The purpose of this study was to standardize a Big Five personality inventory for generations separately. At first, Japan was divided into seven blocks, and forty-six local administrative areas in 23 prefectures were extracted. Then, four thousand and one hundred (4100) men and women were randomly selected from the areas' resident registers. In some blocks, like Shikoku and Okinawa, telephone directories were used instead. Forms with request for cooperation were sent by mail, and 1166 valid forms were returned. The respondents were divided into four generations: adolescence (12-22 years old), youth (23-39), middle age (40-59), and elderly (60 years old and over). An ' F' scale was constructed, to measure the tendency to choose infrequent responses. One-way ANOVAs were performed to check generation differences for each of Big Five scales, for men and women separately. No sex difference was found in fundamental scales, except O score. So, normalized T-scores were calculated on the cumulative proportions. E score was high in adolescence and elderly, but the higher the generation, the higher the other (A, C, N. and O) scores. These results were discussed in the context of societal demands and socially desirable responding.

Key words: Big Five personality inventory, generational standardization, adolescence norm, youth norm, middle age norm, elderly norm