Theoretical and Methodological Considerations in the Translation of the 16PF5 into an African Language

Abstract

A Tshivenda version of the Sixteen Personality Factor Questionnaire Fifth Edition (16PF5) was developed using a back-translation design. The translated version was administered to a Tshivenda-speaking sample. Analyses were done at an item level, specifically in terms of the contribution of the items to the reliability of the 16 primary factors. Results indicated that many of the items would have to be excluded if the aim were to achieve acceptable reliability, and even then the reliability coefficients would remain low. In the case of a number of the ineffective items, translation changed the meaning of the items because of the absence of an equivalent concept in the target language, difficulty in translating idiomatic expressions, potential confusion due to the use of the negative form, and translation errors. Trends that indicate cultural differences in the manifestation of constructs were also observed and these were related to cultural norms (e.g., the expression of emotions) and experiential factors (e.g., job-related experience). A literal translation of the questionnaire proved to be insufficient and the findings of this study need to be considered in further adaptations of the test. The theoretical and methodological issues highlighted here have implications for the translation of this questionnaire into other African languages as well as for the adaptation of personality questionnaires in general.