Levels of child malnourishment in South Asia remained much higher than the good progress in developments indicators would suggest. These contradictory findings known as the Asian Enigma are caused by low female empowerment. The relationship between female empowerment and nutritional outcomes is the focus of this paper that uses the baseline data of a survey of 4,000 households in rural Bangladesh from October to December 2015 for the project Agriculture, Nutrition, and Gender Linkages (ANGeL) Project of the International Food Policy and Research Institute (IFPRI). This data allows to construct the recently developed Abbreviated Women’s Empowerment in Agriculture Index (A-WEAI), a multi-dimensional and agricultural-specific index based on the concept of agency, resources, and achievements. The score, its sub-indices of gender parity and five dimensions of empowerment, and three indicators including input in productive decisions, group membership, and workload are employed as independent variables. Another and partly contradictory measure to the A-WEAI is represented by the subjective measure of “feeling empowered”. Thus, this study investigates weather female empowerment is associated with an increase in nutritional status measured as dietary diversity of female, male, and an average household member, and, secondly, as relative gain of women’s nutritional status compared to men. Finally, it is hypothesized that the A-WEAI has greater empirical power than women’s perception of self-determination. The stated associations are estimated with multivariate linear regressions with district fixed effects. The results support the positive relationship of female empowerment and nutritional outcomes for female, male, and average household members. Female empowerment measured as the A-WEAI indicator work burden is positively associated with a relative increase of male dietary diversity compared to female. Therefore, self-determined “feeling of being empowered” seems to have greater empirical power in explaining nutritional outcomes than the multidimensional A-WEAI and its composite variables. Reasons for the weak performance of the A-WEAI in this study might be an endogeneity bias due to simultaneity, selection bias due to large number of inadequate male A-WEAI surveys, and perhaps weaker concept validity as the WEAI for statistical regression analysis. The results imply the necessity to include “felt empowerment” as additional performance indicator for interventions with the objective to increase female empowerment.