An important issue in the design of stated-preference surveys is whether the information provided to respondents within a survey instrument is adequate to yield valid value estimates. Providing respondents with on-site experience about forest ecosystem management alternatives may influence their expectation of the effects from new policies and programs. In the research reported here, we investigate whether preference parameters for attributes of low-impact timber harvesting programs differ between respondents to a mail survey versus respondents provided with an on-site forest experience (walk through a research forest). The empirical analysis in our application shows that stated preferences for timber harvesting attributes are not statistically different between the mail and on-site applications of the survey, and this result is robust to pretest (before experience) and post-test (post experience) applications.