American Behavioral Scientist · 2004 · 257 citations · 16 references
EducationEducational AssessmentEducational MeasurementAfrican American EducationCausal InferenceExperimental DesignAchievement TestingAfrican American Educational AttainmentEducation VouchersEquality Of Educational OpportunityHypothesis TestingVoucher PolicyAfrican American Student AchievementControl Group DesignEducational Attainment EffectsLow-income StudentsMatched Control GroupSample Size CalculationSample Size EstimationStandardized Test Scores
Random assignment in the NYC voucher experiment should make treatment status unrelated to baseline characteristics. This article reexamines data from the largest and best‑implemented private school scholarship experiment in New York City. The study followed low‑income kindergarten‑to‑Grade‑4 students who entered a monthly lottery for a private‑school scholarship in May 1997, collecting baseline and annual follow‑up data from students and parents over three years. Excluding students with missing baseline scores—those in kindergarten and 11 % of those in Grades 1–4—distorted achievement analyses, and the voucher effect varied with how race/ethnicity was defined.
This article reexamines data from the New York City school choice program, the largest and best-implemented private school scholarship experiment yet conducted. In the experiment, low-income public school students in kindergarten to Grade 4 were eligible to participate in a series of lotteries for a private school scholarship in May 1997. Data were collected from students and their parents at baseline and in the spring of each of the next 3 years. Students with missing baseline test scores, which encompasses all those who were initially in kindergarten and 11% of those initially in Grades 1 to 4, were excluded from previous analyses of achievement, even thoughthese students were tested in the follow-up years. In principle, random assignment would be expected to lead treatment status to be uncorrelated with all baseline characteristics. In addition, it was found that the effect of vouchers was sensitive to the particular way race/ethnicity was defined.
16