A number of school choice advocates claim that there is a research consensus indicating that vouchers for private schools lead to higher academic achievement. The authors review and critique these local studies of voucher programs, contrasting them with findings from larger-scale analyses of nationally representative samples of public and private schools, which illuminate patterns that appear to undercut the assumption of superior private school performance that is a premise of voucher programs. The authors note limitations inherent in different methodological approaches to this question, focusing on the shortcomings of randomization as an exclusive "gold standard" for research on the issue of achievement in school choice plans. The concluding discussion reconsiders the question of a consensus, highlighting the emerging research environment that bypasses traditional review processes and emphasizes instead the promotion of ideas to support policy agendas. (Contains 23 notes.)