Impacts of a Prekindergarten Program on Children's Mathematics, Language, Literacy, Executive Function, and Emotional Skills
Weiland, C., & Yoshikawa, H. · 2013
grade Cquasi-experimentindependentmixed
Sample
~2,018
Population
Boston public preschool, short-run outcomes.
Design
Age-cutoff RD. Measures end-of-treatment skills only.
Key findings
Large end-of-treatment gains for Boston pre-K. Set against Gray-Lobe's long-run precise zero on the same program, this pair is the clearest illustration in the database that impressive end-of-treatment effects predict nothing about durability.
Counterfactual
Children about to enter the SAME program (one year younger), not an alternative-care comparison — a regression-discontinuity on age of eligibility.
Genetic confound
RD on birth-date cutoff; but the design compares treated to not-yet-treated, which inflates apparent effects by including a year of ordinary maturation.
Replication notes
The same program's long-run lottery evaluation (Gray-Lobe 2023) finds ZERO test-score effect — a caution about reading end-of-treatment RD effects as durable learning.
DOI / URL
10.1111/cdev.12099
Effects
| Outcome | Metric | Value | Measure | Timing | Vs | Horizon | Class |
|---|---|---|---|---|---|---|---|
| Early literacy (WJ Letter-Word ID) | d | +0.62 | standardized | end-of-treatment | business-as-usual | end-of-treatment | domain-skill |
| Mathematics | d | +0.59 | standardized | end-of-treatment | business-as-usual | end-of-treatment | domain-skill |
| Receptive language (PPVT) | d | +0.44 | standardized | end-of-treatment | business-as-usual | end-of-treatment | domain-skill |
| Executive function | d | +0.20 to +0.28 on measures reaching significance | standardized | end-of-treatment | business-as-usual | end-of-treatment | near-transfer |
Cited by
- Preschool at scale — Head Start, state pre-K, and what universal provision deliversmixedconf: highgc: low