Öz The objective of this study is to determine the most significant variables for distinction of household types and rural — urban areas. In this study, raw data of household budget survey of the year 2003 was used. Discriminant analysis was applied to this data set and household type and rural-urban area distinction were examined considering several independent variables. Discriminant analysis is multivariate statistical techniques and concerned with separating distinct sets of objects or observation. As a result of this study it was find out that the equivalent person size and food expenses are most significant variables for household type determination while non-food expenses and equivalent person size variables are most important variables for rural-urban distinction determination.