부산대학교 도서관

Data sets from five original trait taxonomies from different languages, Dutch, Italian, Hungarian, American English, and German, are used for a cross-cultural study. Factor structures, presumably Big-Five structures, within these languages, based on ratings from about 400–800 subjects on approximately 500 trait variables per language, are used for comparison. The five five-factor structures are pairwise compared. To enable these comparisons, the best one-to-one translations of trait terms between the languages are used. These translations yielded subsets of the data for direct comparisons. Congruence coefficients were calculated for the corresponding factors in the different languages, based on their independent positions, on their positions after target rotations, and on their positions after simultaneous rotations. The congruences show replicability of the first three factors of the Big Five across the five languages, and replicability of the fourth factor among Dutch, Italian, and Hungarian. Relatively high congruences between all five factors were found between Italian and Dutch. © 1997 John Wiley & Sons, Ltd. [ABSTRACT FROM AUTHOR]