Thirteen years after. Using hierarchical linear modeling to investigate long-term assessment center validity

Using hierarchical linear modeling the author investigated temporal trends in the predictive validity of an assessment center for career advancement (measured as salary growth) over a 13-year period, for a sample of 456 academic graduates. Using year of entry and tenure as controls, linear and quadratic properties of individual salary curves could be predicted by the assessment center dimensions. The validity of the (clinical) overall assessment rating for persons with tenure of at least 12 years corrected for initial differences in starting salaries, and restriction in range was .35. Corrected for predictor unreliability it became .40. In particular substantial long-term validities were obtained for the group discussion and analysis/presentation exercises. In accordance with findings from research in managerial effectiveness and development, the validity of the dimension firmness increased with tenure. Validity trends for two types of interviews and a mental test were also studied. The analytical and verbal ability dimensions of the mental test had negative long-term validities.