All-inclusive machine learning approach to modeling achievement predictors

Research Method · All research methods
Evidence · 2 claims (2 for) · 1 study (1 associational), q2 · 0 of 1 report an effect size · 2 claims rest on one study

Description

The article coins an "all-inclusive approach": modeling the entire set of potential predictors, including their nonlinear and complex interactions, at once, rather than testing a small subset of predictors. Machine learning makes this feasible because regularization and out-of-sample evaluation protect against overfitting and multicollinearity that destabilize traditional regression with many predictors. The approach is applied to the PALMA longitudinal data with Elastic Net and Random Forest models, using a pipeline of training/test splitting, five-fold cross-validation, and SHAP-based variable importance for interpretation.

Accounts

• All-inclusive machine learning approach to modeling achievement predictors: The article coins an "all-inclusive approach": modeling the entire set of potential predictors, including their nonlinear and complex interactions, at once, rather than testing a small subset of predictors. Machine learning makes this feasible because regularization and out-of-sample evaluation protect against overfitting and multicollinearity that destabilize traditional regression with many predictors. The approach is applied to the PALMA longitudinal data with Elastic Net and Random Forest models, using a pipeline of training/test splitting, five-fold cross-validation, and SHAP-based variable importance for interpretation. (Lavelle-Hill et al. (2024))

Claims

• A machine learning model using 105 predictors predicts next-year standardized math achievement with out-of-sample prediction R-squared of .70–.78 [+M]
• Beyond prior achievement, predictor groups (IQ, grades, motivation and emotion, cognitive strategies, classroom/home environments, demographics) show smaller yet consistent contributions to predicting math achievement over time [+M]

Related Research Methods

-

Key Sources

• Lavelle-Hill, R., Frenzel, A. C., Goetz, T., Lichtenfeld, S., Marsh, H. W., Pekrun, R., Sakaki, M., Smith, G., & Murayama, K. (2024). How the Predictors of Math Achievement Change Over Time: A Longitudinal Machine Learning Approach. Journal of Educational Psychology, 116(8), 1383–1403. https://doi.org/10.1037/edu0000863

Page metadata

Field	Value
type	research-method
id	all-inclusive-machine-learning-approach-to-modeling-achievement-predictors
title	All-inclusive machine learning approach to modeling achievement predictors
description	The article coins an "all-inclusive approach": modeling the entire set of potential predictors, including their nonlinear and complex interactions, at once, rather than testing a small subset of predictors.
status	review
generated	by process:settle-candidates on 2026-10-25
trust	unverified — no one has confirmed this page's content

Frontmatter is the page's machine-readable record — see the Schema & Guide.