This paper describes the background, methodology, preliminary findings, and anticipated future directions of a large-scale multi-year randomized field experiment addressing the efficacy of ST Math [Spatial-Temporal Math], a fully-developed math curriculum that uses interactive animated software. ST Math's unique approach minimizes the use of mathematics symbols, terminology, and language to teach standards-based mathematical concepts by utilizing students' innate spatial ability. In 2007, fifty ethnically diverse low performing schools in Orange County, California were randomly assigned to receive ST Math in either 2nd and 3rd or 4th and 5th grades, with 36 of these schools implementing the program during the 2007-2008 school year. Initial findings using aggregated data from the math subtest of the California Standards Test (CST) show an improvement in CST scores among the treatment grades (p = 0.05) after one year of ST Math instruction. This association had an effect size of 0.37. Although treatment was not found to interact with student characteristics, further analysis of the treatment effect for specific groups, such as English Language Learners, is ongoing. Future directions include research on individual student-level effects and motivation as well as analysis of the effects of specific components of the program. (Contains 3 tables and 5 figures.)