Charter Schools and Vouchers
Urban charters work broadly (not just 'No Excuses'); vouchers and virtual charters often harm students.
Urban 'No Excuses' charter schools produce large, replicable gains. Non-urban charters and large-scale voucher programs often show null or negative effects.
Overview
Research on school choice reveals highly heterogeneous effects that depend heavily on the specific model, regulatory environment, and student population served. The school choice literature is also among the most politically contested in education research, with advocates and critics often selectively citing evidence to support their prior positions.
The 'No Excuses' Charter Model
The most robust positive findings come from urban 'No Excuses' charter schools. Angrist et al. (2013) found that Boston charter schools generated substantial test score gains: approximately d = 0.20 in ELA and d = 0.40 in math per year of middle school attendance. Abdulkadiroğlu et al. (2016) confirmed similar large effects in New Orleans and Boston takeover schools. Cohodes, Setren, and Walters (2021) demonstrated that Boston charters maintained large positive effects even as the sector scaled up, suggesting the model is replicable within urban contexts. Conversely, non-urban charters and virtual charter schools generally show null or negative effects. The CREDO (2015) national report found that virtual charter schools produce large negative effects — students losing the equivalent of 180 days of learning in math.
Voucher Programs and the Supply-Side Mechanism
Evidence on private school vouchers is mixed and increasingly negative in recent large-scale programs. While early evaluations of the Milwaukee voucher program (Rouse, 1998) showed modest positive effects, Abdulkadiroğlu, Pathak, and Walters (2018) found that the Louisiana Scholarship Program caused severe test score declines — approaching d = −0.40 in math — for voucher recipients in the first year. The divergence between early small-scale results and recent statewide programs likely reflects supply-side dynamics: when programs scale beyond the capacity of high-quality private schools, average effects decline precipitously.
The school choice literature illustrates the dangers of generalizing from the best examples to the average. Urban 'No Excuses' charters produce genuine, large, and replicable gains for disadvantaged students in cities. But these results cannot be extrapolated to voucher programs, virtual charters, or rural charter schools. The regulatory environment and quality control mechanisms are decisive: choice without quality assurance is likely to harm the students it is intended to help.
Key Papers
- Abdulkadiroğlu et al. (2016)Charters without Lotteries: Testing Takeovers in New Orleans and Boston
- Abdulkadiroğlu & Pathak & Walters (2018)Free to Choose: Can School Choice Reduce Student Achievement?
- Angrist & Pathak & Walters (2013)Explaining Charter School Effectiveness
- Cohodes & Setren & Walters (2021)Can Successful Schools Replicate? Scaling Up Boston's Charter Schools
- (2015)Urban Charter School Study: Report on Charter School Performance in 41 Urban Regions
- Rouse (1998)Private School Vouchers and Student Achievement: An Evaluation of the Milwaukee Parental Choice Program