Sex Segregation Within Occupations
研究检验了人力资本模型和统计歧视理论对职业内性别构成的解释力，基于美国加州400多家企业的数据发现统计歧视更符合实际，但质疑其仅为最优决策的假设。
Do the paychecks and career prospects of men and women differ because of rational, optimizing decisions made by men and women, given their respective skills, experiences, home responsibilities, and intentions regarding future work and family activities? The human capital model-coherent, logical and internally consistent-has a certain elegance to it. Nevertheless, empirically, it cannot explain the sex composition of jobs in lines of work pursued by both men and women. The results reported below are more consistent with the notion that employers statistically discriminate; that is, they reserve some jobs for men and others for women based on their perceptions of the average traits of the two groups. However, our results also call into question the assumption that statistical discrimination is merely an optimal decisionmaking strategy under conditions of uncertainty. We summarize below our recent research on determinants of the sex composition of jobs in mixed occupations. That research is based on data on staffing patterns, job requirements, and personnel policies collected by the U.S. Employment Service in over 400 California establishments between 1959 and 1979. Details on sample representativeness, procedures of data collection, and the like can be found in our 1984 and 1986 papers.