Learning Strategies Shape Math Achievement Through Motivation and Engagement, Study Finds

Students who know how to study may only reap the full benefit of that knowledge when they also feel motivated and stay engaged, according to a new study of 444 middle school and high school students published in Discover Education. The research tested whether learning strategies — the cognitive, metacognitive, and self-regulatory tools students use to plan, monitor, and evaluate their own learning — are linked to mathematics achievement directly, or whether that connection travels through two intermediate processes: motivation and engagement in the classroom. The answer, based on a serial mediation model, was that the direct link between strategies and achievement disappears once motivation and engagement are taken into account, while all three hypothesized indirect pathways remain statistically significant.

The study, conducted by Şahin Danişman, used a relational survey design with a convenience sample of 458 students, from which 14 outliers were removed after screening with z-scores, leaving 444 participants: 207 secondary school students (96 female, 111 male) and 237 high school students (124 female, 113 male). Achievement was measured through students’ self-reported mathematics grades for the year, which ranged from 28 to 100 and were standardized into z-scores before analysis. Learning strategies were assessed with the Motivated Strategies for Learning Questionnaire, a widely used instrument covering both motivational orientations and strategy use, while engagement was measured with the Engagement and Disaffection Scale, which captures behavioral and emotional engagement as well as their negative counterparts. In the present sample, the reliability coefficients were 0.86 for the motivation portion of the questionnaire and 0.94 for the learning strategies portion, and 0.82 for the engagement scale.

Before testing the mediation model, the researchers ran a four-step hierarchical regression to see how much variance in mathematics achievement each block of variables explained. Grade level and gender, entered first as control variables, accounted for 3 percent of the variance. Learning strategies added a further 9 percent when entered in the second step, indicating a meaningful unique contribution. But the picture changed as motivational strategies were added in the third step, contributing another 9 percent, and engagement in the fourth step, adding 2 percent more. Crucially, once motivational strategies entered the model, the effect of learning strategies became statistically non-significant, and the effects of both learning strategies and motivational strategies diminished further after engagement was included — a pattern the authors read as a signal of potential mediating effects that warranted formal testing.

That formal test used bootstrapping with 10,000 resampling iterations, implemented through the PROCESS macro (Model 6) for SPSS, with 95 percent bias-corrected confidence intervals used to judge significance. The results showed that learning strategies had significant positive effects on motivation (B = 0.62, p < .001) and on engagement (B = 0.16, p < .001), and that motivation in turn positively predicted engagement (B = 0.24, p < .001). Both motivation (B = 0.38, p < .001) and engagement (B = 0.34, p < .01) positively predicted mathematics achievement. The total effect of learning strategies on achievement was positive and significant (B = 0.37, p < .001), yet the direct effect was not (B = 0.03, p > .05) — the statistical signature of full mediation.

All three hypothesized indirect pathways were supported. Learning strategies predicted mathematics achievement through motivation alone (B = 0.23, 95 percent CI [0.14, 0.33]), through engagement alone (B = 0.05, 95 percent CI [0.02, 0.11]), and through the serial route in which strategies first raise motivation, which then boosts engagement, which in turn lifts achievement (B = 0.05, 95 percent CI [0.02, 0.09]). When the researchers compared the sizes of these indirect effects, the pathway through motivation proved significantly stronger than both the pathway through engagement and the serial pathway. The difference between the serial pathway and the engagement-only pathway was not significant. In practical terms, motivation emerged as the dominant channel through which students’ strategic learning translated into mathematics grades.

The findings fit into a broader body of evidence on self-regulated learning. Meta-analytic research has reported a positive relationship between self-regulation and student achievement across educational contexts, and mathematics-specific studies have linked self-regulated learning and cognitive and metacognitive strategy use to mathematics performance. Among middle school students, those with higher mathematics achievement have been found to report a greater variety of learning strategies and a broader range of strategy categories than lower-achieving peers, even though the groups did not differ in the overall number or frequency of strategies used. Undergraduate students with higher achievement have likewise shown stronger use of strategies including information processing, concentration, self-testing, and test strategies. Together, these findings suggest that successful learning depends not simply on using more strategies, but on developing a sufficiently varied strategic repertoire matched to the demands of the task.

Previous work has also hinted that the strategy–achievement link may not be direct. One study found that learning engagement fully mediated the relationship between memorization and elaboration strategies and mathematics performance, suggesting that the effectiveness of strategy use is tied to the regularity and persistence with which students engage in learning. Other research has shown that motivation can mediate the relationship between collaborative learning and mathematics achievement, and that mastery-approach goals predict subsequent behavioral engagement, which in turn predicts achievement. Longitudinal evidence spanning Grades 5 to 10 indicates that both motivation and cognitive learning strategies predict long-term growth in mathematics achievement over time, even though intelligence is related to initial achievement levels. The new study extends this literature by examining motivation and engagement simultaneously as indirect pathways within a single integrated model.

The authors are careful to frame the mediation as statistical rather than definitive evidence of temporal or causal ordering. The study is cross-sectional and correlational, relying on convenience sampling and self-reported grades, so the direction of effects cannot be established from these data. Indeed, prior longitudinal research has found that the relationship can run the other way: motivation has been shown to predict subsequent strategy use in mathematics learning, not only the reverse. Mediation findings are also not uniform across contexts — in one study, mathematics motivation did not mediate the relationship between mathematics self-efficacy and achievement, suggesting that the explanatory role of motivation may depend on the antecedent variable and the learning setting. These caveats place the present results firmly in the category of associational evidence about plausible mechanisms rather than proof of cause.

Even so, the practical implications the authors draw are measured and specific. If the association between strategies and achievement operates largely through motivation and engagement, then simply teaching students how to study may not be enough; instruction should also create conditions that sustain students’ willingness to use those strategies. The authors suggest that teachers can encourage students to set learning goals, monitor their own progress, reflect on the effectiveness of the strategies they use, participate actively in mathematical problem solving, and discuss different solution approaches with peers. Meaningful tasks, constructive feedback, and opportunities for active participation may help students apply strategies more consistently and purposefully, particularly during examination preparation, when students may know appropriate strategies but fail to deploy them when motivation or engagement is low.

The findings may be especially useful for identifying a familiar type of student: one who possesses knowledge of learning strategies but shows low motivation or limited classroom engagement. For such students, the authors argue, additional strategy instruction alone may have limited value unless motivational and engagement-related difficulties are also addressed, meaning that interventions — including examination-preparation programs — may work best when they simultaneously target strategies, motivation, and engagement. Given the correlational nature of the study, however, these implications should be read as directions for instructional practice and future intervention research rather than as evidence of causal effects. What the study does offer is a more process-oriented account of how students’ strategic approaches to learning become associated with mathematics achievement: not as a straight line from technique to grade, but as a chain that runs through whether students want to learn and whether they show up, mentally and behaviorally, to do so.

Subject of Research: The mediating roles of motivation and engagement in the relationship between learning strategies and mathematics achievement among middle and high school students

Article Title: Learning strategies predict mathematics achievement through motivation and engagement

Article References: Danişman, Ş. (2026). Learning strategies predict mathematics achievement through motivation and engagement. Discover Education, 5(1), Article 1222. https://doi.org/10.1007/s44217-026-02331-x

Image Credits: AI Generated

DOI: 10.1007/s44217-026-02331-x

Keywords: learning strategies, mathematics achievement, motivation, student engagement, self-regulated learning, mediation analysis, hierarchical regression, middle school students, high school students, MSLQ, Discover Education, educational psychology