This essay develops a new method to diagnose inconsistency in dichotomous choice contingent valuation with follow-up questions: in particular, downward bias in the mean WTP. It is shown that the previous methods aimed to explain this inconsistency in responses have ignored statistical inconsistency: non-perfect correlation between the initial and follow-up responses and thus have provided wrong predictions to explain respondents' inconsistency pattern. In addition, from an application of our method, it has been proven that one model can not encompass all other possible inconsistency patterns in responses. Test results show that the behavioral inconsistency patterns are different both within and between data sets