The purpose of this study is to discern the effects of instruction type on minority students in urban schools, mathematics achievement. Two hundred thirteen third and fifth grade (136 African American and 77 Latino) students attending schools in low-income urban communities were provided mathematics instruction in one of two ways: multiple intelligence instruction (MI) or traditional instruction. Quasi-experimental results (Creswell, 2005) reveal that students exposed to multiple intelligence instruction score significantly higher on the mathematics posttest than students in the traditional instruction context. MI students also demonstrate significantly higher improvement from pre to post test than traditional instruction students do. Implications of these findings and future research directions are discussed. (Contains 3 tables and 1 figure.)