A Problem-Based Learning Meta-Analysis: Differences Across Problem/Implementation Types, Disciplines, and Assessment Levels
Walker, A., & Leary, H. · 2009
grade Cmeta-analysisindependentunreplicated
Sample
82 studies / 201 outcomes
Population
Higher-ed & professional education across disciplines; PBL vs lecture.
Design
Broadens PBL beyond flattering medical-skills outcomes; non-random cohort designs dominate.
Key findings
Across disciplines the AVERAGE PBL advantage is small (d=0.13); its edge is concentrated in application/principle assessments and skills, not aggregate achievement.
Genetic confound
Domain outcomes; cohort selection confound.
Replication notes
Deflates the 'PBL works' headline once non-medical-skills outcomes are included.
DOI / URL
10.7771/1541-5015.1061
Effects
| Outcome | Metric | Value | Measure | Timing | Vs | Horizon | Class |
|---|---|---|---|---|---|---|---|
| Overall PBL vs traditional (all outcomes) | d | 0.13 (small positive) | mixed | mixed; classroom | active-alternative | unclear | domain-skill |
| By assessment level | direction | PBL relatively better at principle/application levels than concept recall | mixed | mixed | active-alternative | unclear | domain-skill |
Cited by
- Explicit instruction vs discovery/inquiry — how much guidance?moderate supportconf: mediumgc: low