This study draws on theories of Self-Regulated Learning and Situated Expectancy-Value Theory (SEVT) to understand how logs and self-reports of reviewing behaviors relate to each other and to motivation and performance within a university introductory computer science course. Only logs of reviewing from within coding problems were associated with comparable surveys when total course engagement clicks were controlled. Moreover, only course cost, an aspect of SEVT predicted logs and self-reports of reviewing behaviors. When motivation and both types of reviewing behaviors were considered together as predictors of final course grade, only course cost was a statistically significant predictor, with a negative relationship. Results can inform assumptions of validity between logs and self-reports of reviewing behaviors within a system of Self-Regulated Learning and demonstrate the important role perceptions of cost may play in both self-regulation and performance.