Same Data Set, Different Conclusions: Preschool Delay of Gratification Predicts Later Behavioral Outcomes in a...

Area of Science:

• Psychology
• Developmental Psychology
• Child Development

Background:

• The marshmallow test, assessing preschool delay of gratification, has long been linked to significant life outcomes.
• Recent conceptual replications have questioned the robustness and interpretation of these findings.
• This study addresses the need for independent verification and re-examination of the predictive validity of delay of gratification.

Purpose of the Study:

• To conduct an independent, preregistered secondary analysis of existing data.
• To test the conceptual replication of longitudinal associations between preschool delay of gratification and adolescent outcomes.
• To explore alternative explanations for the observed predictive relationships.

Main Methods:

• Secondary analysis of a dataset used in a recent replication study.
• Application of bivariate, multivariate, and multilevel statistical models.
• Comparison of analytic approaches between original studies and replication attempts.

Main Results:

• Significant associations were found for three out of five tested adolescent outcomes.
• Longitudinal relationships between delay of gratification and problem behavior were replicated.
• However, these relationships were not significant in a recent, parallel replication using the same data.
• Social support emerged as a stronger explanatory factor than self-control.

Conclusions:

• The predictive power of the marshmallow test may be linked to environmental factors like social support rather than solely self-control.
• The findings suggest a nuanced interpretation of the marshmallow test's predictive utility.
• This reinterpretation opens new avenues for early childhood interventions focused on environmental influences.