MARC BENDICK, JR.,
CHARLES W. JACKSON, AND
VICTOR A. REINOSO
To what extent does discrimination operate in the American labor market today? Has the United States achieved a color-blind society, or do personal characteristics still condition the rewards to personal qualifications?
This paper utilizes a new technique for empirical research on these questions, employment "testing" or "auditing." The paper first outlines gaps in empirical information that testing can address. It then describes the testing approach and illustrates its power with results from initial applications. These results demonstrate that hiring discrimination remains far more prevalent than is commonly assumed. The paper concludes with suggested directions for future public and private efforts against bias.
During the past several decades, substantial research has been conducted on employment discrimination, the vast majority of it suggesting that racial and ethnic bias survives to a significant extent.
This conclusion is reinforced by the continued operation of race/ethnic distinctions throughout American society. These patterns include wide-spread segregation in housing and social life, as well as incidents of discrimination experienced by minorities in daily living. Additionally, public opinion surveys indicate that substantial segments of the American population continue to hold stereotyped beliefs and prejudiced attitudes toward minority groups. 1
Studies of the labor market also suggest the continued presence of discrimination. While some racial and ethnic gaps have diminished over recent decades, econometric research continues to find that minorities do less well than equally qualified nonminorities on such employment outcomes as representation in higher-level occupations, wages, returns on investment in educational credentials,