Degree Name

Master of Arts (MA)

Semester of Degree Completion

1999

Thesis Director

Harold D. Nordin

Abstract

White people earn more money than non-whites. This is true in most cases even if non-white people perform the same job as white people. Are there reasons for these differences? The purpose of this paper is to discover what those reasons are for the differences in wages and to attempt to discover how much of the difference is really discrimination. Any inequalities that are unexplained may be considered by the author to be discriminatory. Therefore, the hypothesis for this paper is: Income differences between whites and non-whites may be attributed in part to race discrimination.

The statistical method used will be ordinary least square. This will allow the author to run a regression and observe what the correlation or relationship is between the dependent and the independent variables. The regression equation for this model will be Yt = A + B1X1 + B2X2 + B3X3+ B4X4 + B5X5 + B6X6 + εt .

This is saying that the income ratio between non-whites and whites, (Y), is a function of the ratios of percentages of median education levels of non-whites and whites, Ed, (X1); the ratio of non-whites and whites who are married, Marital, (X2); the ratio of the number of non-whites and whites living in an urban area of at least 50,000 or more people, Urban, (X3); the mobility of non-whites to whites, Mob, (X4); the ratio of the unemployment rate of non-whites and whites, Unem, (X5); the ratio of the number of non-whites and whites employed in manufacturing in thousands, Ind, (X6. B1 – 6 are the partial regression coefficients for variables one through six.

The final model (which included the mobility of non-whites, manufacturing, and martial statistics ratios of non-whites and whites) accounted for 55.85 percent of the income differences between the white and black races. The remaining 44.15 percent that was not accounted for may be due to discrimination. However, the remaining proportion could also be caused by the absence of other significant variables that were not used by this model or it could be caused by random factors.

Download

Share

COinS