Clinical simulations, in a variety of forms, is a viable educational tool, allowing CSD students to acquire professional competencies and skills. Simucase is a computer-based simulation program designed for this objective. The goal of this study was to determine what elements of simulation engagement predicted success on a student's overall ability to make the correct recommendation for patient care, and what those predictors can tell us about how students navigate computer-based simulations. The data set used for this study comprised 149 graduate students in communication sciences and disorders (CSD) programs who completed a computer-based assessment simulation for a patient with aphasia. To determine which areas of the simulation predicted student success, a logistic regression was performed to determine which of the 12 types of decision points offered predictive data for making the correct final recommendation. The 12 types of decisions used comprised case history, collaborator, assessment, and diagnosis sections with reflective, acceptable, and rejected options in each. Results indicate that student patterns of case engagement can predict overall case success. The overall model was significant and individual predictors were significantly responsible for predicting which students would choose the correct outcome at the end of the case. This study revealed that students who engage in more careful navigation of preliminary assessment steps such as case history and collaborators were more likely to reach the correct recommendation at the end of the case. This finding has implications for the implementation of computer-based simulations for clinical education.