Abstract
Empirical studies have analyzed the destination image formation of potential tourists. This study examined visitor characteristics in order to predict destination images of on-site visitors. Factors influencing destination images and visitor characteristics were selected based on information from existing empirical studies. The method of this study used to search for the best subset of predictors by calculating all possible regression equations. Using the hierarchical solution, variables were entered into regression on the basis of predetermined logical scheme. According to the results, it was possible to determine post-visit images of a destination by classifying into three destination images. Age, frequency of visitation, information use, and familiarity were statistically significant variables in predicting tourism destination images. Advanced study is needed to improve destination images of the most significant predictor, information use.