A Big Five personality inventory in two non‐Indo‐European languages

TLDR

The study reports two successful replications of the NEO‑PI in Estonian and Finnish, aiming to determine which parts of the Big Five inventory are universal versus language‑ or culture‑specific. Costa and McCrae’s NEO‑PI was adapted to these two Uralic languages. Both Estonian and Finnish versions of the NEO‑PI exhibit psychometric properties comparable to the original, with reliabilities slightly lower (11 % and 36 % respectively) and factor structures explaining 71.7 % and 67 % of variance, while also revealing language‑specific differences such as higher extroversion and conscientiousness among women and culturally distinct patterns of gregariousness and excitement seeking.

Abstract

In this study we report on two successful replications of a five‐factor personality inventory in two non‐Indo‐European languages, Estonian and Finnish, which both belong to the group of Uralic languages. Costa and McCrae's (1985) NEO Personality Inventory was adapted to these two languages. By all relevant psychometric parameters neither developed construct differs from the original construct: the reliabilities of only 11 per cent for the Estonian and 36 per cent for the Finnish subscale were lower than those of the respective NEO‐PI scales. The factor structure of both Estonian and Finnish inventories was very close to the five‐factor structure of the NEO‐PI, accounting for 71.7 per cent and 67.0 per cent of the variance, respectively. In spite of this generally good agreement, some language‐ or culture‐dependent differences were observed. Both Estonian and Finnish women were more extroverted and conscientious than men, compared with their English‐speaking counterparts. Also, some differences exist in the need for other people's company and excitement seeking. In the Balto‐Fennic culture gregariousness appears to presuppose some emotional stability and openness and excitement seeking is conceptualized more as a tool of rescuing from anxiety, depression, and low self‐esteem. This study is considered as a step towards determination of which parts of the most popular instrument for the measurement of the Big Five personality dimensions are truly universal and which parts of it are specific to a particular language and culture.