Teachers must increase effectiveness of teaching to prepare quantitatively literate students. The study’s purpose was to determine the effectiveness of the Mobile Math Initiative (MMI) in three high poverty schools. Changes in achievement of 89 students over 3 years were examined using the Stanford 10 (SAT 10) Procedures and Problem Solving subtests and the Alabama Reading and Mathematics Test (ARMT) Math subtest. Independent variables were (a) ethnic background, (b) gender, (c) socioeconomic status, (d) level of English proficiency, (e) level of special education needed, and (f) school student attended. Data were analyzed with the SPSS (Version 15.0) software using factorial ANOVA with repeated measures.

For the SAT 10 Procedures subtest, the null hypothesis of no significant interaction for school was rejected (F = 7.78, df = 4, p < .001). There was also significant interaction between scores and testing occasions (F = 25.57, df = 2, p < .001). For SAT 10 Problem Solving subtest scores, the null hypothesis of no significant interaction for school was rejected (F = 3.35, df = 4, p = .011). There was also significant interaction between scores and testing occasions (F = 11.18, df = 2, p < .001). For the ARMT Math subtest, the null hypothesis of no significant interaction for school was rejected (F = 3.43, df = 2, p = .037). For all subtests, independent t tests were only significant for special education. Paired t tests comparing mean NCE scores for the SAT 10 Procedures subtest for each category of independent variable at baseline and final testing were significant for all levels of all independent variables except White and special education students. Paired t tests comparing mean NCE scores for the SAT 10 Problem Solving subtest for each category of independent variable at baseline and final testing were significant for all levels of all independent variables except White, paid lunch, and special education students. Paired t tests comparing mean scale scores for the ARMT Math subtest for each category of independent variable between testing occasions were significant only for special education students. The researcher concluded that the MMI was effective in high poverty schools.