A comparison of job analytic and conventional economic variables as explanations of occupational earnings differentials

Abstract

Occupational carnings are analyzed in order to determine the relative importance of micro-oriented job analytic variables vis-a-vis more conventional market variables as explanatory factors of earnings differentials. Earnings functions are constructed in order to determine the extent to which these variables can explain occupational earnings differentials. These functions are analyzed by means of both ordinary least squares and ridge regression techniques. Job analytic variables were obtained from the Position Analysis Questionnaire (PAQ), which measures job characteristics that are similar to many job evaluation systems. When these approaches are combined, a comprehensive model can be estimated that reflects the responsibilities, experience, training, skills, job environment, and work regularity characteristics of different occupations as well as conventional economic factors. Although the job analytic variables measure potentially important differences between jobs, the conventional economic variables have greater explanatory power.