CoChat

The DOI matches its Crossref record; 67 of the 71 references it cites were found in a registry or an index, and 4 are not checked yet; none of the 9 sources its claims cite is checked for support yet.

How effective is second language incidental vocabulary learning? A meta-analysis

Open access
Webb, Stuart, Uchihara, Takumi, Yanagisawa, Akifumi · 2023 · Language Teaching

Abstract

There is a great deal of variation in gains found between studies of second language (L2) incidental vocabulary learning, as well as many factors that affect learning. This meta-analysis investigated the effects of exposure to L2 meaning-focused input on incidental vocabulary learning with an aim to clarify the proportional gains that occur through meaning-focused learning. Twenty-four primary studies were retrieved providing 29 different effect sizes and a total sample size of 2,771 participants (1,517 in experimental groups vs. 1,254 in control groups). Results showed large overall effects for incidental vocabulary learning on first and follow-up posttests. Mean proportions of target words learned ranged from 9–18% on immediate posttests, and 6–17% on delayed posttests. Incidental L2 vocabulary learning gains were similar across reading (17%, 15%), listening (15%, 13%), and reading while listening (13%, 17%) conditions on immediate and delayed posttest. In contrast, the proportion of words learned in viewing conditions on immediate posttests was smaller (7%, 5%). Findings also revealed that the amount of incidental learning varies according to a range of moderator variables including learner characteristics (L2 proficiency, institutional levels), materials (text type and audience), learning activities (spacing, mode of input), and methodological features (approaches to controlling prior word knowledge).

Checked 2026-11-19 · 4 references not checked yet

Overview

CoChat Summary

TL;DR

This meta-analysis of 24 primary studies (29 effect sizes; N = 2,771) found large overall effects for L2 incidental vocabulary learning from meaning-focused input on immediate (g = 1.14) and delayed (g = 0.93) posttests, but mean proportions of target words learned were modest (9–18% immediate, 6–17% delayed). Gains varied by mode of input and moderator variables, with viewing generally showing smaller gains than reading, listening, or reading while listening.

Key findings

  • Overall incidental learning effects were large: immediate posttest g = 1.14, 95% CI [0.86, 1.41], p < .001; follow-up posttest g = 0.93, 95% CI [0.44, 1.42], p < .001. Mean proportions of target words learned ranged from 9–18% on immediate posttests and 6–17% on delayed posttests.
  • By mode of input, immediate learning rates were reading 17%, listening 15%, reading while listening 13%, and viewing 7%; delayed rates were reading 15%, listening 13%, reading while listening 17%, and viewing 5%. Immediate effect sizes were reading g = 1.45, listening g = 0.97, reading while listening g = 0.78, and viewing g = 0.60; delayed effect sizes were reading g = 0.94, listening g = 0.76, reading while listening g = 1.07, and viewing g = 0.43.
  • Significant immediate-posttest moderators included text type (narrative g = 1.43 vs. expository g = 0.61), text audience (L2 learners g = 1.56 vs. L1 users g = 0.71), mode of input (Q = 9.50, p = .023), and pre-knowledge control (pseudowords g = 1.90, pretests g = 1.01, pilot testing g = 0.80, no pretest g = 0.27). L2 proficiency approached significance (beyond basic g = 1.40 vs. basic g = 0.70, p = .051), and spacing approached significance (spaced g = 1.51 vs. massed g = 0.97, p = .080). Test format did not differ significantly: form recognition g = 1.04, meaning recognition g = 1.10, meaning recall g = 1.08.
  • On delayed posttests, spacing became significant (spaced g = 1.71 vs. massed g = 0.58, Q = 6.22, p = .013), text audience remained significant (L2 learners g = 1.34 vs. L1 users g = 0.44, Q = 4.00, p = .045), and mode of input was non-significant (Q = 0.80, p = .849).

Methods

  • Meta-analysis of 24 primary studies, with 29 independent effect sizes for the first posttest and a total sample of 2,771 participants (1,517 experimental vs. 1,254 control); the follow-up posttest included 9 effect sizes and 741 participants (416 treatment vs. 325 control), with an average retention interval of 34.1 days. One outlier study (Vidal, 2011) was excluded before analysis, leaving k = 28 for the overall first-posttest effect.
  • Searches covered databases and 19 journals up to October 2022; inclusion required incidental learning from meaning-focused input, no explicit instruction/glossing/captions, a between-participants control condition, and sufficient effect-size data. Effect sizes used Hedges’ g; a random-effects model was used for overall effects and a mixed-effects model for moderator analyses.
  • Eight moderator variables were coded: L2 proficiency, institutional level, text type, text audience, spacing, mode of input, pre-knowledge control, and test format. Coding reliability was 98%; learning rates were calculated as experimental-minus-control gain scores divided by total target words.

Limitations

  • Stated in the paper: controlled learning conditions may not reflect typical incidental learning inside or outside the classroom, where learners may consult dictionaries, teachers, classmates, or parents, so results may be lower than in less controlled conditions. Target items are typically low-frequency words encountered more often than other low-frequency words, while non-target and higher-frequency unknown or partially known words may also be learned. The meta-analysis examined form recognition and form-meaning connections, not other aspects such as spoken form, collocation, word parts, or association.
  • Stated in the paper: few studies included delayed posttests (k = 9) compared with immediate posttests (k = 29), so delayed moderator analyses should be interpreted cautiously; practice effects may occur from immediate to delayed posttests, and counterbalancing or delayed-only designs are recommended. There are also fewer studies of listening, reading while listening, viewing, and younger learners (primary k = 5, secondary k = 6, university k = 17).
  • Evident from the text: several delayed moderator categories rely on very small numbers of studies (e.g., listening k = 2 or 3, reading while listening k = 2, viewing k = 2; pseudoword control removed with k = 1), making those estimates unstable. Significant heterogeneity was also present for both immediate (Q = 269.08, p < .001) and delayed (Q = 55.85, p < .001) posttests, indicating substantial variation across studies.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1017/s0261444822000507

Cites

67 of 71 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

0 Valid · 3 Not valid · 9 Not checked · 79 Own results and interpretations

35 more claims

Create a free account to see all 91 claims and to verify the 8 references still open.

See all claims — free account
  • It may state two separate claims as one. 33
  • Probably not a substantial claim. 20

Create a free account to see these claims.

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

5 of 7 · the rest aren’t loaded

2 more findings

Create a free account to see every finding of this check.

See all findings — free account

Citations

Cites

Show all 67 references

Research works printed but not linked: 1

  1. Borenstein, 2006, Comprehensive meta-analysis (version 3.3.070) [computer software]

Cited by

See all 200

Metadata

Authors
Webb, S., Uchihara, T., & Yanagisawa, A.
Affiliations
  • Western University (CA)
  • Waseda University (JP)
  • University of Tsukuba (JP)
Corresponding author
Webb, Stuart
Venue
Language Teaching
Published
2023-01-13
Volume
56
Issue
2
Pages
161-180
ISSN
  • 0261-4448 (ISSN-L)
  • 1475-3049
OpenAlex
W4316038657
Cited by
200
References
67
Citations by year
2026: 100 · 2025: 55 · 2024: 35 · 2023: 9
FWCI
5.67
Type
journal article
Language
en
Open access
Locations
Language Teaching (publishedVersion, cc-by)
Topics
  • Social Sciences › Psychology › Developmental and Educational Psychology › Second Language Acquisition and Learning
  • Social Sciences › Arts and Humanities › Language and Linguistics › EFL/ESL Teaching and Learning
  • Social Sciences › Psychology › Developmental and Educational Psychology › Reading and Literacy Development
Keywords
incidental vocabulary learning, second language vocabulary acquisition, reading, listening, viewing, spacing effect, second language proficiency, text types

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

A language model drew the claims from the paper on 2026-11-19, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-11-19 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: How effective is second language incidental vocabulary learning? A meta-analysis. https://app.cochat.ai/papers/10.1017/s0261444822000507
The paper
Webb, S., Uchihara, T., & Yanagisawa, A. (2023). How effective is second language incidental vocabulary learning? A meta-analysis. Language Teaching. https://doi.org/10.1017/s0261444822000507

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.