European Journal of Personality Eur. J. Pers. 17: S67–S76 (2003) Published online in Wiley InterScience (www.interscience.wiley.com). DOI: 10.1002/per.482 Evidence of Cross-Cultural Invariance of the Big Five Personality Dimensions in Work Settings JESU ´ S F. SALGADO 1 *, SILVIA MOSCOSO 1 and MARIO LADO 2 1 University of Santiago de Compostela, Spain 2 Metis, personas, tecnologı ´a y sistemas, Spain Abstract This article explores the cross-cultural invariance (construct validity) of two work-related personality inventories based upon the Five Factor Model (the HPI and the IP/5F). The results show a good convergent and discriminant validity between scales that measure the Big Five personality dimensions. A factor analysis indicates that all personality scales load on the hypothesized Big Five dimensions. Some implications of these findings for the research and practice of personality measurement in personnel selection are discussed. Copyright # 2003 John Wiley & Sons, Ltd. INTRODUCTION For many years, personality measures were considered poor predictors of organizational criteria (see e.g. Ghiselli & Barthol, 1953; Guion & Gottier, 1965; Schmitt, Gooding, Noe, & Kirsch, 1984) as the reported validity coefficients were in general small to negligible. A reason for these results was that many different personality characteristics were collapsed into only one data set and the average validity was then calculated (see e.g. Ghiselli & Barthol, 1953; Schmitt et al., 1984). Moreover, the personality measures included in the validity studies had been developed using very different theoretical frameworks and were conceptualized for measuring different personality characteristics (e.g. emotional stability, extroversion, achievement, empathy, and many others). Therefore, collapsing all these different measures into only one data set produces the effect of mixing apples and oranges, as the relationship between a different personality characteristic and the same work criterion may be different or highly specific. In order to be collapsed in the same data set, the personality measures have to show good convergence. However, many years of research have demonstrated the independence of a variety of personality dimensions. For example, it is now accepted that emotional stability and extroversion are different personality dimensions, and the same is true for other personality dimensions. Consequently, if we collapse extroversion coefficients and emotional stability coefficients Received 30 April 2002 Copyright # 2003 John Wiley & Sons, Ltd. Accepted 10 July 2002 *Correspondence to: Jesu ´s F. Salgado, Departamento Psicologı ´a Social y Ba ´sica, University of Santiago de Compostela, 15706 Santiago de Compostela, Spain. E-mail: psjesal@usc.es