Prior research regarding practice effects on ability tests has focused primarily on the differences in average score gains across different types of preparation (e.g., test retaking vs. focused item practice vs. coaching). In contrast, there has been little concerted effort towards understanding the significant variance in score gains across people and tests, within the same preparation condition. In response to calls for such research, this study examines the efficacy of the degree of a test's g-saturation and test-taker characteristics for explaining variance in score gains within a single preparation condition (i.e., retesting). Our results show that the sizes of the score gain due to practice is negatively associated with the degree of g-saturation of the ability scales. Results also show that score gains are generally positively associated with test-taker beliefs about tests and motivation; however, these associations were specific to certain scales.