Computer-mediated and face-to-face peer feedback: student feedback and revision in EFL writing
Open accessThe DOI matches its Crossref record; 62 of the 62 references it cites were found in a registry or an index; none of the 44 sources its claims cite is checked for support yet.
Abstract
This study investigates how two feedback forms and sequences influence peer feedback and revision. The two feedback forms included: written asynchronous computer-mediated communication (hereafter WACMC) in Google Docs and traditional oral face-to-face interaction (hereafter OF2F). These two forms were used in two sequences: WACMC followed by OF2F (WACMC–OF2F) and OF2F followed by WACMC (OF2F–WACMC). Participants were twenty-six Vietnamese EFL students. After feedback training, students completed four writing tasks, provided feedback in pairs, and finally revised their work. This study employed a mixed-method approach. Written and oral feedback, student revision, as well as student interviews were collected, transcribed, coded, and analysed. Paired-samples t-tests and ANOVA tests were employed to compare student feedback and revision in the two forms and sequences. The findings showed that: (1) more revision-oriented comments were associated with the WACMC form and with the WACMC–OF2F sequence; (2) written comments had a higher uptake rate in both sequences; (3) global revisions were more often found in the WACMC–OF2F sequence; (4) while students perceived WACMC feedback in Google Docs as more helpful than OF2F feedback, most suggested using both feedback forms. Pedagogical implications for the L2 writing class will also be included.
Overview
CoChat Summary
TL;DR
This mixed-methods study examined how two peer feedback forms (written asynchronous computer-mediated communication in Google Docs vs. oral face-to-face) and two sequences of those forms (WACMC–OF2F vs. OF2F–WACMC) shaped the comments 26 Vietnamese EFL students gave and the revisions they made. WACMC produced more local revision-oriented comments and had more influence on revision, while students generally preferred WACMC and the WACMC–OF2F sequence but wanted both forms used together.
Key findings
- WACMC feedback yielded significantly more local revision-oriented (LC+) comments than OF2F (M = 1.14, SD = 0.7 vs. M = 0.53, SD = 0.36; t(25) = 5.56, p = 0.00001), and LC+ was the most frequent category overall in both forms.
- Discourse functions differed by form: suggestion (M = 1.22 vs. 0.57, t(25) = 4.46, p = 0.0001), praise (M = 1.19 vs. 1.01, t(25) = 3.09, p = 0.004), and confirmation of understanding (M = 0.24 vs. 0.36, t(25) = 5.29, p = 0.000) were statistically significant after Bonferroni correction.
- Between sequences, only global non-revision-oriented comments (M = 2.02 vs. 2.41, t(25) = 3.48, p = 0.001) and confirmation of understanding (M = 0.45 vs. 0.76, t(25) = 4.95, p = 0.000) remained significant, both higher in OF2F–WACMC.
- Revision: one-way ANOVA showed a significant effect of source (F(3, 100) = 19.60, p < 0.0000); WACMC (M = 0.57, SD = 0.53) differed significantly from OF2F (M = 0.05, SD = 0.07), and self-initiated revisions (M = 0.99, SD = 0.63) exceeded all other sources, accounting for almost half (48%) of all revisions. Local revisions outnumbered global ones (1.74 vs. 0.38 per 100 words).
- Perceptions: 21/26 students preferred WACMC, 20/26 found it less face-threatening, and 18/26 (two-thirds) preferred the WACMC–OF2F sequence; all students said both forms should be used together.
Methods
- Design: mixed-method study with 26 Vietnamese EFL students (6 male, 20 female, aged 18–20, CEFR B2) at a tertiary institution in Ho Chi Minh City, all training to teach English for Specific Purposes. Students completed four writing tasks (cause-effect, problem-solution, compare-contrast, argumentative) over a 15-week semester, with essays 1 and 3 following WACMC–OF2F and essays 2 and 4 following OF2F–WACMC; two weeks of peer feedback training preceded data collection.
- Data: 104 first drafts and 104 discussions (26 students × 4 essays), student revisions recorded via MS Word's Compare function, and semi-structured interviews (~30 minutes each) with all students; 208 papers total (26 × 4 assignments × 2 drafts).
- Analysis: feedback coded using Liu and Sadler's (2003) scheme for areas (global/local), nature (revision-oriented/non-revision-oriented), and discourse functions; revisions coded for origin and area using an adapted Hall (1990) framework. Paired-samples t-tests and one-way ANOVA in SPSS v.23 with Bonferroni correction; comments and revisions standardised per 100 words; thematic analysis (Braun & Clarke, 2006) for interviews. Inter-coder reliability: κ = 1 and 0.95 for areas and nature of feedback, κ = .75–.98 for discourse functions, κ = .99 for revisions.
Limitations
- Stated by the paper: the exploratory approach "could not provide an absolute answer to the effects of the two feedback forms and sequences"; the author calls for future experimental or quasi-experimental research with a counterbalanced design, and for a more sophisticated analysis scheme to pinpoint exact aspects of contribution (e.g. content, idea development, grammar, vocabulary).
- Stated by the paper: the study attended more to feedback training than to revision training, and the author notes the scenario "might be different if more training on revision had been done."
- Evident from the text: the sample is small (26 students) and drawn from a single institution and proficiency level (B2), and each essay went through only two drafts, which the author notes may explain differences from Chang's (2012) findings; the author also suggests the short (~250-word) essays may have limited the relevance of the purpose-confirmation step, and recommends future research with longer texts.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
0 Valid · 4 Not valid · 28 Not checked · 124 Own results and interpretations
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“while students perceived WACMC feedback in Google Docs as more helpful than OF2F feedback”
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“written comments had a higher uptake rate in both sequences”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“more revision-oriented comments were associated with the WACMC form and with the WACMC–OF2F sequence”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“more revision-oriented comments were associated with the WACMC form and with the WACMC–OF2F sequence”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
A result the paper reports from its own data — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“global revisions were more often found in the WACMC–OF2F sequence”
plainstates an association- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
78 more claims
Create a free account to see all 156 claims and every source verdict.
- It may state two separate claims as one. 33
- Probably not a substantial claim. 50
- The quote only partly states this claim. 2
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“This finding diverges from Pham's (2022) study, which reported that self-initiated revisions constituted nearly half of all revisions.”
From Enhancing EFL Writing Revision Practices: The Impact of AI- and Teacher-Generated Feedback and Their Sequences (2025) · cites this paper as Pham's (2022)
In the citing paper
“Since the essays varied in length, revisions were standardized by calculating revisions per 100 words, following the approach taken in prior research (Pham, 2022).”
From Enhancing EFL Writing Revision Practices: The Impact of AI- and Teacher-Generated Feedback and Their Sequences (2025) · cites this paper as Pham, 2022
Statistics that don’t add up
Each reported p-value recomputed from its test statistic and degrees of freedom.
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
“t(25) = 3.09, p = 0.004”
Reported p = .004 · recomputed p = .005
Rated minor: it would not touch a result — a value no result rests on, a p-value that stays on the same side of the significance level, or a conflict that is clerical, resolved elsewhere in the paper, or outside its own findings.
In the paper
“t(25) = 3.48, p = 0.001”
Reported p = .001 · recomputed p = .002
Statistics recomputed
Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.
- Doesn’t add upt(25) = 3.09reported p = .004 · recomputed p = .005
“t(25) = 3.09, p = .004”
- Doesn’t add upt(25) = 3.48reported p = .001 · recomputed p = .002
“t(25) = 3.48, p = .001”
- Consistentt(25) = 5.56reported p = .00001 · recomputed p < .001
“t(25) = 5.56, p = .00001”
- Consistentt(25) = 4.46reported p = .0001 · recomputed p < .001
“t(25) = 4.46, p = .0001”
- Consistentt(25) = 5.29reported p = 0 · recomputed p < .001
“t(25) = 5.29, p = .000”
2 more recomputed tests
Create a free account to see every test the Stats check recomputed.
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.
Rated major: it would touch a result — a value a result rests on, reported and recomputed p-values on opposite sides of the significance level, two passages that conflict with no sign which is right, or a cited work retracted before this paper was published.
In the paper
§ 6. Conclusion
“These comments are also incorporated at a higher rate in revision, which eventually leads to a larger amount of revision in local areas in the WACMC–OF2F sequence.”
§ 4. Results
“more local revisions were made in the OF2F–WACMC sequence despite fewer local revision-oriented comments being given in this sequence than in the WACMC–OF2F sequence”
Citations
Cites
Using thematic analysis in psychology
Braun, V., & Clarke, V. (2006) · Qualitative Research in Psychology · doi:10.1191/1478088706qp063oa · Cited by 194,414
(2005) · PubMed · Cited by 7,011
(2008) · Cited by 6,727
Verification strategies for establishing reliability and validity in qualitative research
Morse, J. M., Barrett, M., Mayan, M., Olson, K., & Spiers, J. (2002) · International Journal of Qualitative Methods · doi:10.1177/160940690200100202 · Cited by 4,978 · cited 1×
“Member checking techniques were applied to ensure the credibility of the transcripts (Morse, Barrett, Mayan, Olson, & Spiers, 2002).”
MethodInterviews – Learning the Craft of Qualitative Research Interviewing
Vaivio, J. (2012) · European Accounting Review · doi:10.1080/09638180.2012.675165 · Cited by 4,535 · cited 11×
“Student opinion is one the most commonly investigated topics, and there has been abundant evidence showing that L2 students preferred CMC to F2F peer feedback (Ciftci & Kocoglu, 2012; Ebadi & Rahimi, 2017; Liu & Sadler, 2003; Pham & Usaha, 2009).”
BackgroundMethodInfluentialLanguage Learning with Technology: Perspectives from Asia
(2021) · Springer eBooks · doi:10.1007/978-981-16-2697-5 · Cited by 2,206
Comparing Face-To-Face and Electronic Discussion in the Second Language Classroom
Warschauer, M. (1995) · CALICO Journal · doi:10.1558/cj.v13i2-3.7-26 · Cited by 955 · cited 2×
“This body of research takes a combine orientation (Chang, 2009, 2012; Ho, 2012, 2015; Ho & Savignon, 2013; Tuzi, 2004; Warschauer, 1996).”
BackgroundWritten Corrective Feedback in Second Language Acquisition and Writing
Bitchener, J., & Ferris, D. R. (2012) · Routledge eBooks · doi:10.4324/9780203832400 · Cited by 711 · cited 1×
“This result corroborates a number of studies which show that EFL/ESL students tend to make changes at local level (Bitchener & Ferris, 2012; Chang, 2012; Jiang & Wang, 2013; Liu & Sadler, 2003; Nelson & Carson, 1998; Tsui & Ng, 2000).”
ResultDo secondary L2 writers benefit from peer comments?
Tsui, A. B. M., & Ng, M. (2000) · Journal of Second Language Writing · doi:10.1016/s1060-3743(00)00022-9 · Cited by 648 · cited 2×
“This result corroborates a number of studies which show that EFL/ESL students tend to make changes at local level (Bitchener & Ferris, 2012; Chang, 2012; Jiang & Wang, 2013; Liu & Sadler, 2003; Nelson & Carson, 1998; Tsui & Ng, 2000).”
BackgroundResultThe influence of teacher commentary on student revision
Ferris, D. R. (1997) · TESOL Quarterly · doi:10.2307/3588049 · Cited by 618 · cited 1×
“Earlier studies report that peer feedback does not help student revision, tends to yield superficial comments (e.g. praise), and is less valued in comparison to teacher feedback (Carson & Nelson, 1996; Ferris, 1997; Zhang, 1995).”
Result
Research works printed but not linked: 2
- Moloudi, M. (2011). Online and face to face peer review: Measures of implementation in ESL writing classes. Asian EFL Journal, 52, 4–22.
- Nemati M. (1999). The effect of mode of discourse on EFL writing performance (Unpublished doctoral dissertation). UK: University of Leicester.
Cited by
Liu, C., Liu, S., Hwang, G., Tu, Y., Wang, Y., & Wang, N. (2023) · Education and Information Technologies · doi:10.1007/s10639-023-11697-6 · Cited by 113
“As technology has evolved, traditional face-to-face feedback has been replaced by computer-assisted electronic feedback, which in the context of EFL writing instruction refers primarily to automated feedback provided by computers (Pham, 2020).”
BackgroundTran, T. T. T. (2025) · Education Sciences · doi:10.3390/educsci15020232 · Cited by 68
“Since the essays varied in length, revisions were standardized by calculating revisions per 100 words, following the approach taken in prior research (Pham, 2022).”
Awada, G. M., & Diab, N. M. (2021) · Computer Assisted Language Learning · doi:10.1080/09588221.2021.1912104 · Cited by 56
“CALL also supports learning by allowing students to apply learning strategies asynchronously at their own pace and in comfortable non-threatening surroundings (Pham, 2021).”
BackgroundInfluentialLi, Y. (2023) · Frontiers in Psychology · doi:10.3389/fpsyg.2023.1165221 · Cited by 104
“The technology-enhanced writing environment may provide opportunities for real-time collaboration, revision, and feedback, which can positively impact writing performance (Pham, 2020; Barrot, 2021).”
BackgroundA review of the ESL/EFL learners’ gains from online peer feedback on English writing
Cao, S., Zhou, S., Luo, Y., Wang, T., Zhou, T., & Xu, Y. (2022) · Frontiers in Psychology · doi:10.3389/fpsyg.2022.1035803 · Cited by 76
“, in terms of questionnaires and interviews) (Pham, 2020).”
BackgroundMethodInfluential
Metadata
- Authors
- Pham, H. T. P.
- Affiliations
- FPT University (VN)
- Corresponding author
- Pham, Ha Thi Phuong
- ORCID
- Pham, Ha Thi Phuong
- Venue
- Computer Assisted Language Learning
- Publisher
- Informa UK Limited
- Published
- 2021-01-15
- Volume
- 35
- Issue
- 9
- Pages
- 2112-2147
- ISSN
- 0958-8221 (ISSN-L)
- 1744-3210
- OpenAlex
- W3122375756
- Cited by
- 51
- References
- 62
- Citations by year
- 2026: 5 · 2025: 10 · 2024: 14 · 2023: 9 · 2022: 12 · 2021: 1
- FWCI
- 35.55
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- Locations
- Computer Assisted Language Learning (publishedVersion, cc-by-nc-nd)
- Topics
- Social Sciences › Arts and Humanities › Language and Linguistics › EFL/ESL Teaching and Learning
- Social Sciences › Education › Student Assessment and Feedback
- Social Sciences › Education › English Language Learning and Teaching
- Keywords
- peer feedback, computer-mediated communication, face-to-face interaction, EFL writing, writing revision, feedback uptake, Google Docs, Vietnamese EFL students, second language writing
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 2026-10-30, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-30 (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: Computer-mediated and face-to-face peer feedback: student feedback and revision in EFL writing. https://app.cochat.ai/papers/10.1080/09588221.2020.1868530
- The paper
- Pham, H. T. P. (2022). Computer-mediated and face-to-face peer feedback: student feedback and revision in EFL writing. Computer Assisted Language Learning. https://doi.org/10.1080/09588221.2020.1868530
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.