We are interested in documenting gender gaps in Ethiopia by observing the administrative sector at a large University and decomposing factors that may contribute to the wage (and position) gap using institutional data. We propose to rigorously explore two potential explanations for the gender gap: 1) whether the evaluation of potential job candidates are a function of the candidate's gender, and 2) whether following advice is a function of the gender of the advisor.
Intervention
We randomly assign gender to resumes and advisors to measure differences in appraisals and willingness to follow advice.
Experimental design
The experiment randomizes gender on resumes and of "team leaders" an experimental logic games.
Design details
Randomization method
Randomization done in office by computer.
Randomization unit
Individual
Clusters
For the resume component, sample size is 600. For the experimental game component, sample size is 300.
Observations
For the resume component, number of observations is 600 subjects * 2 resumes = 1200.
For the game component, number of observations will be at the subject level (300 observations) and per round of the game (300 subjets * 11 rounds) = 3300.
Arms
For the resume component, 300 individuals per treatment arm (female resume, male resume).
For the game component, 75 individuals per treatment arm (male + no ability signal, female + no ability signal, male + ability signal, female + ability signal). In addition, we will randomly determine the ordering of how beliefs are elicited (150 individuals per ordering).
Primary outcomes
Our key outcome variables for one experiment is competancy, likeability, willingness to hire, and salary offers.
Our key outcome variables for the second experiment is perceived ability of self, perceived ability of team leaders, and likelihood of following advice.
Cited by
No citation data available for this registration yet.