Academic journal article Social Work

Gender-Based Salary Inequity in Social Work: Mediators of Gender's Effect on Salary

Academic journal article Social Work

Gender-Based Salary Inequity in Social Work: Mediators of Gender's Effect on Salary

Article excerpt

Building on earlier research on racial discrimination in salary, Yamatani and Koeske (1984) proposed two approaches to the detection of discrimination. In the approach described as the classical validation model, merit factors mediate the effect of a status or attribute variable, such as race or gender, on salary. The 1984 article addressed race as the attribute; the current application was gender. Merit or human capital factors include contributions or investments, such as experience, job responsibility, and training. Research has shown an enduring difference in men's and women's salaries, with women earning less than 90 cents per dollar paid to men (Fortune & Hanks, 1988; Marini & Fan, 1997). In a presumed just world, if all the merit mediators were entered into a path model, or otherwise statistically controlled, gender would no longer directly affect salary. In the real world we know that a myriad of non-merit factors affect salaries, even within gender groups. We might, nonetheless, expect that in the absence of gender discrimination, controlling for highly relevant merit factors would result in a reduction or elimination of any direct effect of gender.

This classical validation model involving control of merit factors is an analytic offshoot of the principle of distributive justice or equity, which has been examined from a political-philosophic perspective (Rawls, 1971, 2001) and an interpersonal-psychological perspective (Adams, 1965). The essential proposition of equity theory is that the ratio of an individual's contribution (or investment) to outcomes (or rewards) should be equal. Equity has been presented as the "fairness" principle, although Rawls (2001) suggested refinements to protect the least advantaged, and tests of Adams's theory have led to an awareness that egocentric and other motives influence the calculation of justice determination in practice (see, for example, Leventhal, 1976). In the context of salary equity and gender, the fairness notion requires that men and women performing the same job with equal merit receive equal salary. Equity can be claimed only when there is no significant salary differential after merit factors, such as education, experience, and job activity, have been controlled.

When statistical procedures have been applied to assessing gender-based salary inequity in large, essentially representative, and heterogeneous samples in the United States, women workers have been shown to receive about 70 percent of the salary of men workers (Barusch, 2002; Wittig & Lowe, 1989). Human capital (or merit) factors seem to account for only about one-half of this disparity, leaving a 15 percent inequity, an amount somewhat higher than found in controlled studies of social worker salaries. It is unknown how much of the lesser disparity in social worker salaries is due to the greater control of irrelevancies achieved by studying a single occupation and how much is attributable to the centrality of social justice as a professional value in social work. Human capital proponents (O'Neill, 1985) have claimed that the studies of heterogeneous samples have not controlled for more subtle merit factors, but liberal social justice advocates might argue that the 15 percent inequity actually underestimates true inequity by missing more subtle, unaccounted for, discriminatory issues.

Fortune and Hanks (1988), however, found that female social workers who had been at their jobs for more than three years made nearly $1,000 less than their male counterparts, even after controlling for job role and other factors. Other researchers found similar results (McNeely, 1989; Sutton, 1982; Yamatani, 1982). Rubin (1981) found statistically significant, but very small and arguably inconsequential, salary differences after controlling for experience, rank, and degree. This study, however, involved only social work academics who had at least an assistant professor rank and were employed in schools with graduate programs. …

Search by... Author
Show... All Results Primary Sources Peer-reviewed

Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.