Wage Labor Regression Analysis
... Labor Economics theory gives us a foundational insight as to how wage discrimination among the two genders and races appear. Discrimination is said to exist when female and/or minority workers, other things equal, are accorded inferior treatment with respect to hiring, promotion, wage rate, or working conditions. Narrowing our scope to only dealing with wage rates, two popular models that help explain firm behavior are the “taste” and “statistical” discrimination model. ... Looking at it a little more in depth, an employer’s cost or physic to hiring a non white-male worker added to the wage offered to that particular worker must be less than the prevailing white-male wage in order for the employer to hire the non white-male employee. My paper will however be dealing with the latter model of wage differential estimation. ... The first such technique is that of a regression model testing increases in the log of wages using simple dependent dummy variables distinguishing race and gender, keeping the white-male as the control group. ... Regression results, interpretations, and possible critiques are reported in section IV. ... The second such article, dealing with that of race and gender wage differentials, is authored by Professor George Borjas of the University of California, Santa Barbara. In his 1983 article, Borjas looks at wage differentials with respect to agencies within the federal government, and also questions the traditional usage of simple dummy variables in determining unbiased differential wage coefficients. ... From this random sample of observations I proceeded to truncate the data set by excluding the self employed individuals, as well as those who are not in the labor market (i. ... The second excluded group is that of those individuals not in the labor force; for the obvious fact that the zero wages earned would lead to a downward bias of the estimated coefficients. ... The methodology used for the following analysis started out with the simple log wage simply dummy equation given as: (1) Log (wage) = a + b(Female) + d(Black) + l(Black * Female) + n Where the dummy variables are categorizing the data set into white female, black male, and black female groups with the base being white male. Given this equation all the estimated coefficients (not including the constant) should be negative, to coincide with wage discrimination theory. ... The second regression used compliments the previous by including socioeconomic factors such as education, marital status, and potential experience. ... The equation is defined as: (2) Log (wage) = a + e(Race/Gender) + g(Education) + q(Marriage) + s(Experience) + m Where e is the gender/race trait taken from the combined coefficient stated in the previous regression; education is a set of dummy variables separating the highest level of attainment and marriage is also a dummy variable. ... More importantly, the regression will be ran across three separate years (95, 97, 99), which will yield separate estimates for each year. ... Significance tests will be ran for each variable as well as each regression, to discourage any implications a possible low R2 could bring about.