Cooperation and Gender Discrimination

How to cite

Wiborg, Vegard Sjurseike; Nyborg, Karine; Lind, Jo Thori; Brekke, Kjell Arne. 2021. "Cooperation and Gender Discrimination." AEA RCT Registry. AEARCTR-0007869. https://doi.org/10.1257/rct.7869-1.0

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

Vegard Sjurseike Wiborg* , Karine Nyborg , Jo Thori Lind , Kjell Arne Brekke

* Primary investigator

Registry ID
AEARCTR-0007869
DOI
https://doi.org/10.1257/rct.7869-1.0
First registered
Last updated
Status
In development
Trial dates
2021-06-24 – 2021-08-10
Countries
Private
JEL codes
C99, J71, A13, D83
Registry record
http://www.socialscienceregistry.org/trials/7869

Abstract

In this experiment we test whether informativeness of the ordering of collaborators matters for the extent of gender discrimination. We match individuals (candidates) who have previously performed mathematical quizzes, in pairs, and ask a new group of subjects to pick (the best) candidates based on information about pair performance, gender, age and type of university degree. Subjects are paid based on their chosen candidates' individual performance on a mathematical quiz. First, we vary the informativeness of a candidate's mathematical performance by ordering pair members either alphabetically (ABC treatment) or according to merit (123 treatment), listing first the pair member with the highest number of correct answers. Second, we measure the presence of gender discrimination by varying the gender compositions of candidates in the choice sets that subjects choose from. In this plan we describe hypotheses to be tested, the coding of variables, and the empirical strategy that will be used.

Intervention

In our study, we ask subjects to choose candidates based on information about the candidates' performance on a set of mathematical exercises together with a partner. Subject's earnings depend on their chosen candidates' individual performance on a math quiz.

We implement interventions simultaneously. In intervention 1 we randomly vary the information subjects receive about the candidates' individual contributions to pair work. In intervention 2, we randomly vary the gender composition of choice sets (group of candidates from which subjects choose one) in order to study how gender discrimination of candidates depend on the information scheme.

Experimental design

In this experiment, Experiment 2, subjects will evaluate the mathematical abilities of a second group of individuals, from hereon called candidates, that participated in a previous experiment, Experiment 1. First, we provide subjects with information about the candidates' mathematical performance together with a partner (a pair) and some demographic characteristics. Second, we ask them to pick a subset of these candidates. The subjects in Experiment 2 are paid based on the number of correct answers their chosen candidates provided on a particular mathematics quiz in Experiment 1. In total, subjects in Experiment 2 make 13 choices of candidates from 13 choice sets, each consisting of four eligible candidates.

We randomly vary the information subjects receive about candidates' individual contribution to pair-work (intervention 1) to gauge the effect of productivity relevant information on candidate choices. We also vary the gender composition of choice sets (intervention 2) to estimate the effect of information on gender discrimination.

Design details

Randomization method
Randomization performed by computer.
Randomization unit
Subjects are randomized to our between-subjects treatments on the individual level.
Clusters
500 subjects
Observations
500 subjects
Arms
In each treatment arm of our 2*2 design there are 125 subjects. Subjects are allocated to each treatment in intervention 1 with equal probability, and within each of these treatments, subjects are allocated to the two treatments in intervention 2 with equal probability.
Primary outcomes
The main outcome variables are whether a candidate is chosen, the gender of the chosen candidates and the performance of chosen candidates on the payoff relevant quiz.

Cited by

No citation data available for this registration yet.

Keywords