This paper evaluates an online charter school that serves children in grades K-8 in a southern state in the United States. We compare growth in math and literacy learning on state standardized assessments between students enrolled in this school and "matched twin" students enrolled in traditional public school students statewide each year from 2010 to 2012. We also examine the impact of the school subgroups that include minorities, students with special needs, and students in the bottom and top quartiles within their schools. We then use propensity score matching to match students who enrolled in the online school at least three consecutive years with an individually matched comparison group of public school students and estimate the differential value-added in math and literacy between the two groups. We find evidence of negative transition effects in the first year for students who enroll in the online school. These effects, however, dissipate after the first year and, in some instances, turn positive.