The Prospects and Limitations of Latent Variable Models in Educational Psychology

Authored by: Benjamin Nagengast , Ulrich Trautwein

Handbook of Educational Psychology

Print publication date:  July  2015
Online publication date:  July  2015

Print ISBN: 9780415894814
eBook ISBN: 9781315688244
Adobe ISBN: 9781317420569

10.4324/9781315688244.ch4

 Download Chapter

 

Abstract

Latent variable models, in particular, structural equation modeling (SEM), are a staple of research in educational psychology. SEM combines measurement models, which provide links between observed responses to items and unobserved latent variables, and structural models, which represent the multivariate relations between latent variables and manifest variables (Jöreskog, 1970; see Matsueda, 2012, for an overview of the history of SEM). The main application of latent variable models in educational psychology lies in the development and validation of measures of core constructs. Second, SEM is used to study and test relations between latent variables in structural models. Often, these structural relations are not only used descriptively but are also given a causal interpretation (e.g., in mediation analyses). There have been many methodological advances with respect to measurement models, structural models, and causal inference that have the potential to impact the ways in which researchers in educational psychology use SEM to analyze their data. In this chapter, we critically review the scope and limitations of applications of SEM in educational psychology and highlight the challenges that researchers will face in the upcoming decade.

 Cite
Search for more...
Back to top

Use of cookies on this website

We are using cookies to provide statistics that help us give you the best experience of our site. You can find out more in our Privacy Policy. By continuing to use the site you are agreeing to our use of cookies.