SUMMARY The issue of free entrance is debated in many countries. We investigate changes in visitor composition associated with an introduction of an entrance fee to a state funded museum, and the validity of the Contingent Valuation (CV) method. We conducted two surveys to collect information about the visitors' socio‐economic backgrounds, one before and one after the fee was in effect. While entrance was still free, we also asked visitors about their willingness to pay for a visit, using the CV method. We then compare the results of the CV survey with the actual change in visitor composition caused by the fee. We thus have a unique opportunity to test the validity of the CV, which, as far as we know, has never been done in a similar way before. The results of the CV indicate that several target group visitors are less likely to visit the museum after an implementation of a relatively low fee. Consequently, charging for entrance does affect who visits the museum. The validity test of the CV method shows that a majority of the changes in visitor composition were correctly predicted. We conclude that the CV method is particularly successful when applied on goods familiar to the respondents.