Historically, secondary students with emotional behavioral disorders (EBD) have made poor progress in mathematics putting them at risk for school failure and placement in an alternative setting. Two under studied areas essential to success in mathematics are fractions and algebra. The purpose of this study was to test the effects of a multi-component intervention on the math performance for middle school students with EBD in an alternative educational setting. This study used a one-group nonequivalent dependent variables design (Shadish, Cook, & Campbell, 2002) with multiple measures in multiple waves to assess the effects. Repeated measures ANOVA indicated that students significantly improved their math performance on both fractions and algebra using researcher developed measures. Social validity results indicated that teacher and students found the intervention to be an acceptable intervention. Implications from fidelity and social validity data are discussed in addition to intervention components for this population.