Workplace Diversity and Inclusion

How to cite

Flory, Jeffrey; List, John; Hickman, Brent. 2021. "Workplace Diversity and Inclusion." AEA RCT Registry. AEARCTR-0008020. https://doi.org/10.1257/rct.8020-2.0

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

Jeffrey Flory* , John List , Brent Hickman

* Primary investigator

Registry ID
AEARCTR-0008020
DOI
https://doi.org/10.1257/rct.8020-2.0
First registered
Last updated
Status
In development
Trial dates
2024-03-15 – 2024-12-01
Countries
Private
JEL codes
J15, J16, C93, D22
Registry record
http://www.socialscienceregistry.org/trials/8020

Abstract

To advance our understanding of diversity and inclusion in labor markets, this field experiment will examine the effects of workplace diversity on worker behavior and employee composition, and identify channels driving the effects, using actual workers in a real workplace. Employee diversity, defined as demographic variation among workers (primarily race and gender), has been argued to have significant impacts on worker- and firm-level outcomes such as employee behavior, recruiting, and productivity of companies and organizations. Yet much of this is theoretical and conjectural: rigorous empirical evidence on the existence and extent of impacts, as well as the causal mechanisms through which they occur, is lacking. This field experiment fills that gap by generating new causal evidence that tests for effects, measures their size, and identifies their channels. Specifically, the project will examine the impacts of racial diversity and gender diversity among workers on diversity and skill level of applicants, worker performance, labor supply, and other critical dimensions of worker behavior and characteristics. It will also identify the extent to which these impacts derive from (i) effects on the types of people that apply to a job (e.g. what types of workers are attracted to diverse or inclusive workplaces); (ii) effects on the types of people that stay at the firm (employee retention); and (iii) effects on behavior from working in a diverse or inclusive environment separate from impacts on types of employees recruited or retained. (The first two channels are often referred to as “selection effects” and the third as “treatment effects”.) Tightly linking the experiment design to a structural model will ensure the insights we uncover feed directly into our existing body of knowledge and help expand theory to better address questions and problems relating to diversity and inclusion.

Experimental design

We advertise a job with variation in employee diversity and wages and use data generated during the work-stage of the experiment to determine impacts of this variation on worker selection. (Randomization at individual level, between-subjects.) We also exogenously vary the diversity environment of workers while they are working, to measure impacts of variation in coworker diversity on production characteristics (speed, accuracy, labor supply, cooperation). (Randomization at the shift-level, within-subject.) The work is done in teams, with a joint production output.

Design details

Randomization method
Randomization done in office by a computer.
Randomization unit
Hiring Stage: Individual level randomization. Work-Stage: Individual + Shift level randomization
Clusters
Hiring Stage: 10,000 Work Stage: 600
Observations
Hiring Stage: 10,000 Work Stage: 600
Arms
Hiring Stage: 2,500 non-diverse low wage, 2,500 non-diverse high wage, 2,500 diverse low, 2,500 diverse high
Primary outcomes
productivity (time to produce each unit of output), accuracy (number of errors per unit of output), labor supply (number of hours worked), cooperation (agreement level in teams), application probability, as well as four underlying parameters in a structural model specifying each productivity dimension (productivity, accuracy, labor supply/leisure, cooperation)

Cited by

No citation data available for this registration yet.

Keywords