The effect of behavior within a word-problem intervention for students with mathematical difficulties

Students with or at-risk for challenging behaviors (e.g., students with emotional and behavioral disorders [EBD]) represent a group of children who often experience difficulty with both behavior and academics. Researchers and educators have designed interventions to improve the conduct for students with challenging behaviors, however, few mathematics interventions are specifically designed to support students with challenging behaviors. One mathematics intervention with a strong evidence base for use with students with learning disabilities is using schemas to solve word problems. Notably, researchers also have identified schema-based instruction as an effective intervention to solve word-problems for students with EBD. While a promising intervention, little is known about the effectiveness of a word-problem intervention relative to student behavior. That is, do behavioral profiles collected before the start of intervention predict response to intervention? This study sought to compare the behavioral profiles of students with and without mathematics difficulty (MD) and investigate the efficacy of a randomized control trial of a word-problem intervention for students with MD. Results of the present study indicated that students with MD demonstrate higher externalizing and internalizing behaviors than students without MD. Additionally, students in the word-problem intervention groups outperformed their peers in a no-treatment comparison group on all outcome measures, but most notably, on double-digit word problems with an effect size of 1.01. While students with high externalizing scores in the treatment condition performed significantly lower on a double-digit word-problem outcome measure than students without high externalizing scores, their externalizing score was not necessarily predictive of their word-problem gains. When comparing students with and without high internalizing scores in the treatment group, there was no statistical difference. Findings from this study establish a need to continue to study the link between mathematics difficulties and behavioral challenges.