Formative Assessment: A Meta-Analysis and a Call for Research (with the Briggs et al. 2012 / McMillan et al. 2013 critiques)
Kingston, N., & Nash, B. · 2011
grade Cmeta-analysisindependentmixednumbers spot-checked
Sample
13 usable studies (of 300+ screened) / 42 effect sizes
Population
K-12, US-dominated.
Design
The only computed K-12 FA meta. Measure type uncoded. Contested in BOTH directions: Briggs et al. 2012 (selection, effect computation, non-independence); McMillan et al. 2013 ('not credible' — as an UNDERestimate, naming no alternative).
Key findings
The deliberate corrective: the commonly cited d≈0.7 'is not supported by the existing research base.' Only 13 of 300+ studies were usable — the FA literature is mostly unusable for causal inference.
Genetic confound
Mostly classroom quasi-experiments; ability composition often uncontrolled.
Replication notes
Contested but never superseded; the honest bracket for the K-12 FA meta layer is ~0.1-0.3 with wide uncertainty.
DOI / URL
10.1111/j.1745-3992.2011.00220.x
Effects
| Outcome | Metric | Value | Measure | Timing | Vs | Horizon | Class |
|---|---|---|---|---|---|---|---|
| Achievement, overall | d | 0.20 (random effects) | mixed | end-of-treatment | business-as-usual | end-of-treatment | domain-skill |
| By subject | d | ELA 0.32 / math 0.17 / science 0.09 | mixed | end-of-treatment | business-as-usual | end-of-treatment | domain-skill |
Cited by
- Feedback and formative assessmentmixedconf: highgc: low