This exploratory study determined the relationship between math instructional activities in Wave (year) 1 and English language learner (ELL) children's math problem-solving performance in Wave (year) 2. The math performance of ELL children (N = 291) in Grades 1, 2, and 3 was tested in two testing waves. Two major findings emerged. First, strategy instruction, peer interaction, and explicit instruction in Wave 1 uniquely predicted English problem-solving scores in Wave 2 independent of grade level, vocabulary, calculation, and problem-solving at Wave 1. Second, the frequency of explicit instruction and peer interactions was significantly related to the odds of predicting ELL children at risk for math disabilities. The results are discussed within the context-specific instructional activities that were positively related to later math outcomes.