Supporting women workers in factories: experimental evidence from India

How to cite

Gadhok, Ishrat; Amengual, Matthew; Guasti, Alessandro. 2024. "Supporting women workers in factories: experimental evidence from India." AEA RCT Registry. AEARCTR-0013620. https://doi.org/10.1257/rct.13620-2.0

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

Ishrat Gadhok* , Matthew Amengual , Alessandro Guasti

* Primary investigator

Registry ID
AEARCTR-0013620
DOI
https://doi.org/10.1257/rct.13620-2.0
First registered
Last updated
Status
On going
Trial dates
2023-12-04 – 2025-12-31
Countries
Private
Registry record
http://www.socialscienceregistry.org/trials/13620

Abstract

Multinational companies (MNCs) are increasingly implementing social initiatives that aim to improve the health and wellbeing of production workers in their supply chains. Yet, there is limited rigorous evidence on their effectiveness. We collaborate with a large MNC in the apparel sector to conduct a cluster randomised controlled trial (RCT), to study the effects of one such program on outcomes among women employed in Indian factories that are in the MNC's supply chain. This study will evaluate the program's impact on behavioural outcomes, particularly attrition. Additionally, we will investigate the program's effects on self-reported outcomes such as health literacy, health status, and embeddedness in the factory.

Experimental design

This is a cluster randomised trial in the supply chain of an MNC. We use administrative data provided by the MNC to draw up a list of factories (clusters) that are eligible for the study. Ultimately, we had a sampling frame of 118 factories producing apparel, textiles, footwear, and accessories. We then sampled 75 factories proportionally to the strata size using the following criteria: size (above and below 500 workers), and geography (north and south).

Design details

Randomization method
The unit of randomisation of this cluster trial is the factory. To maximise the precision of our estimates, we used a matched quadruplet design following recommendations from the literature. We first calculated the Mahalanobis distance between factories on a set of blocking covariates; we then matched factories that were as similar as possible into quadruplets (18 quadruplets and 1 triplet). We randomly assigned the treatment to two factories in each quadruplet. We did the randomisation in R. Our blocking variables include the stratification variables of geography (North-South) and size, as well as the two factory-level outcomes of interest that we had access to at the time of randomisation: average turnover rates and average absenteeism.
Randomization unit
factory
Clusters
75 factories
Observations
4500 women workers
Arms
38 factories in treatment group, and 37 factories in the control group
Primary outcomes
retention in the factory; health literacy; health status; and embedded-ness in the factory
Secondary outcomes
absenteeism; returning to the factory; job satisfaction; empowerment at home, empowerment at work; subjective wellbeing

Cited by

No citation data available for this registration yet.

Keywords