This paper utilizes panel data from rural India on a daily rated labour market to examine how workers deal with the uncertainty which arises from the existence of involuntary unemployment. In particular, it measures the relative explanatory power of the three models which the literature suggests should better fit the conditions prevailing in this market: the Expected Utility Model (EUM) with linear objective probabilities, the EUM with linear subjective probabilities and the EUM with non-linear subjective probabilities. The econometric analysis indicates that the EUM with linear has to be preferred to that with non-linear probabilities. Moreover, it supports the hypothesis that the decision making model is under uncertainty and that people work out their subjective probabilities through their past experiences. Finally, agents in the sample turn out to be risk-averse and not to have a positive reservation wage.