Structural equation modeling (multivariate analysis with latent variables, also called causal modeling or covariance structure analysis) is a valuable methodological tool for use in counseling psychology research. Essentially the broad framework that subsumes many well-known procedures (e.g., multiple linear regression, factor analysis, path analysis), structural equation modeling allows for analysis of causal patterns among unobserved variables represented by multiple measures. It permits testing of causal hypotheses and theory, examination of psychometric adequacy, and enhancement of the explanatory power of correlational data.that characterize counseling psychology research. I present and illustrate structural equation modeling, followed by a discussion of (a) issues and problems related to the use of this methodology, (b) possible applications of structural equation modeling to counseling psychology research, and (c) resources for those wanting further study. Borgen (1984) asserts that the field of counseling psychology is in epislemological disarray, and points out the importance of integrating our fragmented counseling theories and methodological approaches into coherent frameworks that better represent the complexity of the counseling process. He stresses the point previously made by others (e.g., Gelso, 1979;Meehl, 1978) that limitations in our methodological sophistication often impede theoretical progress and hinder complete understanding of the myriad variables thought to be important in the process of personal development and change in counseling. Borgen suggests that models of causation deserve increased attention and that they may allow us to begin this process of theoretical and empirical synthesis. What appears to be needed, then, is a methodology that allows the examination of causal and interactional relations among variables in an integrated, testable form.A promising methodological advance in this regard is structural equation modeling or multivariate analysis with latent variables, also called causal modeling and covariance structure analysis (Bentler, 1980). This method allows the use of correlational and nonexperimental (as well as experimental) data to determine the plausibility of theoretical models in specific populations. Hypothesized in the structural equation model is a specified causal structure among a set of unobservable constructs, each measured by a set of observed indicator variables; this model can then be tested for fit in a particular population. A full structural equation model consists of two components: (a) a structural model that specifies the hypothesized causal structure among latent variables {theoretical constructs not directly observable); and (b) a measurement model that defines relations between measured variables or indicators {variables that are observed directly) and the latent I would like to thank Bruce E. Wampold and two anonymous reviewers for their comments on an earlier draft of this article.Correspondence concerning this article should be addressed to Ru...