How to make policy assessment is one of the biggest issues for local governments. Questionnaire surveying has been one of the main means to surface the needs from local residents for the local government or community and the survey results have been utilized to draw out concrete policies or measures in almost all local governments in Japan. However, an empirical research made by our research group has suggested that questions which ask satisfaction and importance level for residents' living environment items by means of ordinal scale tend to overestimate their needs due to external effects from local public facilities and services such as option value and prestige value and thus we proposed the introduction of quantification methods such as CVM(Contingent Valuation Method), AHP (Analytic Hierarchy Process) and Conjoint Analysis to surface local issues and make local community management effective. In this study, we considered the policies in local governments to be a package of concrete measures for attaining goals in various policy fields and verified the applicability of conjoint analysis to policy assessment. Based on a questionnaire survey we conducted in Tatsuno, Hyogo Prefecture in 2006, we maintain that conjoint analysis is a viable method to grasp the preferences of residents for policies made by local governments. The results of conjoint analysis are presented and some tips for carrying out conjoint analysis through the questionnaire survey which we conducted. As a further extension of this study, we plan to propose a viability of GIS based feedback system of questionnaire results to local residents. The goal of our research study is to verify the concept of “Informed Decision by Informed Citizens through GIS” and support local residents to participate all the steps of local community management process: Plan, Do and See or Check and Action. JEL Classification: C35, R22, R58