The purpose of this study is to investigate the impact of academic language on the standardized test scores of special needs students and their peers in elementary (n = 1140) and middle (n = 451) public schools across the state of Georgia. Several univariate (ANOVA) and multivariate (MANOVA) analyses of variance are conducted on student classification (English language learners “ELL”, non-ELL, students with disabilities “SWD,” non-SWD, economically disadvantaged students “EDA”, and non-EDA) and student test scores in ELA, math, science, and social studies. Univariate (ANCOVA) and multivariate (MANCOVA) analyses of covariance are also conducted where ELA is treated as a predictor of students’ test scores in math, science, and social studies. MANOVA results reveal that the combined-subjects modeling of student test scores is significantly different by student classification with relatively large effect sizes (0.44 to 0.63) for all grade levels. Follow-up ANOVAs indicate that individual modeling of core subjects is significantly different by student classification, with effect sizes between 0.37 and 0.61. The results of ANCOVA and MANCOVA suggest a statistically significant effect of ELA on student test scores results. SWD and ELL groups benefit the most when controlling ELA test scores.