This paper investigates the effects of irrational responses on stated willingness-to-pay (WTP) in a contingent valuation study. A significant portion of the respondents stated that they were willing to pay a higher price for irradiated beef which they would avoid consuming due to their concerns about the side effects of irradiation. Such responses may not reflect true WTP and may cause bias in WTP estimate. Excluding these responses from estimation may result in sample selection bias. Whereas setting the bid values presented to these responses close to zero both helped to improve the estimation model and to reduce the potential bias in WTP estimate.