While Asians are overrepresented in science and engineering (S&E), they receive limited scholarly attention in sociology of science. To fill the knowledge gap about this understudied group, this study examines the effects of race, nativity, degree origin, gender, field, employment sector, and national origin on the annualized earnings of Asian computer scientists and engineers working in the U.S.
To understand the above effects, this study uses descriptive analyses and quantile regressions. Data are derived from the National Survey of College Graduates (NSCG) conducted by the National Science Foundation. To track the changes of the above effects over time, this study uses 1993 and 2003 NSCG data.
Using quantile regression, this study has the following major findings. First, race and nativity had some statistically significant effects on the earnings of Asian computer scientists and engineers in 1993 at both 90th and 50th quantiles. The race effect disappeared in 2003, and the nativity effect disappeared with an exception at the 50th quantile. Degree origin had a statistically significant effect in 1993 in some cases at the 90th quantile but across gender, field, and two employment sectors at the 50th quantile. While this effect existed in 1993, it disappeared in 2003 except among engineers and in educational institutions at the 50th quantile.
Second, in terms of gender differences in earnings, all the four women’s groups, namely, white, Asian American, U.S.-educated immigrant, and Asian-educated immigrant women, earned less than their male counterparts in either 1993 or 2003 at either the 50th or the 90th quantile. In addition, U.S.-educated immigrant women suffered from the double bind effect, or being disadvantaged due to both their gender and race, at the 50th quantile.
Third, computer scientists earned slightly more than their engineer counterparts in both years at both quantiles. Fourth, in terms of employment sector differences, educational institutions and state/local government paid less than industry in both 1993 and 2003 at both quantiles. Federal government, which paid comparable workers less than industry in 1993, eliminated the gap in 2003 at the 50th quantile but not at the 90th quantile.
Finally, when disaggregating U.S.- and Asian-educated immigrants by national origin, this study finds that a few but not all nationality groups, including the Chinese, Filipinos, the Vietnamese, Koreans, and the Taiwanese, suffered from earning disadvantages in 1993 or 2003 at either quantile. In addition, the findings at different quantiles suggest that the earnings of workers in the upper tail (90th quantile) are less influenced by their personal or employment characteristics that have been examined in this study than those at the median (50th quantile). In other words, workers of different backgrounds, such as race, nativity, degree origin, gender, etc., in the upper tail are closer to each other in earnings than those at the median.
Overall, the findings partly confirm the structural arguments that some groups, notably women, racial/ethnic minorities, and immigrants, are disadvantaged in the U.S. workplace. The degree origin effect in 1993 could be due to the lower quality of degrees obtained from Asian higher education institutions and to the marginalized structural positions of Asian-educated immigrants in the American society. The disappearance of such an effect in 2003 could be due to the interactions between structural forces and human capital. The change of the effect of human capital has to be placed in a context of globalization and the resulting structural changes in various aspects, such as the improvement in higher education in Asia and changes in immigration policies in the U.S.