discussion paper

A multi-country validation and sensitivity analysis of the project level Women’s Empowerment in Agriculture Index (Pro-WEAI)

by Greg Seymour,

Simone Faas,

Nathaniel Ferguson,

Jessica Heckert,

Hazel J. Malapit,

Ruth Meinzen-Dick,

Agnes R. Quisumbing and

Chloe van Biljon

ifpri discussion paper | 2023 | Pages: 49

DOI : https://doi.org/10.2499/p15738coll2.136974

Open Access

Citation

Seymour, Greg; Faas, Simone; Ferguson, Nathaniel; Heckert, Jessica; Malapit, Hazel J.; Meinzen-Dick, Ruth; Quisumbing, Agnes R.; van Biljon, Chloe; and Gender Agriculture Assets Project Phase 2 Study Team. 2023. A multi-country validation and sensitivity analysis of the project level Women’s Empowerment in Agriculture Index (Pro-WEAI). IFPRI Discussion Paper 2201. Washington, DC: International Food Policy Research Institute (IFPRI). https://doi.org/10.2499/p15738coll2.136974

We discuss the evolution of the project-level Women’s Empowerment in Agriculture Index (pro-WEAI) from its initial launch in 2018 until early 2023. We explain the reasons motivating changes to the composition of pro-WEAI and the adequacy thresholds of several indicators and discuss the implications of both for the overall measurement of project impacts on women’s empowerment. We present supporting empirical results comparing projects’ impacts calculated using the abbreviated Women’s Empowerment in Agriculture Index (A-WEAI) (the predecessor to pro-WEAI with fewer indicators and less stringent indicator cut-offs), the pilot 12-indicator version of pro-WEAI, and the final, revised 10-indicator version of pro-WEAI, based on longitudinal data from six agricultural development projects in East and West Africa and South Asia as part of the Gender, Agriculture, and Assets Project, Phase 2 (GAAP2). In addition, we assess the sensitivity of the revised pro-WEAI to an alternative weighting scheme, namely inverse covariance weighting (ICW). Overall, we find that the revised pro-WEAI performs well: In comparison to A-WEAI, pro-WEAI—regardless of version—identifies larger and more frequently significant impact estimates, indicating that pro-WEAI is more sensitive to detecting project impacts on women’s empowerment than A-WEAI. And we find only minor differences in impact estimates produced using the 12-indicator, 10-indicator, or alternate weighting scheme versions of pro-WEAI. We conclude with reflections on six years of work on pro-WEAI during GAAP2.