Abstract

Are women less often invited to job interviews? The theory of statistical discrimination states that representatives of social groups, which are judged as less productive by the employer, are systematically disadvantaged irrespective of the actual productivity of the individual. Women (at least in a certain age range) are generally assumed to interrupt their work due to family times. According to the theory, they should therefore be rarely hired or even invited to interviews. But is there any evidence for the statistical discrimination? And are these assumptions also true in typical female occupations, where employers know more or less exclusively the productivity of women as a benchmark? Or is there rather a discrimination against men? And what is it the case in typical men's and hybrid occupations?
Whether employers prefer women or men in filling vacancies is hardly observable in direct surveys, due to a strong social desirable response behavior. Therefore we analyze this issue by a quasi-experimental design in form of a vignette survey (factorial survey). In the online survey over 4200 personnel managers from a representative sample of German companies participated. They had to evaluate short fictional applications in which the individual characteristics of applicants were varied experimentally. In addition to the characteristics of the applicants further company attributes were collected in order to isolate the effect of gender discrimination in the analyses.