• This research was supported by grant No. 01103293 from the Research Council of the University of Ghent, awarded to Ivan Mervielde. Part of this manuscript is based on research by the first author submitted to the University of Ghent in partial fulfillment of Ph.D. requirementsunder the supervision of the second author. Filip De Fruyt is now post-doctoral research fellow of the Flemish Science Foundation. The authors are indebted to Gary D. Gottfredson for his comments on previous versions of this manuscript. This manuscript has greatly benefited from editorial suggestions and commentaries by three anonymous reviewers.

and requests for reprints should be addressed to Filip De Fruyt, Department of Psychology, University of Ghent, H. Dunantlaan 2, B-9000 Belgium; Filip.DeFruyt@rug.ac.be.


This prospective study investigated the validity of the Five-Factor Model (FFM) of personality and Holland's RIASEC vocational interest typology in predicting employment status and the nature of employment in a sample of graduating college seniors as they entered the job market. A sample of 934 senior college graduates enrolled in various academic subjects filled in Costa and McCrae's NEO-PI-R (1992) and Holland's Self-Directed Search (1979). One year after graduation, they were requested to describe their labor market positions and jobs, using the Position Classification Inventory (PCI; Gottfredson & Holland, 1991). Six hundred and twelve people responded to the second call, of whom 335 were employed and 66 unemployed. The incremental validity of the 2 models over and above each other was investigated in the sample of employed and unemployed subjects (N= 401) using stepwise regression analysis. The results showed that Extraversion and Conscientiousness were the only valid predictors of employment status and that vocational interests did not show incremental validity over and above these factors. The RIASEC types, however, were clearly superior in explaining the nature of employment, underscoring the validity of Holland's hexagonal calculus assumptions. Employment reflecting Realistic, Social and Enterprising characteristics was to a limited extent predicted by four of the Big Five, except Neuroticism, over and above the RIASEC types. The findings are discussed in the framework of Schneider's Attraction-Selection-Attrition (ASA) theory (1987) concluding that Holland's RIASEC model is more employee-driven, being better at predicting the nature of employment, whereas the FFM is more employer-oriedted, with greater validity in evaluating the employability and employment status of applicants.