professionals, who progress steadily from the entry level grades to at least grades 13 and 14 and who have enjoyed such predictable progression for a number of years. White males, then, with long government service tend to be in the higher grades. For black male professionals, however, the average length of government service at each grade level was found to be relatively constant across the grades. Thus, length of government service is a much less significant indicator of salary for black males than for white males. Black males have not benefited from predictable promotion patterns as have white males. And this is what is shown by the large differences among the coefficients for government service of the various salary models. beta coefficients. 33/ Such differences are also revealed by a comparison of For each variable of a regression model, its beta coefficient is the product of its regression coefficient and its variance, divided by the variance of the dependent variable. The beta coefficient represents 33/ Note that the use of beta coefficients is not a different way of comparing regression coefficients as much as it is a more meaningful way. It is certainly not independent from other comparisons. a standardization of the coefficient-variance products by the variance of the dependent variable, permitting a valid comparison of these products. One would expect independent variables with large variances and large regression coefficients to have large effects on the value of the dependent variable. Thus, we can use the beta coefficients an indication of the relative importance of individual independent variables in determining salary. The table on For page 94 ranks the beta coefficients by magnitude by regression model for all of the models. Comparing the ranks for the government service variable, we again see the important differences existing among salary models. example, the service beta coefficient for white males ranked first among beta coefficients in the white male model, while the service beta coefficient for black males ranked sixth in the black male model. Thus, the great difference between black and white males in the relative importance of government service as a salary determiner is illustrated again. g. Racial and sexual implications of the results. Referring to the table (page 90 ) listing the sta tistically significant explanatory factors for each salary Ranking of Explanatory Factors by Regression Model Using the Magnitude of Beta Coefficients t + All variables having a beta coefficient of at least .1 are ranked. * Using black, nonblack minority, and female as the race/sex dummies. Using white female, black male, black female, nonblack minority male, and nonblack minority female as the race/sex dummies. model, we see that, except for some of the nonblack min ority variables, all of the race and sex variables were statistically significant in all of the models in which they were used. 34/ Furthermore, they were all signifi 35/ cant at a 1 percent alpha-level. This is a very strong and important result. It means that when we account for the variation in salary which can be explained by variations in years of government service, age, education, geographic location, and organizational unit, a significant portion of variation remains unexplained which can be accounted for by the race and sex variables. That is, race and sex are important determinants of an employee's salary in the Labor Department; they account for salary differences that are not explained by differences in age, 34/ No race and sex variables were used in the models with the finest possible race and sex breaks in themselves. They were used wherever they were necessary to give a finer break. To see in which models they appeared, refer to the table on page 88. 35/ In this case, statistical significance is affirmed if a hypothesis stating that a given variable is not significant can be rejected. The alpha-level is the probability of rejecting a hypothesis that was true. experience, and education alone. This conclusion is even more important because it speaks to the basic research question which the correlation analysis was seeking to answer. That question was ; having demonstrated through straight quantitative techniques that age, education, and experience considered individually do not account for the major portion of the salary gap between minority and female employees and all employees, do they, when considered simultaneously, explain the major portion of these salary differences? has been shown, the question can now be answered in the Minority and female employees in the Department negative. As do earn less than all employees, and a major portion of this salary gap exists just because of their race or sex. The magnitude of the salary differences explained by race and sex variables in the regression models is shown by the table on page 97. Only statistically sig nificant differences are shown. For example, considering the salary model for all employees using the race-sex dummies giving sex within race, we see that after computing a salary prediction for an employee by plugging in specific values for all of the variables not relating to race and |