How the Predictors of Math Achievement Change Over Time: A Longitudinal Machine Learning Approach.

Researchers have focused extensively on understanding the factors influencing students' academic achievement over time. However, existing longitudinal studies have often examined only a limited number of predictors at one time, leaving gaps in our knowledge about how these predictors collectively contribute to achievement beyond prior performance and how their impact evolves during students' development. To address this, we employed machine learning to analyze longitudinal survey data from 3,425 German secondary school students spanning 5 to 9 years. Our objectives were twofold: to model and compare the predictive capabilities of 105 predictors on math achievement and to track changes in their importance over time. We first predicted standardized math achievement scores in Years 6–9 using the variables assessed in the previous year ("next year prediction"). Second, we examined the utility of the variables assessed in Year 5 at predicting future math achievement at varying time lags (1–4 years ahead)—"varying lag prediction." In the next year prediction analysis, prior math achievement was the strongest predictor, gaining importance over time. In the varying lag prediction analysis, the predictive power of Year 5 math achievement waned with longer time lags. In both analyses, additional predictors, including intelligence quotient, grades, motivation and emotion, cognitive strategies, classroom/home environments, and demographics (including socioeconomic status), exhibited relatively smaller yet consistent contributions, underscoring their distinct roles in predicting math achievement over time. The findings have implications for both future research and educational practices, which are discussed in detail. Educational Impact and Implications Statement: Understanding the predictors of students' academic achievement is one of the foremost concerns in research on education. Most studies analyze the effects of only a handful of predictors at one time. However, in the real world, many factors likely interact and jointly contribute to explaining achievement. We use machine learning methods to model a large number of variables and their interactions to better understand how accurately data collected from school documents, cognitive tests, and self-report questionnaires can predict students' math achievement, above and beyond prior achievement. We also assess how the predictive utility of groups of variables changes over time. The insights produced are useful for understanding what data are most useful to collect when predicting math achievement, as well as when to plan interventions to be maximally effective. [ABSTRACT FROM AUTHOR]