The Interplay of Personal and Situational Factors in Situated Expectancy-Value Theory: Cross-Classified Analyses of Students’ Situation-Specific Motivations in the Math Domain

Many students in STEM programs experience motivational declines early in their studies, which predict later academic difficulties and dropout. The psychological processes that make some students more susceptible to motivational declines in achievement situations are still not well understood. Drawing on situated expectancy-value theory, this study examined the motivational beliefs of beginning STEM students in Germany (N = 3,213) who completed a low-stakes, statewide test of math prerequisites. Students rated their test-specific motivation four times during the test as they worked on different math content (e.g., basic arithmetic, geometry, calculus). Cross-classified variance decomposition analyses indicated that variation in students’ test-specific motivation and performance was primarily due to differences between individuals (person-level variance) and, to a lesser extent, differences between tested math content (content-level variance) and residual situation-specific variance. Cross-classified random effects models further showed that test performance and the difficulty of different content areas predicted students’ motivational beliefs after working on each test section. Importantly, even after controlling for differences in actual test performance, students with higher prior math achievement and higher-quality learning opportunities in high school reported higher test-specific motivation. Cross-level interactions indicated that positive prior experiences boosted students’ test-specific motivation independent of their test performance. Students’ test performance predicted their later performance in math-intensive STEM courses. These findings suggest that efforts to support STEM students’ motivation should consider the lasting impact of previous experiences with math, as these experiences shape students’ perceptions and engagement with new achievement situations within STEM—even in the absence of situation-specific performance differences.