StructuralEquationModeling.doc

Structural Equation Modeling Introduction Things you should know before attempting SEM Kinds of research questions addressed by SEM Limitations and assumptions relating to SEM Preparing your data for analysis Developing the Model Hypothesis and Model Specification Drawing your hypothesized model: procedures and notation Specifying equation sets Specifying covariances, residuals, and variances Estimating, fixing, and constraining variables Estimation methods Determining the number of data points for testing Overidentification Preparing a program to analyze your model Other issues Assessing Model Fit Comparative fit indices Variance-accounting indices Parameter-based indices Model Modification Modifying a model to improve fit The Chi-square difference test The Lagrange-multiplier test The Wald test Examples Introduction A. Things You Should Know Before Using Structural Equation Modeling. Structural equation modeling (SEM) is a series of statistical methods that allow complex relationships between one or more independent variables and one or more dependent variables. Though there are many ways to describe SEM, it is most commonly thought of as a hybrid between some form of analysis of variance (ANOVA)/regression and some form of factor analysis. In general, it can be remarked that SEM allows one to perform some type of multilevel regression/ANOVA on factors. You should therefore be quite familiar with univariate and multivariate regression/ANOVA as well as the basics of factor analysis to implement SEM for your data. Some preliminary terminology will also be useful. The following definitions regarding the types of variables that occur in SEM allow for a more clear explanation of the procedure: Variables that are not influenced by another other variables in a model are called exogenous variables. As an example, suppose we have two factors that cause changes in GPA, hours studying per week and IQ. Suppose there is no causal relationship between hours studying