Curriculum Reform in the Common Core Era: Evaluating Elementary Math Textbooks Across Six US States
Blazar, D., Heller, B., Kane, T., Polikoff, M., Staiger, D., et al. · 2020
grade Cquasi-experimentindependentunreplicatednumbers spot-checked
Sample
~6,255 schools / 10,797 school-year observations, 6 states
Population
US public elementary schools, grades 4-5, post-CCSS era.
Design
First multi-state textbook efficacy study; school value-added with switcher models; powered to detect 0.05-0.10 SD effects.
Key findings
In the post-Common-Core textbook market, WHICH mainstream text a school adopts makes essentially no detectable difference (textbook SD ≈ 0.00-0.03) — plausibly because standards homogenized texts and teachers barely use/are barely trained on them. The pre-CCSS curriculum lever largely collapsed.
Genetic confound
Selection-on-observables; school-level.
Replication notes
Numbers verified (incl. corrected implementation stats). Contradicts the pre-CCSS 'cheap lever' magnitudes.
DOI / URL
10.1002/pam.22257
Effects
| Outcome | Metric | Value | Measure | Timing | Vs | Horizon | Class |
|---|---|---|---|---|---|---|---|
| Between-textbook variation in achievement gains | SD of textbook effects | 0.000 pooled (excl. CA); 0.027 CA; 0.028 WA; joint test ns (p=0.323) | standardized | annual gains | active-alternative | end-of-treatment | domain-skill |
| Best-case textbook advantage (CA 95th percentile) | SD | +0.054 | standardized | annual | active-alternative | end-of-treatment | domain-skill |
| Implementation reality | survey | 7% of teachers used the adopted text exclusively; 46% supplement heavily; median textbook-aligned PD = 3 days over entire careers | mixed | n/a | none | not-applicable | behaviour |
| EdReports alignment ratings vs efficacy | association | none | standardized | annual | none | not-applicable | domain-skill |
Cited by
- Curriculum choice as a school-level levermixedconf: highgc: low
- Math curriculum choice — how much does the textbook matter?mixedconf: highgc: low