Peer Tutoring Improves Achievement¶

Claim · All claims
Evidence · 2 studies · 2 quant-synthesis r? · q3–q4 · i1 small
Currency · newest study 2017 · checked against its sources 2026-10-20 (automated, 2 of 2 entries) · not reviewed by a person

Learners who take part in school peer-tutoring programs, as tutees or as tutors, tend to show higher achievement than comparable learners without tutoring. One meta-analysis (65 evaluations) found gains for tutees and the same pattern of gains for tutors, without a pooled effect size in the abstract read; another (50 mathematics studies) found a small average effect for participants (g = 0.333). Neither recorded entry shows that tutors gain more than tutees.

Subclaims¶

q3 i? A meta-analysis of 65 independent tutoring-program evaluations found tutoring raised tutees' academic performance and attitudes relative to controls, with parallel gains for the students who served as tutors. → Cohen et al. 1982

q4 i1 A meta-analysis of 50 peer-tutoring studies in mathematics found a small-to-moderate average positive effect on academic achievement, moderated by educational stage, study design, program duration, tutor knowledge level, time of day and sample size. → Alegre-Ansuategui et al. 2017

Evidence¶

Cohen et al. 1982¶

Cohen, P. A., Kulik, J. A., & Kulik, C.-L. C. (1982). Educational Outcomes of Tutoring: A Meta-analysis of Findings. American Educational Research Journal, 19(2), 237–248. doi:10.3102/00028312019002237

q3 · meta-analysis · i? · no pooled effect size reported in the abstract · n=65 independent evaluations · quant-synthesis · r?

A meta-analysis of 65 independent evaluations of school tutoring programs, spanning tutee and tutor outcomes. Tutored students outperformed control students on examinations and developed more positive attitudes toward the subject matter; students who served as tutors showed the same pattern of academic and attitudinal gains. Participation had little or no effect on the self-esteem of either tutors or tutees. Only the abstract was read (the full article is paywalled and no open-access copy was found), so no pooled effect-size statistic could be confirmed from the text — the abstract itself states the finding only in directional terms.

Alegre-Ansuategui et al. 2017¶

Alegre-Ansuategui, F. J., Moliner, L., Lorenzo, G., & Maroto, A. (2017). Peer Tutoring and Academic Achievement in Mathematics: A Meta-Analysis. EURASIA Journal of Mathematics, Science and Technology Education, 14(1). doi:10.12973/ejmste/79805

q4 · meta-analysis · i1 · small effect, Hedge's g = 0.333 · n=50 studies · quant-synthesis · r?

A meta-analysis of 50 independent studies of peer-tutoring programs in mathematics across multiple educational stages found that 88% of the programs had a positive effect on participants' academic performance, with a pooled Hedge's g of 0.333. Moderator analyses found tutee/tutor age, assigned roles, tutee ability level (disabled/at-risk vs. not), session length and frequency were not significant moderators, while educational stage, study design, program duration, tutor knowledge level, time of day (in-school vs. out-of-school) and sample size were significant moderators of the effect. Crossref carries no page range for this record; the article's own header gives pp. 337–354 (2018;14(1)), which is noted here but not asserted as the Crossref-verified value.

Studies bearing on this claim¶

Does structured peer tutoring, a paired form of cooperative work, raise achievement, and does its structure matter? This table gathers the studies behind this claim and the 3 narrower claims that bear on it: 4 distinct studies (4 quant-synthesis), q3–q4, 2 of 4 with an effect size coded. One row per study, codes as each entry records them; no effect is averaged and no verdict is drawn. Built by scripts/synthesize_claim.py from eval/synthesis/; a row that looks wrong is fixed on the claim it comes from.

Discussion¶

Tutor gains. Cohen et al. (1982) report that tutors showed the same pattern of academic and attitudinal gains as tutees; the recorded entries do not compare the size of the two, so the claim often made in the peer-tutoring literature that the tutor benefits as much as or more than the tutee is not established here. Preparing to explain, generating explanations, and detecting and repairing one's own knowledge gaps during tutoring are forms of elaborative retrieval and self-explanation, which plausibly explains why tutoring functions as a powerful learning activity for the tutor rather than merely a delivery mechanism. This aligns with the broader "protégé effect" documented for teaching-related activities — see Teaching as learning.

Moderators to establish. Two meta-analyses are recorded, one read only as an abstract; Alegre-Ansuategui et al. (2017) found educational stage, study design, program duration, tutor knowledge level, time of day and sample size moderated the effect in mathematics. Open questions the recorded evidence does not settle include: structured versus unstructured tutoring formats, same-ability versus cross-ability pairing, age and domain of learners, and whether effects persist on delayed or transfer measures rather than only immediate post-tests. Reciprocal tutoring designs, where learners alternate roles, are one common attempt to capture both tutor and tutee benefits.

Boundary conditions. Peer tutoring is not a substitute for expertise: tutors with inaccurate understanding can propagate errors, and benefits depend on tutors having sufficient mastery of the material they explain. Implementation quality — training tutors, providing structured materials, and monitoring sessions — is likely to moderate outcomes substantially. From a cognitive load theory perspective, explaining to a peer can benefit the tutor through germane elaboration, but an overloaded or under-prepared tutor may fail to generate useful explanations at all.

Interpreting against the right baseline. Because tutoring is an active, interactive format, any comparison against passive instruction conflates tutoring-specific effects with the general advantage of active learning. Well-designed studies compare tutoring against other active formats, not only lecture.

Role assignment and structure. Because tutor and tutee roles engage different mechanisms — elaborative explanation for the tutor, guided practice with feedback for the tutee — designs that fix one learner in each role may systematically favor one party. Reciprocal or role-swapping designs, and structured protocols that script the explanation and questioning moves, are the main implementation responses; neither is tested by the evidence recorded here.

Merged from “Tutoring benefits both tutors and tutees” (tutoring-benefits-tutors-and-tutees): Why tutors benefit. The dominant explanation is that preparing to teach forces the tutor to organize, elaborate, and verbalize their knowledge, exposing gaps that passive review leaves hidden. This aligns with the generative-processing logic behind Active learning improves exam performance and with Constructivism: knowledge is consolidated through use rather than reception. Explaining to another person also adds social accountability, which Social learning theory treats as a powerful motivator for effortful engagement.

Why tutees benefit. Tutees receive individualized pacing, immediate feedback, and opportunities to ask questions that whole-class settings rarely permit. One-to-one and small-group interaction also supports the social presence conditions described in Community of inquiry.

Moderators and boundary conditions. Tutor benefits are largest when tutors are adequately prepared and the material is within reach of their current competence; asking a near-peer to tutor content they have not yet mastered can entrench errors for both parties. Structured protocols (prompt cards, scripted question sequences) tend to outperform unstructured "just help each other" arrangements, consistent with the scaffolding logic of Cognitive load theory — unstructured tutoring can impose extraneous load on both roles. Reciprocal designs, where partners alternate roles, distribute these benefits and avoid fixed status hierarchies.

Design implications. Effective programs typically (a) train tutors in specific questioning and explanation routines before sessions begin, (b) provide materials that structure the interaction rather than leaving roles open-ended, and (c) build in role rotation so neither partner is permanently cast as the weaker learner. These design features echo the scaffolding-and-fading approach described in Cognitive apprenticeship, where responsibility is gradually transferred to the learner.

Open questions. The relative magnitude of tutor versus tutee gains across age bands and domains, and the durability of tutor-side gains over time, still need systematic synthesis on this page. The recorded meta-analysis of tutoring programmes reports gains for tutors but no effect size for them.

Related Claims¶

• Collaborative learning improves outcomes — peer tutoring is a structured form of collaborative learning with assigned roles.
• Teaching as learning — the "protégé effect": preparing to teach improves the teacher's own learning.
• Active learning improves exam performance — tutoring is an active-learning format; effects should be interpreted against that baseline.
• Cognitive load theory — explains why explaining to a peer can benefit the tutor through germane elaboration, and why overloaded tutors may fail.
• Intelligent tutoring systems improve achievement over non-individualised instruction — related
• Engelmann's Direct Instruction curricula improve academic outcomes across reading, mathematics, language and spelling — a broader claim this one bears on
• Learning By Teaching Improves Tutor Learning — related
• Mastery Learning Improves Outcomes — related
• The quality of feedback students give peers predicts gains in their own work, but no recorded study isolates a causal benefit of the assessor role — related
• A descriptive report defines peer assisted learning and asserts, without a tested result, that it promotes knowledge construction, collaboration ability and interpersonal interaction — a broader claim this one bears on
• Structured Peer Tutoring Outperforms Unstructured — related
• Peer-assisted learning interventions improve achievement for elementary students — related
• Cross-age peer tutoring by either bilingual or English-only tutors shows no statistically significant effect on oral language proficiency (Serrano, 1987) — a narrower finding that bears on this claim
• GO Tutor Corps high-impact math tutoring produced significantly greater mathematics achievement gains than no tutoring for matched grade 6-10 students — a narrower finding that bears on this claim
• Math tutoring programs may improve students' math confidence and sense of belonging — related
• Meta-analytic studies show moderate to large effect sizes for collaborative learning on both achievement and attitudes — related
• High-quality tutoring produces large learning gains cost-effectively, in person and virtually — a broader claim this one bears on
• Tennessee's education department notes tutoring during the school day is more effective than before- or after-school programming — a narrower finding that bears on this claim
• Summary research table reports tutoring impacts of roughly 90 to 222 days of additional learning across five studies — related
• A recent meta-analysis of 14 types of educational interventions found high-impact tutoring the most effective at improving student achievement — related
• Interrupted time series analysis found tutor quality improved gradually over time (β=0.01, p=.022) with no immediate level change or slope change at the training intervention — related
• Scenario-based tutor lessons produced a significant pooled 7.4% learning gain, driven by three lessons with established construct validity while three newer lessons showed no significant gains — a narrower finding that bears on this claim
• Social learning theory — social accountability and modeling explain tutor-side motivation gains
• Constructivism — knowledge consolidation through explanation and use
• Cognitive apprenticeship — modeling, coaching, and fading map onto the tutor's role progression
• The different elements of the Engagement Framework are interrelated in practice, with gains in one dimension reinforcing others — related
• Sociocultural learning theory holds that knowledge develops through engagement in social interaction with others, so English learners benefit from meaningful interaction with peers — related
• Tutoring is highly effective, and hiring older peers as tutors supports learning gains for both older and younger students — a narrower finding that bears on this claim
• Tutoring programs consistently outperform common school-based alternatives such as class-size reduction and summer school, per the review's synthesis of Kraft — a narrower finding that bears on this claim
• The article advances a five-part rationale for benefits of adding a peer element to extensive reading — related
• The review reports that college tutors who espoused a malleable view of intelligence to their tutees improved their own academic performance, with benefits for tutees as well — related
• Cross-age tutoring produces a small-to-moderate positive effect on academic outcomes for both tutors and tutees — a narrower finding that bears on this claim
• Cross-age tutoring effects on tutees do not significantly differ by number of sessions, tutor type, tutee risk type, or subject area — a narrower finding that bears on this claim