Abstract

In this study, we developed and modified the Korean version of the NEO-PI-3 test, a revision of a test measuring personality based on one of the most widely used personality theories, the Five-factor personality traits, to validate its structure. The process involved three stages: a preliminary test, a modified test, and a final test, with analyses conducted at each stage. Initially, with permission from the original copyright holders, content experts translated the original test to create the preliminary test. Data were collected from 1,934 employees and university students for the preliminary test. Based on this data, exploratory factor analysis, classical test theory, and item response theory-based item analysis were conducted to identify and modify items in subfactors with inadequate factor structures, leading to the creation of the modified test. The modified test, which was responded to by 1,280 participants, underwent confirmatory factor analysis to further modify items in subfactors that did not demonstrate satisfactory factor loadings, to construct the final test. The final test was then subjected to target rotation factor analysis to confirm that the test structure adhered to the Five-factor personality structure. Measurement invariance across genders was established, confirming weak invariance at the level of metric invariance. In addition, test-retest reliabilities of all factors were relatively high except for only sub-factor ‘values’. Finally, limitations and implications of this study were described in discussion part.