The Evidence on Teaching

Homeschooling: What do we know and what do we need to learn?

Valiente C, Spinrad TL, Ray BD, Eisenberg N, Ruof A · 2022

grade Creviewdeveloper-involvednot-applicable
Sample
narrative review
Population
US home-schooled children; academic and socioemotional outcomes
Design
Review in Child Development Perspectives. Notable for its authorship — mainstream developmental psychologists writing jointly with the head of NHERI — and for landing on a cautious verdict despite that.
Key findings
"Preliminary evidence suggests that homeschoolers perform equal to or better than their conventionally schooled peers on measures of achievement and socioemotional functioning, but methodological limitations, especially selection effects, make it premature to draw definitive conclusions." The review also documents the heterogeneity of the population (religious, secular, special-needs, gifted-acceleration) and of parental practice, and argues that pooled estimates across those groups are not interpretable.
Genetic confound
High; the review names selection as the binding limitation without attempting to bound it.
Replication notes
Reaches the same bottom line as the independent Kunzman & Gaither surveys despite very different authorship, which is the closest thing this literature has to a cross-checked conclusion.

Effects

OutcomeMetricValueMeasureTimingVsHorizonClass
Achievement and socioemotional functioning, home-schooled vs conventionally schooledqualitative summaryequal to or better, but "premature to draw definitive conclusions" given selection effectsmixedpooled across studiesactive-alternativeunclearunclear

Cited by

Homeschooling: What do we know and what do we need to learn? · The Evidence on Teaching