This study addressed whether prior successes with educational interventions grounded in the theory of successful intelligence could be replicated on a larger scale as the primary basis for instruction in language arts, mathematics, and science. A total of 7,702 4th-grade students in the United States, drawn from 223 elementary school classrooms in 113 schools in 35 towns (14 school districts) located in 9 states, participated in the program. Students were assigned, by classroom, to receive units of instruction that were based either upon the theory of successful intelligence (SI; analytical, creative, and practical instruction) or upon teaching as usual (weak control), memory instruction (strong control), or critical-thinking instruction (strong control). The amount of instruction was the same across groups. In the 23 comparisons across 10 content units in 3 academic domains, there were only a small number of instances in which students in the SI instructional groups generally performed statistically better than students in other conditions. There were even fewer instances where the different control conditions outperformed the SI students. Implications for the future of SI theory and the scalability of research efforts in general are discussed.