Measurement models for studying child executive functioning: Questioning the status quo.

Area of Science:

• Developmental Psychology
• Cognitive Science
• Psychometrics

Background:

• Executive functioning (EF) is a key developmental construct with broad interest.
• Current research predominantly uses reflective latent variable models to measure EF.
• The appropriateness of these reflective models for representing EF in children is questioned.

Purpose of the Study:

• To critically evaluate the prevalent use of reflective latent variable models for child executive functioning.
• To compare reflective latent variable models with alternative composite models for measuring EF.
• To investigate how measurement model choice impacts substantive findings in child EF research.

Main Methods:

• Literature review examining the theoretical basis for reflective versus composite EF models.
• Analysis of longitudinal data from the Family Life Project (N=920) including preschool and first-grade children.
• Comparison of substantive analysis outcomes when EF is modeled as a reflective latent variable versus a composite variable.

Main Results:

• Findings indicate that measurement model specification significantly influences conclusions drawn from EF research.
• The choice between reflective latent variable and composite models leads to divergent interpretations of child executive functioning.
• The study highlights potential limitations of the dominant reflective model approach in developmental research.

Conclusions:

• Reflective latent variable models may not be the most suitable representation for the construct of executive functioning in children.
• Composite models offer a viable alternative that can yield different, potentially more accurate, research findings.
• Recommendations are provided for researchers to consider measurement model implications for robust developmental research on child EF.