Two questions are raised about the accuracy of the rates of disguised unemployment in rural Bangladesh estimated on the basis of farm surveys using ‘labour utilization’ or ‘gainful work’ approach. One is the possible exclusion of migratory labour use from labour utilization data. The other is the adequacy of ‘work duration’ as the sole basis for measuring employment. Results of a large survey throughout the country show that seasonal migration of rural labour is a more common phenomena than is ordinarily recognized and it needs to be explicitly considered in labour tilization/unemployment estimation. Another survey in eight villages show that work duration has little relationship with income, so on its own, it may not be an adequate criteria for measuring unemployment.