Effectiveness of Schema-Based Instruction for Improving Seventh-Grade Students' Proportional Reasoning: A Randomized Experiment
Jitendra AK, Star JR, Dupuis DN, Rodriguez MC · 2013
grade Brctdeveloper-ledmixed
Sample
1,163 students / 42 classrooms
Population
US 7th graders learning ratio, rate and percent.
Design
Pretest-intervention-posttest-retention cluster RCT of schema-based instruction: teach the underlying problem STRUCTURE (identify type, represent with a diagram, solve, check) with explicit self-monitoring heuristics. Developer-run, and every outcome measure was built by the research team — no independent standardized outcome in this trial.
Key findings
The efficacy trial that made schema-based instruction famous: g > 1.2 on trained content, retained at six weeks, with a null transfer test. Both halves matter — the technique teaches what it teaches, and only what it teaches, and the magnitude is an aligned-measure efficacy number, not a scale estimate.
Genetic confound
Randomized at classroom level. Not contraindicated for strugglers.
Replication notes
Scaled up by the same team in 2015 with 82 classrooms, where the effect shrank to ~0.32 — the within-programme shrinkage is itself instructive.
Effects
| Outcome | Metric | Value | Measure | Timing | Vs | Horizon | Class |
|---|---|---|---|---|---|---|---|
| Proportional problem solving (ratios/rates and percents) | g | 1.24 at immediate posttest; 1.27 at 6-week retention | researcher-designed | end of 6-week unit and +6 weeks | business-as-usual | under-1yr | domain-skill |
| Transfer to untrained problem types | g | not significant — transfer hoped for is not transfer got | researcher-designed | end of unit | business-as-usual | end-of-treatment | near-transfer |
Cited by
- Word problems, schema instruction, and the conceptual-vs-procedural questionmoderate supportconf: highgc: low