The Evidence on Teaching

The Relationship Between Inquiry-Based Teaching and Students' Achievement: Longitudinal PISA Evidence from England

Jerrim, J., Oliver, M., & Sims, S. · 2019

grade Dlongitudinalindependentmixed
Sample
PISA 2015 England linked to National Pupil Database
Population
~15-year-olds, England; standardized national science achievement.
Design
Better-controlled than cross-national PISA (prior attainment, demographics), but exposure is self-reported and cross-sectional — fundamentally correlational.
Key findings
The best-controlled test of the 'PISA shows inquiry is bad' claim: in England's linked data the raw negative largely DISAPPEARS, becoming curvilinear/null, with inquiry+high teacher support modestly positive. Neither extreme survives clean identification.
Genetic confound
Correlational; confounded by ability sorting and reverse causation (weaker classes get more hands-on remediation).
Replication notes
Undercuts BOTH the strong pro-inquiry case and the anti-inquiry PISA correlation.
DOI / URL
10.1016/j.learninstruc.2018.12.004

Effects

OutcomeMetricValueMeasureTimingVsHorizonClass
Inquiry-teaching frequency -> science achievementassociationlittle/no significant association; naive negative attenuates; curvilinearstandardizedcross-sectionalnonenot-applicabledomain-skill
High inquiry + high teacher supportassociationsmall POSITIVEstandardizedcross-sectionalnonenot-applicabledomain-skill

Cited by