The Evidence on Teaching

Delay of Gratification in Children

Mischel W, Shoda Y, Rodriguez ML · 1989

grade Dreviewdeveloper-ledmixed
Sample
Narrative synthesis of the Stanford Bing Nursery School delay programme
Population
Upper-middle-class preschoolers in the Stanford university community.
Design
The programme summary that carried the marshmallow finding into public consciousness. It is a review rather than a study and reports no effect sizes in its abstract. Graded D accordingly.
Key findings
The claim, verbatim, is that four-year-old children who delayed gratification longer in certain laboratory situations developed into more cognitively and socially competent adolescents, achieving higher scholastic performance and coping better with frustration and stress. Delay time is an objective behavioural measure; the adolescent outcomes are parent-rated.
Genetic confound
High. Observational prediction in a homogeneous sample with no family-background controls.
Replication notes
The predictive claim was tested at scale by Watts, Duncan and Quan 2018 and shrank by roughly half before disappearing under controls, and the original cohort's own mid-life follow-up found the preschool measure predicts nothing.
DOI / URL
10.1126/science.2658056

Effects

OutcomeMetricValueMeasureTimingVsHorizonClass
Preschool delay predicting adolescent competenceclaimlonger delayers became more cognitively and socially competent adolescents; no effect sizes in the abstractmixedabout 14 yearsnonenot-applicablenon-cognitive

Cited by

Delay of Gratification in Children · The Evidence on Teaching