Geographic Variation in the Gender Differences in Test Scores

I n the 1960s, the gender gap in college enrollments was 1.55 males for every female. By the 1980s, this gap had been erased. By 2003, the college gender gap was 1.30 females for every male undergraduate (Goldin, Katz, and Kuziemko, 2006). All cohorts of U.S. women born since 1960 have had higher average years of schooling than their male counterparts (Charles and Luoh, 2003). This educational convergence has even taken place in the historically male-dominated areas of science and engineering, with women earning 42.7 percent of bachelor’s degrees in these fi elds in 1990 and 50.6 percent by 2001 (National Science Foundation, 1998, 2001). Despite this convergence at the undergraduate level, women are still greatly underrepresented in the upper echelons of many fi elds, particularly in the ranks of faculty in science, math, and engineering at prestigious universities. In the science faculty at MIT, for example, 8 percent of the faculty were women in 1990, rising slightly to 12 percent by 1999 (Committee on Women Faculty at MIT, 1999). As of 2000, the proportion of tenured U.S. faculty who are women was under 5 percent for engineering and around 10 percent for economics and physical sciences ( Ginther and Kahn, 2004). This pattern of rough gender equality in averages, but differences at the extremes, is also found in standardized test scores of pre-collegiate students. Geographic Variation in the Gender Differences in Test Scores