The Women's Agency Scale 61 (Was-61): A Comprehensive Measure of Women's Intrinsic, Instrumental, and Collective Agency

Background. Sustainable Development Goal (SDG) 5 prioritizes women's empowerment, alone and to advance other SDGs. Measures of women's empowerment have been resource focused; yet, agency (an aspect of empowerment) captures conscientization — the emergence of critical awareness and action for change. A comprehensive measure of women's agency would capture intrinsic power within to overcome dominated consciousness, instrumental power to pursue aspirations, and collective power with others to pursue shared goals. A valid measure of women's multidimensional agency is needed to monitor real progress toward SDG5 and women's well-being. Methods. In this sequential, mixed-methods study, we developed and validated the Women's Agency Scale 61 (WAS-61) in 930 married women 15–49 years and living with their husband in Matlab, Bangladesh. Using theory, qualitative research, and questionnaire reviews, we developed an item pool to capture women's intrinsic (43 items), instrumental (23 items), and collective (17 items) agency. We performed exploratory then confirmatory factor analyses (EFA/CFA) in random split-half samples, and estimated pairwise and factor score correlations to confirm that the dimensions of agency were distinct but related in expected ways. In future work, we will perform multiple-group CFA to assess to what extent the WAS-61 is measurement invariant across salient sub-groups of women. Results. A 29-item, three-factor CFA model measured women's intrinsic agency as endorsement of gender equity and rights (non-justification of wife beating, gender-equitable attitudes) and confidence in capabilities (comfort expressing opinion and going places outside home) (loading: 0.35-0.93, RMSEA=0.04, CFI=0.94, TLI=0.93). Significant pairwise factor correlations (0.10–0.55) confirmed that the intrinsic-agency dimensions were distinct but related. A 27-item, three-factor CFA model measured women's instrumental agency as use of financial services, voice with husband, and mobility/voice outside home (loading: 0.43-0.99, RMSEA=0.05, CFI=0.94, TLI=0.93). Significant pairwise factor correlations (0.36–0.58) confirmed that the instrumental-agency dimensions were distinct but related. To capture women's collective agency, an 8-item, one-factor CFA model measured leadership of others (loading: 0.52-0.84, RMSEA=0.10, CFI=0.97, TLI=0.96), and a 7-item, one-factor CFA model measured influence in their community (loading: 0.54-0.83, RMSEA=0.10, CFI=0.96, TLI=0.95). Significant, and generally higher, within-construct than cross-construct pairwise correlations suggested that the dimensions of women's agency were distinct but related (intrinsic: 0.13–0.60; instrumental: 0.40–0.72; collective: 0.58; cross-construct: 13 of 21 correlations < 0.30). Intrinsic agency in mobility/voice outside home was strongly correlated with measures of instrumental and collective agency (0.28–0.51). Instrumental agency in use of financial services and mobility/voice outside the home were strongly correlated with measures of collective agency (0.38–0.47). Conclusions. The WAS-61 is a comprehensive measure of women's multidimensional agency validated in married women in rural Bangladesh. These novel scales capture well-defined dimensions of intrinsic, instrumental, and collective agency that are distinct but related. Women's intrinsic agency in voice/mobility outside the home is strongly related to their instrumental and collective agency, and women's instrumental agency in use of financial services and voice/mobility outside the home is strongly related to their collective agency. These findings suggest that all dimensions of women's conscientization are inter-related and necessary for women's empowerment and well-being. We will test measurement in-variance of the WAS-61 across salient sub-groups of women, and the WAS-61 should be validated in urban women and other samples to confirm its widespread utility for impact assessment of health and development programs and national monitoring. Evidence of time in-variance of the WAS-61 would confirm its utility to monitor national and program-level changes in women's agency and its influences on other SDGs.