Race, Class and Conservatism by Thomas D Boston

Author:Thomas D Boston [Boston, Thomas D]
Language: eng
Format: epub
Tags: Social Science, Sociology, General
ISBN: 9781136030727
Google: xeYDAQAAQBAJ
Publisher: Routledge
Published: 2013-09-27T05:50:48+00:00

Black

−0.51365

−3.42162

Age

−0.13673

−0.05765

Agesq

0.00181

0.06396

Ushrswh

0.00405

0.68288

Highed

0.15526

7.96389

Lpreempl

0.00341

0.49543

Hlngwk

0.00223

0.33541

Ncentrl

0.21521

1.60835

South

0.06562

0.59461

West

0.17596

1.25052

Intercept

5.54496

0.11148

Note: The logit procedure used adds 5 to the intercept and divides the logit by 2.

The region variables are referenced to the North, meaning that the coefficients of the three other regions record the probability of being in the primary sector if one lives in a region other than the North. For example, the log of the odds of being in the primary sector is greater in the North Central (0,21521), South (0.0656) and West (0.17596) than in the North. However, none of these regional coefficients are significant at the 0.05 level.

The variables that are most significant and have the greatest impact upon the probability of being in the primary sector are race (Black) and years of education (Highed). Education has the highest coefficient to standard error ratio (7.964) and positively influences the probability of being in the primary sector; its coefficient is 0.15526. Race is the next most significant variable and has a strong negative influence on the probability of being in the primary sector. Its coefficient value is −0.5136. All other variables are insignificant at the 0.05 level. The size, sign and significance of the coefficient for race strongly support our hypothesis on discrimination. Specifically, all other factors constant, blacks have a much greater probability of being in low-status, low-paying occupations.

The best way to interpret the coefficient results is to examine the probability generated by a particular set of values of our explanatory variables. Suppose we have two workers differing only by race: one black and the other white. Apart from this all other factors are the same. Specifically, they both are 42 years of age, work 40 hours per week, have 13 years of education, 5 years’ tenure with present employer, 4 years’ job experience in present occupation and currently reside in the North Central. Under these circumstances, the probability of a white holding a primary-sector job is 81.2 percent while that of a black is 60.7 percent. There is clearly an occupational accessibility difference attributable to race – something we will explore in greater detail in Chapter 4.

The key point is that the endogenous relationship between occupational distribution and race causes estimates of discrimination, as measured by the race variable, to be inefficient. Under these circumstances, when measured discrimination in Table 3.12 decreased from −0.0498 in equation 2 to −0.0126 in equation 3, this is to be expected. That is, by introducing occupations into the equation we are controlling for the most prevalent form of discrimination and thereby eliminating it a priori from the results.

In summary, as we add more and more control variables, it is true that the size of the discrimination coefficient is reduced, but certainly not enough to justify a contention that discrimination is not a significant factor in contemporary labor markets. Our results show (see equation 2 of Table 3.12) that, if one is black, the percent change in hourly wage will be below that of whites. Likewise, Table 3.13 reveals that equally qualified blacks have a 20 percent lower probability than whites of working in the primary sector.

Download

Copyright Disclaimer:
This site does not store any files on its server. We only index and link to content provided by other sites. Please contact the content providers to delete copyright contents if any and email us, we'll remove relevant links or contents immediately.