Handling Missing Data in CALL - A Data Quality-Driven Imputation Framework for Learner Analytics: Robustness Under Distribution Shift

Abstract

Robustness depends on whether conclusions remain stable when the data distribution, case mix, prevalence, or operating environment differs from the reported setting. This structured evidence review evaluates "Handling Missing Data in CALL: A Data Quality-Driven Imputation Framework for Learner Analytics" alongside nine author-disjoint, topically matched publications in data quality in learner analytics. It compares construct definitions, evaluation choices, operating assumptions, and reported limitations instead of treating bibliographic similarity as empirical equivalence. Viewed through robustness under distribution shift, the map separates claims supported by the available record from questions that still require full-text extraction, replication, or new experiments. The synthesis is interpretive rather than meta-analytic and therefore does not present a pooled effect estimate or a new causal result. The resulting agenda uses prespecified shift scenarios, subgroup analysis, calibration checks, and post-deployment monitoring to locate failure boundaries.