Journal Article
Skill formation and the trouble with child noncognitive skill measures
Publication details
Summary
This paper develops a framework to address issues of contamination in parent-reported measures of child noncognitive skills. We estimate a dynamic model in which child and parental skills evolve jointly and leverage information provided by teachers and interviewers to deal with contamination of parent-reported measures. The model also allows us to examine the relative importance of mothers and fathers in the evolution of child skills. Our findings reveal that ignoring contamination significantly underestimates the role of maternal non-cognitive skills in the evolution of child noncognitive skills. Additionally, we find evidence of stronger feedback effects from child skills to mothers than fathers. Simulation exercises demonstrate how contamination can distort evaluations of early childhood policies, underscoring the importance of robust measurement approaches.
Notes
Copyright © 2026 The Authors.
Open Access
Licensed under the Creative Commons Attribution-NonCommercial License 4.0.
Code:The replication package for this paper is available at https://doi.org/10.5281/zenodo.17290197. The authors were granted an exemption to publish their data because either access to the data is restricted or the authors do not have the right to republish them. Therefore, the replication package only includes the codes but not the data. However, the authors provided the Journal with (or assisted the Journal to obtain) temporary access to the data. The Journal checked the restricted data and the provided codes for their ability to reproduce the results in the paper and approved online appendices. Given the highly demanding nature of the algorithms, the reproducibility checks were run on a simplified version of the code, which is also available in the replication package.