Using Structural Equation Modeling in Place of Between-Subjects Analysis of Variance

Datos Bibliográficos

ID21641981
AutoresJonathan L Helm (0000-0001-8580-980X, San Diego State University, autor de correspondencia), Benedikt Langenberg (0000-0002-4757-0698, Bielefeld University), Emma Grossman (San Diego State University), Justine Poulin (San Diego State University), A Mayer (0000-0001-9716-878X, Bielefeld University)
Año2023
Volumen30
Número1
Páginas123-131
Fecha de publicación2023-01-02
Peer ReviewedSí
Open AccessNo
TipoARTICLE
RevistaStructural Equation Modeling: A Multidisciplinary Journal (JOURNAL)
Identificadores de la revistaISSN: 1070-5511 • E-ISSN: 1532-8007
EditorialInforma UK Limited (PUBLISHER • GB)
DOI10.1080/10705511.2022.2033977
OpenAlexW4224241196
IdiomaEN
Referencias citadas7

This article demonstrates how to use structural equation modeling (SEMing) in place of between-subjects analysis of variance (BS-ANOVA). More specifically, this article demonstrates how to closely reproduce the F-values and p-values from ANOVA (for all main and interactions effects) using an SEM model comparison approach (i.e., χ2 difference tests between two SEMs that conform to the main and interaction effects of the null and alternative hypotheses of an ANOVA main or interaction effect). Therefore, researchers less familiar with SEM can use this article to test hypotheses common to BS-ANOVA, and potentially use this article as a steppingstone for implementing more complex SEMs (including modern methods for missing data, estimation of latent variables, and relaxation of homogeneity of variance assumptions)

Analysis of variance · Econometrics · Interaction · Latent variable · Main effect · Mixed-design analysis of variance · Null hypothesis · One-way analysis of variance · Repeated measures design · Statistical analysis · Statistics · Structural equation modeling · Advanced Statistical Modeling Techniques · Behavioral and Psychological Studies · Mathematics · Psychometric Methodologies and Testing