Labels and images have become central in the controversies surrounding affirmative action. To some people, affirmative action means making equal opportunity concrete, while to others it means reverse discrimination. To some people, affirmative action is only a partial compensation for monumental wrongs, while to others it just means replacing competent whites with incompetent blacks. The reality of affirmative action is much more complex than the labels and images, both in concept and in practice.
          To make these intricate and emotionally charged issues manageable, it is necessary (1) to distinguish the basic concepts and legal rationale of affirmative action from the many specific laws, regulations, and practices that have developed under the affirmative action label, (2) to measure in some general terms the magnitude of the problem that affirmative action programs were intended to solve or ameliorate, (3) to consider the actual results achieved and the general trends set in motion by these programs, and finally (4) to weigh the implications of affirmative action policies for those directly affected and for society in general.
          This study draws upon the large general literature on race and sex differentials in employment, pay, and promotion prospects. In addition, it presents some original data specifically focused on academic employment, pay, and promotion. For many occupations, the fact that some of the factors determining individual qualifications for jobs are intangible makes it difficult to determine how much of the observed difference in end results is due to discriminatory treatment and how much to differences in the relevant capabilities. For the academic profession, however, many of the job qualifications that are either conceptually or statistically elusive in other occupations are spelled out-most bluntly in the “publish or perish” rule. For example, the possession or non-possession of a Ph.D. is crucial to an academic career, and the quality of the department at which the Ph.D. was earned is of major importance at the outset of a career and exerts a continuing influence for years thereafter.1 Comprehensive data available from the American Council on Education cover both the degree level of academic individuals and the respective disciplines’ own rankings of the various university departments which issue those degrees, as well as the publication records and academic salaries of individuals by race and sex. In addition, the National Academy of Sciences has made available data collected by the National Science Foundation on holders of doctoral degrees (Ph.D.s, M.D.s, and other doctorates) in various fields by race and sex. In short, the academic profession offers a unique combination of known job requirements and salary data with which to determine to what extent group differences in pay represent group differences in job requirements rather than employer discrimination.

I am grateful to the Liberty Fund for the research grant that made this study possible.

1 Theodore Caplow and Reece J. McGee, The Academic Marketplace (New York: Basic Books, Inc., 1961), p. 225; David G. Brown, The Mobile Professors (Washington, D. C.: American Council on Education, 1967), p. 97.