How Big Are Effect Sizes in International Education Studies?
Author(s) -
David K. Evans,
Fei Yuan
Publication year - 2022
Publication title -
educational evaluation and policy analysis
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 3.636
H-Index - 78
eISSN - 1935-1062
pISSN - 0162-3737
DOI - 10.3102/01623737221079646
Subject(s) - randomized experiment , econometrics , psychological intervention , variance (accounting) , scale (ratio) , distribution (mathematics) , psychology , demographic economics , economics , public economics , statistics , mathematics , geography , accounting , mathematical analysis , cartography , psychiatry
A growing literature measures the impact of education interventions in low- and middle-income countries on both access and learning outcomes. But how should one contextualize the size of impacts? This article provides the distribution of standardized effect sizes on learning and access from 234 studies in low- and middle-income countries. We identify a median effect size of 0.10 standard deviations on learning and 0.07 standard deviations on access among randomized controlled trials. Effect sizes are similar for quasi-experimental studies. Effects are larger and demonstrate higher variance for small-scale studies than for large-scale studies. The distribution of existing effects can help researchers and policymakers to situate new findings within current knowledge and design new studies with sufficient statistical power to identify effects.
Accelerating Research
Robert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom
Address
John Eccles HouseRobert Robinson Avenue,
Oxford Science Park, Oxford
OX4 4GP, United Kingdom