A Kindergarten Lottery Evaluation of Core Knowledge Charter Schools

~2,300 applicants / 14 lotteries / 9 schools

~2,300 Colorado kindergarten-lottery applicants; mostly middle-income districts + one low-income CK charter.

Lottery-based (grade-A identification) evaluation of oversubscribed Core Knowledge (E.D. Hirsch content-rich curriculum) charters. Outcomes = grade 3-6 STATE standardized reading (independent measures), reflecting cumulative multi-year schooling. Single, not-yet-peer-reviewed working paper; whole-school bundle.

The strongest causal support for the build-broad-knowledge hypothesis: random assignment to content-rich schooling raised grade 3-6 reading on INDEPENDENT state tests by ITT 0.24 / TOT 0.47 SD (up to ITT 0.94 at a low-income CK charter). Because effects are on standardized measures and lottery-identified, they resist both measure-inflation and genetic-confound critiques - but rest on a single unreplicated working paper of a whole-school bundle, so the knowledge-specific share is unisolated.

Largely addressed by lottery randomization (winners vs losers balanced on genes/family selection) - its central strength vs the correlational knowledge literature.

Not yet replicated or peer-reviewed; complements Kim MORE content-literacy RCT (ES~0.18, near>far transfer). Single-study => caps confidence at medium.