Sep 06, 2022

Media Coverage: Adapting and Validating WEE Indicators in an Experimental Study of Savings

Media Coverage: Adapting and Validating WEE Indicators in an Experimental Study of Savings

This measurement study was originally published on BRAC Institute of Governance and Development


This measurement study is part of a larger randomized controlled trial (RCT) that aims to produce evidence on how goal-based digital savings accounts and services might improve the financial health and well-being of low-income people in Uganda. Preliminary data from this study revealed that women were more responsive to financial coaching than men and that women’s savings balances were higher than men’s. This discovery is in line with existing research that illustrates the potential for digital financial services (DFS) to empower women. However, in order to advance this understanding, more work is needed to improve and expand the measurement of women’s economic empowerment (WEE) indicators with a focus on context-specific indicators.


Layering a measurement study onto the existing RCT will help uncover why women are more responsive than men to financial coaching and savings. Additionally, the study will identify how these interventions contribute to women’s agency, defined as setting goals, acting on them, and achieving them. The measurement study will be informed by qualitative research that is already underway. Utilizing input from two complementary investigative techniques, focus group discussions (FGD) and cognitive interviews (CI) iteratively, indicators are being developed and adapted to a Ugandan context. The measurement study involves increasing the female sample size in the RCT from 400 to 1,600 women to allow for more variables associated with WEE to be included in the analysis. The women will be evenly split between a treatment and a control group. In the treatment arm, the sample population will receive personalized financial coaching emphasizing the importance of saving habits. Finally, the third stage will consist of construct validation using machine learning algorithms that will identify which WEE indicators have the greatest statistical power and the strongest ability to predict women’s empowerment.

Proposed impact

This measurement study seeks to contribute an improved approach to measuring WEE. By quantitatively testing existing WEE indicators and adapting them to a localized context, this study will demonstrate a pathway for future research to adapt and strengthen the testing and understanding of WEE. Furthermore, this research will contribute to the existing literature that describes women’s agency in decision-making in three steps: setting goals, acting on them, and achieving a desired objective. The final aim of this measurement study is to offer robust, double layer validated (content and construct validated) WEE indicators that have strong classification power in profiling Ugandan women according to their ability to set, act-upon, and achieve savings goals.