Many U.S. students enter elementary school with poor math skills. Students from different socioeconomic backgrounds also differ in math performance, with students from poor families lagging behind those from more affluent families. These differences grow over time, resulting in substantial differences in math achievement by the time students reach fourth grade. This report presents the evaluation design for Mathematica’s large-scale, national study that is comparing the effects of four early elementary math curricula on improving student math achievement in disadvantaged schools. Experimental methods are being used to evaluate the relative effects of the curricula, using a school-level random assignment design.