Association for Public Policy Analysis & Management

Can successful schools replicate? In a climate of school turnarounds, charter conversions, and new school openings, an important question is whether schools that have demonstrated success with in one or several schools can replicate their success with additional campuses. In this paper, we use a change in the Massachusetts law that authorizes charter schools to examine whether charters deemed by the state as “proven providers” have the largest impacts on student outcomes and whether their replication campuses are similarly successful. We find that “proven providers” are indeed proven. Before expansion, proven provider schools had impacts of about a third of a standard deviation per year of charter school attendance in math, which is significantly larger gains per year in “other charters” before expansion. Proven providers also had larger gains in English/language arts (ELA) before charter expansion. After the expansion, proven providers and their replication campuses have very similar impacts on student test scores: 0.36σ for parent campuses and 0.33σ for replicate campuses in math; 0.18σ for parent campuses and 0.23σ for replicate campuses in ELA. There are no significant differences between parent and replicate test score gains. While this replication was on a relatively small scale, we find evidence that successful schools can recreate themselves.