Corrective feedback on second-language errors has a significant effect on L2 development that is maintained on delayed tests¶

Claim · All claims
Evidence · 2 studies · 2 quant-synthesis r2 · q3 · i2 medium
Currency · newest study 2010 · checked against its sources 2026-10-20 (automated, 2 of 2 entries) · not reviewed by a person

Corrective feedback here is a teacher's or interlocutor's response to a learner's error in the target language (recasts, prompts, explicit correction and similar moves). The claim is about feedback as a whole; which type works best is a separate question (see Related Claims).

Subclaims¶

q3 i2 Across 15 classroom quasi-experimental studies (N = 827), oral corrective feedback had a medium effect against control groups (d = 0.74), with a between-group mean of 0.63 at immediate and 0.84 at delayed posttests, and no significant decline between them. → Lyster & Saito 2010 q3 i? Across 33 studies, corrective feedback had an overall effect the author describes as medium that was maintained over time, larger in laboratory than classroom studies and in foreign- than second-language contexts; the abstract prints no effect size. → Li 2010

Evidence¶

Lyster & Saito 2010¶

Lyster, R., & Saito, K. (2010). Oral feedback in classroom SLA: A meta-analysis. Studies in Second Language Acquisition, 32(2), 265–302. doi:10.1017/S0272263109990520

q3 · meta-analysis of classroom quasi-experimental studies · i2 · medium effect, d=0.74 (between-group) · n=15 studies (N=827) · quant-synthesis · r2

Meta-analysis restricted to 15 classroom-based quasi-experimental studies of oral corrective feedback in teacher–student interaction (N = 827), chosen because laboratory results were not taken to predict classroom effectiveness. Against control groups, corrective feedback had d = 0.74; against learners' own pretests it had d = 0.91, though control groups also gained (d = 0.39). Between-group contrasts gave a mean of 0.63 at immediate posttests and 0.84 at delayed posttests, and the overlapping confidence intervals indicate no significant decline. Effects were larger for prompts than recasts, most apparent on free constructed-response measures (d = 0.97 between groups, significantly larger than metalinguistic judgments, d = 0.45, though not distinguishable from constrained constructed-response measures, d = 0.70), and not different between second- and foreign-language classrooms; long treatments (d = 1.13) beat short-to-medium ones (d = 0.57), and a regression on age found younger learners benefiting more than older ones, and adults (mean age above 23; 140 of the 827 learners) were a minority of the pooled learners; the authors caution that few child and adult studies were available. Full text read.

"Both immediate and delayed effects of CF thus exhibit medium-to-large effect sizes"

Li 2010¶

Li, S. (2010). The effectiveness of corrective feedback in SLA: A meta-analysis. Language Learning, 60(2), 309–365. doi:10.1111/j.1467-9922.2010.00561.x

q3 · meta-analysis (abstract only) · i? · n=33 studies · quant-synthesis · r?

Meta-analysis of 33 primary studies of corrective feedback in second language acquisition (22 published studies and 11 Ph.D. dissertations), coded for 17 substantive and methodological features. It found a medium overall effect for corrective feedback that was maintained over time; implicit feedback's effect was better maintained than explicit feedback's; laboratory studies showed larger effects than classroom studies, shorter treatments larger effects than longer ones, and foreign-language contexts larger effects than second-language ones; published studies did not show larger effects than dissertations. Only the abstract was read: it prints no effect size, confidence interval or participant count, so "medium" is the author's word and is not coded here.

"there was a medium overall effect for corrective feedback and the effect was maintained over time"

Studies bearing on this claim¶

Does corrective feedback on second-language errors improve accuracy, and which form of correction (recast, prompt, metalinguistic clue, explicit correction with or without a rule) works for which target and learner? This table gathers the studies behind this claim and the 21 narrower claims that bear on it: 12 distinct studies (7 causal, 3 review, 2 quant-synthesis), q2–q3, 4 of 12 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¶

Both syntheses find the effect of corrective feedback durable, and they overlap in time and topic, so some primary studies are probably in both; the abstract of Li (2010) does not list its studies, so the overlap could not be checked. They disagree on treatment length: Li (2010) reports larger effects for shorter treatments, while Lyster and Saito (2010), in classrooms only, found long treatments larger than short-to-medium ones and not distinguishable from brief ones. On setting they also differ: Li found larger effects in foreign-language contexts, Lyster and Saito found setting not to be a contributing factor. Lyster and Saito's age result (younger learners benefit more) is a caution for designs aimed at adults, who were a minority of their pooled learners.

Related Claims¶

• The literature is supportive of recasts, but recast effectiveness varies with the language area and learner knowledge, focus, and aptitudes — a narrower claim about one feedback type
• Metalinguistic feedback's gains on wh-question performance decline over a ten-day delay while recast gains hold, leaving the two no different — a narrower finding that bears on this claim
• Metalinguistic feedback outperforms recasts on the immediate post-test, but the two do not differ significantly on the delayed post-test — a narrower finding that bears on this claim
• Corrective feedback treatment and time interact significantly in shaping wh-question learning gains across the three testing periods — related
• Feedback in CALL has a significant moderate positive effect on student language learning outcomes (RE g = 0.56) across 21 studies — related
• Corrective feedback facilitates L2 grammatical development, with explicit feedback stronger immediately and implicit feedback better maintained over time — related
• Prompts raised learners' pragmatic awareness of refusals more than explicit feedback or control after a 10-week treatment — related
• All four written CF types, including metalinguistic code CF, outperformed the control group on the delayed posttest of explicit L2 knowledge two weeks later. — a narrower finding that bears on this claim
• On the immediate posttest of explicit L2 knowledge, direct CF and metalinguistic CF with rule explanations outperformed the control group while metalinguistic code CF did not. — related
• All four written CF types, including metalinguistic code CF, outperformed the control group on the delayed posttest of implicit L2 knowledge. — a narrower finding that bears on this claim
• Generative AI-assisted feedback is associated with a statistically significant pretest–posttest improvement in violin students' intonation performance over a four-week period — related
• Feedback channel is a non-significant moderator of feedback effects — related
• High-information feedback has a very large effect (d = 0.99) while reinforcement/punishment has a low effect (d = 0.24) — related
• Providing corrective feedback on practice tests does not significantly change the testing effect (g = 0.63 with feedback vs. g = 0.60 without) — related
• CF effectiveness is mediated by multiple variables, including linguistic targets and learners' age — a narrower finding that bears on this claim
• The frequency of oral corrective feedback differs across instructional contexts — related
• Laboratory and classroom studies assess the effects of recasts and different CF types differentially — related
• Interaction in both SCMC and FTF contexts has a significant impact on L2 development (meta-analysis of 14 studies, 1990–2012) — related
• Theoretical perspectives support corrective feedback while contentious issues remain about uptake, instruction, and CF's purpose — related
• All three groups benefited from the instructional intervention, with both experimental groups benefiting the most — a narrower finding that bears on this claim
• Explicit correction did not produce evidence of learned morphological generalizations to novel stimuli in adult L2 learners — a narrower finding that bears on this claim
• Explicit correction improves adult L2 learners' responses on corrected suffixation items during the feedback session — a narrower finding that bears on this claim
• Explicit instruction facilitates the acquisition of L2 grammar — related
• Form-focused instruction on noun endings significantly increases French immersion students' ability to correctly assign grammatical gender, with gains maintained on delayed posttests — related
• Form-focused instruction combined with prompts is more effective than FFI combined with recasts or no feedback, especially on written tasks — a narrower finding that bears on this claim
• Explicit metalinguistic feedback outperforms recasts on the delayed oral imitation posttest measuring implicit knowledge — a narrower finding that bears on this claim
• Metalinguistic feedback produced system learning, generalizing to verbs not included in the treatment — a narrower finding that bears on this claim
• A controlled experiment tested whether various types of negative feedback affect adult second-language learners' acquisition of abstract constraints on an overgeneralized rule (the English dative alternation). — a narrower finding that bears on this claim
• Prompts were overall more effective than recasts as corrective feedback for L2 grammar learning — a narrower finding that bears on this claim
• Study design and post-test timing show only borderline moderating effects in between-subject analyses, not significant in within-subject analyses — related
• Metalinguistic written CF was equivalent, not superior, to non-metalinguistic direct CF, contrasting with previous studies whose metalinguistic CF also contained correct forms. — a narrower finding that bears on this claim
• Explicit teaching is more effective for EFL copular be learning but its effect is not durable — related
• The study compared feedback-uptake relationships across two immersion settings totaling 18.3 hr of French and 14.8 hr of Japanese classroom interaction — related
• On the immediate posttest of implicit L2 knowledge, direct CF and metalinguistic CF with rule explanations outperformed the control group while metalinguistic code CF did not. — a narrower finding that bears on this claim
• Low-proficiency L2 learners benefited significantly more from prompts than from recasts — a narrower finding that bears on this claim
• Feature-focused information-literacy trainings appear effective, but lack delayed-posttest evidence — related
• Immediate trial-by-trial feedback was more helpful than delayed feedback for children ages 8-11, but timing conclusions remain weak overall — related
• Posttests were coded as delayed if they occurred at least one day after treatment, with two-delay studies and studies lacking day/week delay metrics excluded from mean and median delay calculations — related