Using machine learning and qualitative interviews to design a five-question women's agency index

Abstract

We propose a new method to design a short survey measure of a complex concept such as women's agency. The approach combines mixed-methods data collection and machine learning. We select the best survey questions based on how strongly correlated they are with a "gold standard" measure of the concept derived from qualitative interviews. In our application, we measure agency for 209 women in Haryana, India, first, through a semi-structured interview and, second, through a large set of close-ended questions. We use qualitative coding methods to score each woman's agency based on the interview, which we treat as her true agency. To identify the close-ended questions most predictive of the "truth," we apply statistical algorithms that build on LASSO and random forest but constrain how many variables are selected for the model (five in our case). The resulting five-question index is as strongly correlated with the coded qualitative interview as is an index that uses all of the candidate questions. This approach of selecting survey questions based on their statistical correspondence to coded qualitative interviews could be used to design short survey modules for many other latent constructs.

How do Gender Norms Shape Education and Domestic Work Outcomes? The Case of Syrian Refugee Adolescents in Jordan

“…Disconnects between perceptions and realities of norms can play a particularly important role in gender inequality (Bursztyn, Gonzalez, & Yanagizawa-Drott, 2020). Mixed-methods research will also be particularly important to developing better measures of gender norms (Jayachandran, Biradavolu, & Cooper, 2021). Some of the interesting but puzzling findings in our work relate to the justification of domestic violence.…”

Hitting the target and missing the point? On the risks of measuring women’s empowerment in agricultural development

“…There are some promising ways forward to this end, in both application and conceptually. Jayachandran et al (2021) propose a methodology that combines machine learning with qualitative interviews to design a five-question women's agency index. Glennerster et al (2018) outline a four-step methodology for developing locally tailored empowerment indicators using findings from formative qualitative research.…”