Oya Somer

The structure of Turkish trait-descriptive adjectives.

This description of the Turkish lexical project reports some initial findings on the structure of Turkish personality-related variables. In addition, it provides evidence on the effects of target evaluative homogeneity vs. heterogeneity (e.g., samples of well-liked target individuals vs. samples of both liked and disliked targets) on the resulting factor structures, and thus it provides a first test of the conclusions reached by D. Peabody and L. R. Goldberg (1989) using English trait terms. In 2 separate studies, and in 2 types of data sets, clear versions of the Big Five factor structure were found. And both studies replicated and extended the findings of Peabody and Goldberg; virtually orthogonal factors of relatively equal size were found in the homogeneous samples, and a more highly correlated set of factors with relatively large Agreeableness and Conscientiousness dimensions was found in the heterogeneous samples.

The hierarchical structure of common Turkish person-descriptive adjectives

Findings from analyses of self‐descriptions by 631 native speakers of Turkish, using 498 familiar Turkish person‐descriptive adjectives, are compared to those of Saucier (1997), who analyzed 500 familiar English adjectives. In the total item pools in both studies, variants of the English/German ‘Big‐Five’ factors were recovered, along with a broad Attractiveness factor and another factor composed of items of extremely low endorsement rates. Moreover, in both studies when the item pools were restricted to terms that are less pejorative and more clearly related to personality traits, again variants of the Big‐Five factors were recovered. These latter findings replicate those from Somer and Goldberg (1999), who analyzed Turkish synonym clusters rather than single person‐descriptive adjectives. We discuss some of the pitfalls of comparative studies of personality lexicons, and suggest some procedures for reconciling diverse findings. Copyright

Six Turkish Personality Factors and the HEXACO Model of Personality Structure

This article examines the six largest dimensions of the Turkish personality lexicon, as derived from the personality-descriptive terms of Goldberg and Somers (2000) data set. The six Turkish lexical factors show close correspondences of content to all six dimensions observed in several other languages. In a new data set, the authors then correlate factor scores representing these indigenous Turkish dimensions with the scales of the HEXACO Personality Inventory, which operationalize the six cross-language factors. Results show a pattern of strong convergent and weak discriminant correlations. Overall, findings suggest that the cross-language six-dimensional structure of personality description does generalize to the Turkish lexicon. The Turkish structure also reveals some interesting emic features, particularly with regard to the content of the Openness to Experience factor.