Validity and reliability of the contingent valuation method (CVM), which refers to the ‘accuracy’ and ‘reproducibility’ of the CV results, have become increasing attention. However, there are still challenges. The monetary value of non-use value of Wuyishan tourism attraction was gauged by CVM method using willingness-to-pay elicited by payment card approaches. A test-retest procedure for payment card approaches was used in this case, and a random sample of native environmental assessment engineers were interviewed twice with an interval of around 6 months. These two surveys were named PC1 and PC2 in short. The reliability tested indicates that there is no statistically significant difference between the values obtained in two time periods. PC1 and PC2 had the similar results by way of regressing either the WTP value or positive WTP rates against standard socioeconomic characters(gender, age, monthly income etc), and PC1 and PC2 were good values resulted from WTP median and WTP average due to very small error, e.g, WTP median absolute error was only 158 yuan between PC1 and PC2, yet median and the mean WTP were higher in the test than in the retest.