Background: Structural Equation Modeling (SEM) is a powerful and broadly utilized statistical framework. Researchers employ these models to dissect relationships into direct, indirect, and total effects (Bollen, 1989). These models unpack the "black box" issues within cause-and-effect studies by examining the underlying theoretical…

Researchers in behavioral sciences are often interested in longitudinal behavior change outcomes and the mechanisms that influence changes in these outcomes over time. The statistical models that are typically implemented to address these research questions do not allow for investigation of mechanisms of dynamic change over time. However, latent…

Research questions regarding how, for whom, and where a treatment achieves its effect on an outcome have become increasingly valued in substantive research. Such questions can be answered by causal moderated mediation analysis, which assesses the heterogeneity of the mediation mechanism underlying the treatment effect across individual and…

Latent variable structural equation modeling has become the analytic method of choice in many domains of research in psychology and allied social sciences. One important aspect of a latent variable model concerns the relations hypothesized to hold between latent variables and their indicators. The most common specification of structural equation…

Purpose: The purpose of this paper is to conduct a critical review of the mediation studies published in the field of Human Resource Development (HRD) to discern if the study designs, the nature of data collection and the choice of statistical methods justify the causal claims made in those studies. Design/methodology/approach: This paper conducts…

Structural equation modelling (SEM) is a method for analysis of multivariate data from both non-experimental and experimental research. The method combines a structural model linking latent variables and a measurement model linking observed variables with latent variables. Its use in social science and educational research has grown since the…

Purpose: The overall purpose of the European Union-funded Leadership Improvement for Student Achievement (LISA) project was to explore how leadership styles, as conceptualized in the developed dynamic holistic leadership framework, directly or indirectly affect student achievement at the lower secondary level of education in seven European…

This paper examines the estimation of two-stage clustered RCT designs in education research using the Neyman causal inference framework that underlies experiments. The key distinction between the considered causal models is whether potential treatment and control group outcomes are considered to be fixed for the study population (the…

Discusses linear structural relations, Two Stage Least Squares, and path analysis as statistical procedures that sometimes permit causal inferences from correlational findings. Even though two variables cannot be interpreted causally due to a possible but unknown third variable, these methods are appropriate for handling models with correlated…

Nonlinear relationships and latent variable assumptions can lead to serious specification errors in structural models. A quadratic relationship, described by a linear structural model with a latent variable, is shown to have less predictive validity than a simple manifest variable regression model. Advocates the use of simpler preliminary…

Structural equation models merge multiple regression, path analysis, and factor analysis techniques into a single data analytic framework. Measurement models are developed to define latent variables, and structural equations are then established among the latent variables. Explains the development of these models. (KS)

This overview of nonexperimental causal research methods focuses on latent variable structural equation modeling using the LISREL computer program. An extended example in special education is used to present LISREL as an extension of structural equations analysis (path analysis) and as a method of reducing the effects of error in research.…

In studies of relatives, conventional multiple regression may not be appropriate because observations are not independent. Obtaining estimates of regression coefficients and correct standard errors from these populations through a structural equation modeling framework is discussed and illustrated with data from twins. (SLD)