Structural equation modeling (SEM), a powerful multivariate statistical method which can test and refine theoretical models, is being used in the social and behavioral sciences with increasing frequency. In the psychological literature, for example, SEM citations have steadily risen since 1979; SEM now rivals analysis of variance (ANOVA) in "statistical method popularity." In published health education/health behavior research, however, SEM has yet to reach such popularity. In an electronic search of articles published between 1996 and 2004 in one major health education/health behavior research journal, we found only three papers which utilized SEM, and 21 additional papers which used some other multivariate analytic technique. Thus, only a fraction of health education/health behavior research studies published over the past nine years go beyond using univariate methods, such as ANOVA and regression. This may be due, in part, to the complexity of SEM analyses. For instance, a working knowledge of matrix algebra is required for one to fully understand how to conduct SEM. Courses on SEM (and definitely on matrix algebra!), however, are usually not required, and sometimes not even available, for many doctoral students in public health, health behavior, health education, and health promotion. The purpose of this presentation is to introduce--to health education researchers--the "state of the art" in SEM as a multivariate analytical technique. We will 1) summarize the theory behind SEM, presenting its strengths as a multivariate analytic tool in health education research, 2) define its purpose and provide an overview of the basic steps involved in conducting SEM, 3) introduce the language and concepts unique to SEM, and 4) present cautionary notes related to using SEM. Finally, we will 5) provide a host of resources available to both new and experienced users of SEM, including a review of SEM software packages.