The purposes of this study were to develop a structural equation model of preservice teacher variables and to examine alternative methods of testing this model under different assumptions and measurement conditions. A conceptual model was posited to examine variables related to job expectations, academic ability, pre-student teaching preparation, satisfaction with cooperating teacher, self-rated student teaching performance, and quality of teacher preparation program;From this conceptual model of preservice teacher variables, five empirical models were developed and analyzed. Models I and II utilized single indicators computed as composite means of the respective observed standardized indicators. Model I assumed variables were measured without error. Model II utilized the same indicators as Model I but adjustments were made to allow for measurement error. Models III and IV used a multiple indicator approach with Model III assuming no measurement error, while Model IV adjusted for measurement error. Model V, assuming no measurement error, utilized the multiple indicators from Models III and IV but analyzed them independently. The method of analysis for Models I, III and V was multiple linear regression. Models II and IV were analyzed using the LISREL approach;Results of the analyses lent moderate support for the conceptual model. The results of two single indicator models indicated a satisfactory fit of the models to the data. However, the multiple indicator model adjusting for error did not provide a good fit. In addition to the individual model tests, parameter estimates from the different models were compared and discussed.