Abstract
M.Phil. (Industrial Psychology)
Personality assessments are often used in organisations and have shown to be valuable for selection, to maximise employee performance and to assist with training and development. Personality assessments are argued as providing invaluable information about individuals that may not be gained from other sources of assessment. Personality assessments are valued overall in the workplace because personality is known to be a predictor of workplace behaviours and outcomes. However, internationally and in South Africa, personality assessments are copyrighted. This can be potentially problematic because it hinders scientific progression and its contribution to the workplace due to a potential lack of reliability and validity. In addition it might hinder cross-cultural validation because assessment developers keep item-factor correspondence and scoring keys in-house. This holds implications for the use of personality assessments in organisations, as the reliability and validity of these assessments might be difficult to investigate due to copyright laws. It is therefore argued that personality assessments should move toward an open-source platform in order to navigate the ongoing issues that occur with copyrighted assessments. Goldberg’s Big-Five Factor Markers is an open-source personality assessment that has shown encouraging results throughout research internationally. However, its psychometric properties in South Africa remain uninvestigated. Given that it is open-source it might be of value in terms of addressing issues surrounding scientific progression, and reliability and validity of personality assessments in the workplace. This study set out to investigate the psychometric properties of Goldberg’s Big-Five Factor Markers in the South African context. The main objectives of this study are as follows: (a) to investigate the reliability of Goldberg’s Big-Five Factor Markers scale scores, (b) to investigate the factor structure of Goldberg’s Big-Five Factor Markers items. To investigate Goldberg’s Big-Five Factor Markers, a sample of 224 participants were used. The analyses showed somewhat satisfactory reliability for research purposes but lower than that found in international studies. Factor analysis supported the construct validity with five-factors found and some of the items across all five factors were potentially problematic, with Extraversion performing well and Agreeableness performing poorly in comparison to the other five factors. These results were similar to other studies internationally and therefore suggest that the items that were found to be problematic would need to be revised and changed. Some...