urban institute nonprofit social and economic policy research

Gender Gaps in Math and Reading Gains During Elementary and High School by Race and Ethnicity

Read complete document: PDF

PrintPrint this page
Share on Facebook Share on Twitter Share on LinkedIn Share on Digg Share on Reddit
| Email this pageE-mail
Document date: September 30, 2006
Released online: March 02, 2007

The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders.


Gender differences in academic achievement have long fascinated researchers and policy-makers alike. In this paper we analyze differences in math and reading test score growth rates by gender for four different race and ethnic groups—white, black, Hispanic, and Asian students—for six different time periods. Our data cover both the earliest years of education and the crucial years of adolescence. In addition, we have data bracketing one non-schooling period. Together these data enable us to get a very complete picture of how gender gaps evolve over the course of early elementary and high school years and how these trajectories differ by race and ethnicity. While the gender gaps are not always statistically significant, they are for 15 of 48 comparisons made, all during school. In addition, all of the statistically significant results suggest that males learn more math and females more reading during early elementary school and again during high school.

The text below is an excerpt from the complete document. Read the full paper in PDF format.


Who's smarter, girls or boys? Each year new studies appear examining if gender gaps in achievement exist, and if so, who is favored and by how much. These questions cannot be answered by achievement tests, or perhaps any other means, but policymakers and researchers continue to focus on gender gaps in achievement. While such gaps cannot be used to clearly identify differences in innate ability or the impacts of schools or society, they can help us determine where to direct resources and research. Of primary interest is not an achievement gap but a gap in learning, or the change in achievement over time, since any gap at a point in time can be quickly swamped by a small difference in learning rates. Previous research indicates that girls tend to outperform boys on reading and comprehension tests, and boys outperform girls on math and science tests, both in terms of achievement at a point in time, and to a lesser degree the rate of learning.

While a great deal of research stresses the importance of gender gaps in achievement (Dee 2006; Murray 2005), others argue that the achievement gap between girls and boys has been greatly exaggerated. Friedman (1989) found small differences by gender in average SAT scores and other tests based on results of meta-analyses. More recently, Mead (2006) notes that gaps by ethnicity and social class are far larger than the gender gaps. In addition, American boys are performing about as well as they ever have. Girls are improving faster in math and science, but this only means that they are closing the gaps in those areas. Mead worked with data from the National Assessment of Educational Progress (NAEP), which includes results of reading and math assessments administered periodically to nationally representative cohorts of students around the ages of 9, 13, and 17. These data are useful for looking at differences between states and over time, but tell us little about how gaps develop before the age of 9.

Gaps at a point in time can be quickly swamped by small differences in learning rates. Consequently we focus our research on rates of gain in achievement (i.e. learning rates). Some previous research, such as the work by Mead on NAEP, is based on comparisons of the performance of one cohort of students in one year (e.g., the average math scores of 9-year-olds in 2004) with the performance of another cohort in another year (e.g., the average math scores of 9-year-olds in 1973). While interesting, these comparisons cannot distinguish between differences in learning rates and differences in the composition of the cohorts being studied. In order to focus on learning rates we follow individual children over time.

Our data include two periods in children's lives. The first period starts in the fall of kindergarten and continues through the end of third grade (when children are typically 9 years old). The second period goes from eighth grade to twelfth grade, the critical adolescent and teenage years. In earlier work we looked at differences in test score growth rates by gender and by race separately (LoGerfo, Nichols, and Reardon 2006). In this report we look at the interactions between these two characteristics.

This report is divided into three sections. First, we discuss the data and analytic methods that produce the gender gap estimates. Second, we present and explain our findings. Finally, we conclude with a brief summary of our findings.

The complete paper is available in PDF format.

Topics/Tags: | Education | Race/Ethnicity/Gender

Usage and reprints: Most publications may be downloaded free of charge from the web site and may be used and copies made for research, academic, policy or other non-commercial purposes. Proper attribution is required. Posting UI research papers on other websites is permitted subject to prior approval from the Urban Institute—contact publicaffairs@urban.org.

If you are unable to access or print the PDF document please contact us or call the Publications Office at (202) 261-5687.

Disclaimer: The nonpartisan Urban Institute publishes studies, reports, and books on timely topics worthy of public consideration. The views expressed are those of the authors and should not be attributed to the Urban Institute, its trustees, or its funders. Copyright of the written materials contained within the Urban Institute website is owned or controlled by the Urban Institute.

Email this Page