Gender Peer Effects in the Labor Market: A field experiment in Indian call centers

How to cite

Batheja, Deepshikha. 2019. "Gender Peer Effects in the Labor Market: A field experiment in Indian call centers." AEA RCT Registry. AEARCTR-0003932. https://doi.org/10.1257/rct.3932-1.0

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

Deepshikha Batheja*

* Primary investigator

Registry ID
AEARCTR-0003932
DOI
https://doi.org/10.1257/rct.3932-1.0
First registered
Last updated
Status
On going
Trial dates
2018-11-01 – 2019-07-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/3932

Abstract

Does the gender of one’s co-workers/peers have an impact on labor productivity? This question has primarily been studied in the developed country setting. However, this research question is particularly important in patriarchal and traditional societies where gender is salient. In these settings, gender segregation is practiced widely such that, in many cases, the first prolonged interaction with the opposite gender occurs in a workplace. So, social interaction among opposite genders is likely to lead to psychic costs in the workplace (Akerlof and Kranton 2000; Bertrand et al., 2015). This can negatively impact a firm’s output if it comprises a gender diverse employee pool.

The proposed individual-level randomized controlled trial will study the effect of gender diversity in workplace on worker productivity of male employees, their job satisfaction levels and their gender attitudes. Treatment in the form of gender exposure will be provided to men by varying the gender composition of teams in call centers in India. The productivity of male employees in mixed teams (treatment) and all male teams (control) will be compared for a period of 2-3 months.

Experimental design

A typical domestic call center in India works for many processes/businesses. The workers in each process are divided into teams, which are helmed by team leaders. This study will rely on individual level randomization of workers in a process into teams of varying gender compositions. The treatment or mixed gender group (40-50% male) will be compared to control group of all males. Agents will be matched on baseline productivity and then randomly assigned into treatment and control arms. Team leaders will also be matched on productivity before being randomly assigned to treatment and control teams. The seating of agents will be assigned in a manner that all teammates sit together. In mixed teams, there will be alternate seating of men and women.

Design details

Randomization method
randomization done in office by a computer
Randomization unit
individual level randomization
Clusters
532 workers
Observations
532 workers
Arms
355 workers in treatment/mixed group (177 males and 177 females) 177 male workers in control group
Primary outcomes
1) Productivity or the quantity variable (varies for each process/business. In sales process, it is number of sales made in a day. In inbound processes, it is average call handling time and in outbound processes, it is number of calls made. As I have multiple processes in my study, I will normalize the scores within each process to make the productivity variable comparable across processes. 2) Quality of calls, rated by quality audit team internally at the call center.
Secondary outcomes
1) Job satisfaction level- collected pre-intervention (baseline) and post-intervention (endline) through survey, 2) Gender attitude - collected pre-intervention (baseline) and post-intervention (endline) through survey.

Cited by

No citation data available for this registration yet.

Keywords