of female applications for job titles using the generic masculinum. In the gender-fair treatment, we record the female applicant ratio when jobs are advertised using a gender-fair frame. The RCT will be conducted in cooperation with a German online job platform and all applications towards the eligible jobs, as described earlier, can be part of the study and will, upon checking eligibility, enter the RCT and thus be randomly assigned to one of the two treatments with equal probability.

Intervention

Our study encompasses two treatments per each of the three broad industry categories: In the baseline, we record the number of female applications for job titles using the generic masculinum. In the gender-fair treatment, we record the female applicant ratio when jobs are advertised using agender-fair frame. Target group: The RCT will be conducted in cooperation with a German online job platform and all applications towards the eligible jobs, i.e., full-time jobs not already advertised using gender-fair titles by the hiring company within the categories “IT & Development”, “Business and Management” and “Marketing and Sales”, can be part of the study and will, upon checking eligibility, enter the RCT and thus be randomly assigned to one of the two treatments with equal probability.

Intended purpose: We study the effects of gender-fair framing of job titles in online job platforms in cooperation with a large German online job platform using a randomized controlled trial (RCT) design. We are particularly interested in how the usage of the gender-fair variant affects the female and male applicant ratio as well as the overall number of applications. The goals of our study are (i) to investigate whether the female applicant ratio can be increased by deviating from the “generic masculinum” which is typically used in German, (ii) study whether any changes in overall applicant numbers are due to a specific sex of the applicant, and (iii) investigate whether and how these effects differ across industries and jobs.

Experimental design

We have two treatment groups (baseline, gender-fair) and job ads will be randomly allocated to one. We only use full-time positions for jobs not already advertised using gender-neutral titles by the hiring company within each of the above-mentioned industry categories.

Design details

Randomization method
The server of the platform assigns each new incoming job ad with the letter A or B with probability 50%. After that, each ad is checked for being in one of the above-mentioned industries and the following criteria: Fulltime job or fulltime traineeship. If the letter A was assigned, the job ad will be displayed using generic male titles (baseline), if the letter B was assigned, gender-fair titles will be used (gender-fair) throughout for the respective ad.
Randomization unit
Data is organized at the job-ad-level, that is, each job ad that passed through the procedure described in the randomization method is one observation with the number of applications, views and clicks as variable expressions.
Clusters
n.a.
Observations
Data collection will seize once the last treatment-industry combination has 300 observations. (Total: 2x3x300=1800 observations)
Arms
Data collection will seize once the last treatment-industry combination has 300 observations. (Total: 2x3x300=1800 observations) We will allocate job ads randomly to each treatment.
Primary outcomes
female_ratio treatment indcat
Secondary outcomes
Jobtitle views_spreading_count views_native_duration requirements_appl size_comp jobad_postdate jobad_closedate alt_titles

Cited by

No citation data available for this registration yet.

Keywords