Does the gender of household survey respondents affect the accuracy of reported food consumption? Findings from a randomisation-based survey of beneficiary households in the Productive Safety Net Programme in Ethiopia suggest that it might – with implications for the targeting of such programmes. Females reported greater food insecurity and males reported higher food expenditures. Such striking gender differences raise the possibility that poor households may be classified as non-poor, or less poor than they actually are. Survey design and respondent selection should therefore aim to minimise bias to ensure the poorest and most food-insecure households benefit from assistance.
Authors
