Assessing adult farm labor statistics: Evidence from a survey design experiment in Ethiopia

Abstract

Agricultural labor accounts for most of the employment in Sub-Saharan Africa. Yet, few empirical studies have examined the reliability of farm labor measures emerging from household surveys. In this study, we report the results from a survey design experiment in rural Ethiopia that focus on the effects of survey respondent i.e., proxy vs self-report, on adult farm labor. Our results show that proxy respondents generate lower farm labor statistics relative to self-responses for men but not for women. The magnitude of the impacts for men reaches 3.9 percentage points or 4.3% lower rate of participation relative to the mean participation obtained from the self-report. Our results have implications for the current debate on the measurement of agricultural productivity and the collection of survey data in rural areas of developing settings.