Measuring women’s decisionmaking: Indicator choice and survey design experiments from cash and food transfer evaluations in Ecuador, Uganda and Yemen

•Women’s decisionmaking indicators are widely used as a proxy for empowerment, though insufficient attention is given to measurement.•Using survey design experiments in three countries, we find that small differences in construction and wording of indicators matter.•These differences change how women are ranked in terms of decision making as well as evidence of program impacts.•Further measurement innovation utilizing context-specific understanding of intra-household dynamics is needed. Women’s decisionmaking indicators are widely used in social science research, though insufficient attention is given to measurement issues. We interrogate variations in indicator construction using survey experiments undertaken in the context of transfer programs in Ecuador, Yemen, and Uganda. Findings show that small variations can lead to meaningful differences in how women are ranked on decisionmaking, as well as change conclusions on whether programs have significant impacts on decisionmaking. Results also raise questions as to how well indicators capture dimensions of empowerment as they are not consistently associated with measures of women’s status. Finally, in the context of Uganda, results indicate that measures are susceptible to social desirability bias, however further investigation is needed to assess this potential. We conclude that choices made in indicator construction matter and in the absence of standardized guidance, current practice is ad-hoc. Further, despite the wide use of these indicators, evaluating program success on women’s empowerment solely through standard decisionmaking indicators is not recommended. Finally, measurement innovation utilizing context-specific understanding of intra-household dynamics is needed.