The Current Population Survey (CPS) is used to supplement census information in between census years. Our data consist of a random sample of 534 persons from the CPS, with information on wages and other characteristics of the workers, including sex, number of years of education, years of work experience, occupational status, region of residence and union membership. We wish to determine (i) whether wages are related to any of these characteristics and (ii) whether there is a gender gap in wages.

We began by inspecting the univariate plots of wages separately against each covariate. Wages appeared to be very skewed, and the variation in wages increased with several variables, such as age and experience (Figure 1). We log-transformed the wage variable, and replotted the log of wages against each covariate; plots were less skew and the variance appeared to be stabilized (e.g. Figure 2).

We then performed a univariate regression of log wages against each covariate (Figures 2 to 8). Wages appeared to increase with work experience, number of years of education and age, and were higher for males, union members and northerners (all F-tests for regression were significant, p < .0001). There was also variation among the different occupational categories (F test for regression, p < .01), with professional and management positions having the highest salaries and clerical and service industry positions having the lowest. Wages did not appear to increase strongly with experience; the percent increase (95% confidence interval) per 10 years of experience was 5% (1%, 8.5%). Female wages were 79% (73%, 87%) that of male wages.

Age and work experience were almost perfectly correlated (r=.98). Professional employment was moderately correlated with years of education (r=.49) (p. 84). There did not appear to be any other strong correlations among the covariates. Multiple regression of log wages against sex, age, years of education, work experience, union membership, southern residence, and occupational status showed that these covariates as a group were related to wages (F test for regression, p < .0001) (p. 85); however, the effect of age was not significant after controlling for experience. Plots of the standardized residuals against the predicted log wages showed no patterns. However, all standardized residual plots showed one large outlier with lower wages than expected, who was not expected to be an influential point in the regression. This was a male, with 22 years of experience and 12 years of education, in a management position, who lived in the north and was not a union member. Removing this man from the analysis did not give substantially different results (p. 90), so that the final model included the entire sample.

Adjusting for all other variables in the model, females wages were lower than males. all other factors being equal, females earned 81% (75%, 88%) that of males (p < .0001). Wages increased 41% (28%, 56%) for every 5 additional years of education (p < .0001). They increased by 11% (7%, 14%) for every additional 10 years of experience (p < .0001). Union members were paid 23% (12%, 36%) more than non-union members (p < .0001). Northerns were paid 11% (2%, 20%) more than southerns (p =.016). Management and professional positions were paid most, and service and clerical positions were paid least (pooled F-test, p < .0001). Overall variance explained was R2 = .35.

In summary, many factors describe the variations in wages: occupational status, years of experience, years of education, sex, union membership and region of residence. However, despite adjustment for all factors that were available, there still appeared to be a gender gap in wages. There is no readily available explanation for this gender gap.

|Data Sources| Course Topics|Examples|