We compare two methods of stated preference valuation techniques -dichotomous choice contingent valuation and choice experiments – to test for the prevalence of embedding. The test is based on a survey of willingness to pay for the preservation of heath areas in Denmark. Using only one outcome alternative and a status quo in the choice experiment we can compare the two methods using exactly the same estimation procedure. Thereby the differences found can be attributed to the different questioning formats alone. We find differences between the methods, especially regarding the sensitivity to scale.