Academic journal article Public Administration Review

Diminishing Returns from Statistical Analysis: Detecting Discrimination in Public Employment

Academic journal article Public Administration Review

Diminishing Returns from Statistical Analysis: Detecting Discrimination in Public Employment

Article excerpt

In their article, "Asian Americans in the public Service: Success, Diversity, and Discrimination," Kim and Lewis (1994) note that Asian Americans experience discrimination within the federal service, suggest that the problem is pervasive and serious, and propose remedies. This writer has personally witnessed at least two incidents of employment discrimination against Asian Americans. Neither of these episodes involved federal personnel, but it seems reasonable that the federal service is not totally immune to prejudices evident elsewhere in society. However, Kim and Lewis's statistics shed little light on such discrimination's pervasiveness. Anti-Asian discrimination may not be as important a federal phenomena as their article concludes. Kim and Lewis's lucidly presented statistics can be interpreted in varying ways. The alternative interpretation which follows may also have implications for identification of discrimination against other groups, such as women, Hispanics, and African Americans.

Using 1990 and 1992 statistics, Kim and Lewis describe differences between Asian Americans and nonminority whites. Asian Americans make up 2.6 percent of the civilian work force but constitute less than 1 percent of municipal officials. There are almost no Asian American mayors or city/county managers. Only .9 percent of the Senior Executive Service is Asian, while 3.5 percent of the federal work force and 4.3 percent of the postal service are Asian (286-287). Twenty-seven percent of white men serving the federal government are supervisors, but only 15 percent of Asian American men hold such positions. Twelve percent of white women in the federal service hold supervisory positions, compared to only 7 percent of Asian women (288-289).

Kim and Lewis also discovered grade discrepancies between Asian Americans and whites. "Asian men [in federal service] tended to be .03 of a grade lower than white men with the same amount of education, federal experience, and age who had the same handicap and veteran status" (288). The disparity between Asian American and white females employed by the service is even greater; Kim and Lewis found Asian women to be .06 of a grade behind white women with "the same, education, seniority, age, veteran status and handicap status" (289).

Kim and Lewis assert, "Because other factors also affect career success, the persistence of grade gaps after controlling for age, education, federal experience, veterans preference and disability status does not prove discrimination, but it does indicate problems that the government needs to investigate" (287). Elsewhere they attribute disparities to discrimination more bluntly. Asian American women are said to "face double discrimination" but be "held back more by their gender than their race" (287). Kim and Lewis conclude that "even this model minority faces discrimination. Policy makers should not ignore this evidence and assume that the battle against discrimination has been won for Asian Americans" (289-290).

Readers seem to be presented with a stark choice: either envelop oneself in blanket denial of the existence of any anti-Asian discrimination, thus ignoring the "evidence," designate anti-Asian discrimination as a serious problem requiring responses such as diversity training, special recruitment and placement, and Asian American political mobilization. This writer, however, remains uncertain about the phenomenon's extent; Anti-Asian discrimination may be pervasive within the public service, or it may consist largely of isolated incidents addressable through existing laws and grievance procedures. The statistical disparities cited above shed little light on this issue because plausible alternatives to the discrimination hypothesis can be formulated.

One alternative could be called immigrants' loss or the educational devaluation hypothesis. Education's contribution to productivity is often contingent upon linguistic and cultural contexts, so immigration can devalue human capital. …

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


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.