The purpose of this quantitative causal-comparative research was to assess the effect of classroom structure (single-gender vs. coeducational) on students' performance in mathematics in grades 3 through 5 at three large, urban elementary schools in Las Vegas, Nevada, while controlling for student gender and ethnicity. This study was theoretically founded on four educational theories--Piaget's constructivist learning theory (1952) and theory of cognitive development (2003), Kolb's (1976) learning styles theory, and Bandura's (1977) theory of self-efficacy. The researcher analyzed students' scores on the Nevada Criterion Reference Test in mathematics from the 2013-2014 school-year. The sample consisted of 743 students in grades 3 through 5. For Research Question 1, the results showed that third grade students in coeducational classrooms had a statistically significant higher mean score t(100.239) = -2.191, p = 0.031, while fourth and fifth grade students in single-gender classrooms had statistically significant higher mean score t(149.452) = 1.993, p = 0.048; t(178.000) = 3.969, p < or equal to 0.001. These data suggest that single-gender classrooms are more effective for upper elementary grades. When controlling for gender and ethnicity, the ANOVA results indicated that there were no statistically significant interactive effects of classroom structure and gender, classroom structure and ethnicity, or classroom structure, gender, and ethnicity for Research Questions 2, 3, and 4, respectively. This knowledge allows school leaders to make educated decisions on whether single-gender classrooms will be an effective strategy for closing the achievement gap in their elementary school. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]