The effects of instruction in solving mathematical word problems for students with learning problems: A meta-analysis

This article provides a synthesis of word-problem-solving intervention rese arch with samples of students with learning problems (i.e., mild disabiliti es and at risk for mathematics failure). The effectiveness of word-problem- solving instruction in 25 outcome studies was examined across student chara cteristics (e.g., grade, IQ); instructional features (e.g., intervention ap proach, treatment length); methodological features; skill maintenance; and generalization components. Separate analyses were performed for group-desig n studies and single-subject studies using standardized mean change and per centage of nonoverlapping data (PND), respectively. The overall mean weight ed effect size (d) and PND for word-problem-solving instruction were positi ve across the group-design studies (ES = +.89) and single-subject studies ( PND = 89%). In addition, positive effects for skill maintenance and general ization were found for group design (ES = +.78 and +.84, respectively) and single-subject studies (PND = 100%). Computer-assisted instruction was foun d to be most effective for group-design studies. Effects for representation techniques and strategy training were found to be significantly higher tha n the "other" approach for both group-design and single-subject studies. Lo ng-term (> 1 month) intervention effects were significantly higher than sho rt- or intermediate-term interventions for group-design studies, whereas bo th long-term and intermediate treatments were seen to be more effective tha n short-term treatments for single-subject studies. Other significant effec ts found for group-design studies only in terms of student characteristics, instructional features, and methodological features are reported. Finally, implications of the current analysis for future research in the area of wo rd-problem solving are discussed.