Learning to code or coding to learn? A systematic review

10 studies in a narrative synthesis

School-age learners.

A narrative systematic review of what students learn ALONGSIDE learning to code — mathematical problem solving, critical thinking, social skills, self-management, academic skills. No meta-analysis, no risk-of-bias appraisal, no effect-size synthesis, no moderation by control quality. The full text is paywalled and the abstract is not deposited in Crossref or ERIC, so it was read entirely through secondary characterisations — the weakest read depth in this topic.

Reports conceptual links between coding and mathematical modelling and creative problem solving, and catalogues claimed additional achievements. On the specific mathematics-transfer question the evidence it assembles is MIXED — improved verbal and mathematical problem solving in one study, but no difference in mathematical modelling or procedural comprehension. Its own framing question — learning to code, or coding to learn? — is exactly the near/far distinction, and the review does not resolve it with numbers.

Not applicable — no pooled estimates.

A narrative synthesis; nothing to replicate.