The DOI matches its Crossref record; 136 of the 137 references it cites were found in a registry or an index; none of the 28 sources its claims cite is checked for support yet.
RETRACTED ARTICLE: The effect of ChatGPT on students’ learning performance, learning perception, and higher-order thinking: insights from a meta-analysis
Open accessAbstract
As a new type of artificial intelligence, ChatGPT is becoming widely used in learning. However, academic consensus regarding its efficacy remains elusive. This study aimed to assess the effectiveness of ChatGPT in improving students’ learning performance, learning perception, and higher-order thinking through a meta-analysis of 51 research studies published between November 2022 and February 2025. The results indicate that ChatGPT has a large positive impact on improving learning performance ( g = 0.867) and a moderately positive impact on enhancing learning perception ( g = 0.456) and fostering higher-order thinking ( g = 0.457). The impact of ChatGPT on learning performance was moderated by type of course ( Q B = 64.249, P < 0.001), learning model ( Q B = 76.220, P < 0.001), and duration ( Q B = 55.998, P < 0.001); its effect on learning perception was moderated by duration ( Q B = 19.839, P < 0.001); and its influence on the development of higher-order thinking was moderated by type of course ( Q B = 7.811, P < 0.05) and the role played by ChatGPT ( Q B = 4.872, P < 0.05). This study suggests that: (1) appropriate learning scaffolds or educational frameworks (e.g., Bloom’s taxonomy) should be provided when using ChatGPT to develop students’ higher-order thinking; (2) the broad use of ChatGPT at various grade levels and in different types of courses should be encouraged to support diverse learning needs; (3) ChatGPT should be actively integrated into different learning modes to enhance student learning, especially in problem-based learning; (4) continuous use of ChatGPT should be ensured to support student learning, with a recommended duration of 4–8 weeks for more stable effects; (5) ChatGPT should be flexibly integrated into teaching as an intelligent tutor, learning partner, and educational tool. Finally, due to the limited sample size for learning perception and higher-order thinking, and the moderately positive effect, future studies with expanded scope should further explore how to use ChatGPT more effectively to cultivate students’ learning perception and higher-order thinking.
Overview
AI snapshot
Written from the paper’s full text.
TL;DR
This PRISMA-guided meta-analysis of 51 experimental and quasi-experimental studies (November 2022–February 2025) finds that ChatGPT has a large positive effect on students' learning performance (g = 0.867) and moderate positive effects on learning perception (g = 0.456) and higher-order thinking (g = 0.457). Effects are moderated by course type, learning model, duration, and the role ChatGPT plays, leading the authors to recommend scaffolded, sustained (4–8 week) integration of ChatGPT across grade levels and course types.
Key findings
- Overall effects across 72 independent effect sizes: learning performance g = 0.867 (k = 44, 95% CI [0.683, 1.052], p < 0.001); learning perception g = 0.456 (k = 19, 95% CI [0.287, 0.625], p < 0.001); higher-order thinking g = 0.457 (k = 9, 95% CI [0.255, 0.650], p < 0.001).
- Learning performance was moderated by type of course (Q_B = 64.249, p < 0.001; strongest for skills and competencies development, g = 0.874), learning model (Q_B = 76.220, p < 0.001; strongest for problem-based learning, g = 1.113, weakest for project-based learning, g = 0.239), and duration (Q_B = 55.998, p < 0.001; strongest at 4–8 weeks, g = 0.999). Grade level, role of ChatGPT, and application area were not significant.
- Learning perception was moderated only by duration (Q_B = 19.839, p < 0.001), with >8 weeks showing the largest effect (g = 1.054) versus ≤1 week (g = 0.203).
- Higher-order thinking was moderated by type of course (Q_B = 7.811, p < 0.05; STEM g = 0.737, language/writing g = 0.334, skills development g = 0.296) and role of ChatGPT (Q_B = 4.872, p < 0.05; intelligent tutor g = 0.945).
Methods
- Design: meta-analysis following PRISMA guidelines, with literature search, inclusion/exclusion screening, coding of study characteristics, and data analysis; search covered Web of Science Core Collection, ScienceDirect, Taylor & Francis, Wiley Online Library, and Scopus, restricted to November 2022–February 2025, starting from 6,621 records and yielding 51 included articles.
- Sample: 51 (quasi-)experimental studies with an experimental group using ChatGPT and a control group without it, both using the same learning method; 72 independent effect sizes were extracted. Interrater reliability for study selection was Cronbach's α = 0.96.
- Analysis: Hedges's g computed in Comprehensive Meta-Analysis (3.0/3.3.0), interpreted via Sawilowsky's (2009) thresholds; random-effects models were used given heterogeneity (I² = 89.243% for learning performance, 72.580% for learning perception, 55.813% for higher-order thinking); publication bias assessed via funnel plots and fail-safe N (5490, 399, and 101 respectively), with one-study-removal sensitivity analysis.
Limitations
- Stated by the authors: the sample of 51 documents included only one primary-school study and none at kindergarten level; only 19 and 9 studies examined learning perception and higher-order thinking respectively, versus a much larger sample for learning performance; only seven moderating variables were explored, omitting factors such as cultural background, parents' occupations, and national policies; and only experimental/quasi-experimental studies were included, excluding qualitative and mixed-methods research.
- Evident from the text: heterogeneity was high for learning performance (I² = 89.243%), and the authors note some funnel-plot points fell outside the diagonal lines, indicating potential bias or heterogeneity; several moderator categories (e.g., mixed roles, intelligent partners for higher-order thinking, primary school) were excluded from analyses due to insufficient data.
- Evident from the bibliographic record: the article is labeled "RETRACTED ARTICLE" in the title, though the retraction is not discussed in the full text provided.
Model-written summary. Check it against the paper before relying on it.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
68 Own result · 19 Own method · 17 Not checked yet · 6 No basis given · 5 Own interpretation
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
“The impact of ChatGPT on learning performance was moderated by type of course (Q B = 64.249,P < 0.001), learning model (Q B = 76.220,P < 0.001), and duration (Q B = 55.998,P < 0.001); its effect on learning perception was moderated by duration (Q B = 19.839,P < 0.001); and its influence on the development of higher-order thinking was moderated by type of course (Q B = 7.811, P < 0.05) and the role”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Abstract
plaincorrelational- 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
“The impact of ChatGPT on learning performance was moderated by type of course (Q B = 64.249,P < 0.001), learning model (Q B = 76.220,P < 0.001), and duration (Q B = 55.998,P < 0.001); its effect on learning perception was moderated by duration (Q B = 19.839,P < 0.001); and its influence on the development of higher-order thinking was moderated by type of course (Q B = 7.811, P < 0.05) and the role”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Abstract
plaincorrelational- 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
“The results indicate that ChatGPT has a large positive impact on improving learning performance (g = 0.867) and a moderately positive impact on enhancing learning perception (g = 0.456) and fostering higher-order thinking (g = 0.457).”
- Also stated in: § Results · hedged
“The overall effects of ChatGPT on enhancing learning performance, improving learning perception, and promoting higher-order thinking were calculated as g = 0.867, g = 0.456, and g = 0.457, respectively, indicating that ChatGPT has a large positive effect on learning performance and a medium positive effect on learning perception and higher-order thinking.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Abstract
hedgeddirect- 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.
- Also stated in: § Results · hedged
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
“The Q-test for higher-order thinking is 18.105, and the p-value of the Q-test is less than 0.1.The I² value is 55.813%, indicating moderate heterogeneity in the included studies regarding the effect of ChatGPT on higher-order thinking.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Results
hedged- 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
“The impact of ChatGPT on learning performance was moderated by type of course (Q B = 64.249,P < 0.001), learning model (Q B = 76.220,P < 0.001), and duration (Q B = 55.998,P < 0.001); its effect on learning perception was moderated by duration (Q B = 19.839,P < 0.001); and its influence on the development of higher-order thinking was moderated by type of course (Q B = 7.811, P < 0.05) and the role”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Abstract
plaincorrelational- 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
“According to Table 4, the Qvalue for learning performance is 339.725, and the p-value of the Q-test is <0.1.The I² value is 89.243%, indicating a high degree of heterogeneity in the included studies regarding the impact of ChatGPT on learning performance.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Results
hedged- 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
“The impact of ChatGPT on learning performance was moderated by type of course (Q B = 64.249,P < 0.001), learning model (Q B = 76.220,P < 0.001), and duration (Q B = 55.998,P < 0.001); its effect on learning perception was moderated by duration (Q B = 19.839,P < 0.001); and its influence on the development of higher-order thinking was moderated by type of course (Q B = 7.811, P < 0.05) and the role”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Abstract
plaincorrelational- 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
“The impact of ChatGPT on learning performance was moderated by type of course (Q B = 64.249,P < 0.001), learning model (Q B = 76.220,P < 0.001), and duration (Q B = 55.998,P < 0.001); its effect on learning perception was moderated by duration (Q B = 19.839,P < 0.001); and its influence on the development of higher-order thinking was moderated by type of course (Q B = 7.811, P < 0.05) and the role”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Abstract
plaincorrelational- 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
“The results indicate that ChatGPT has a large positive impact on improving learning performance (g = 0.867) and a moderately positive impact on enhancing learning perception (g = 0.456) and fostering higher-order thinking (g = 0.457).”
- Also stated in: § Results · hedged
“The overall effects of ChatGPT on enhancing learning performance, improving learning perception, and promoting higher-order thinking were calculated as g = 0.867, g = 0.456, and g = 0.457, respectively, indicating that ChatGPT has a large positive effect on learning performance and a medium positive effect on learning perception and higher-order thinking.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Abstract
hedgeddirect- 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.
- Also stated in: § Results · hedged
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
“The impact of ChatGPT on learning performance was moderated by type of course (Q B = 64.249,P < 0.001), learning model (Q B = 76.220,P < 0.001), and duration (Q B = 55.998,P < 0.001); its effect on learning perception was moderated by duration (Q B = 19.839,P < 0.001); and its influence on the development of higher-order thinking was moderated by type of course (Q B = 7.811, P < 0.05) and the role”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Abstract
plaincorrelational- 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.
- It may state two separate claims as one. 30
- The quote only partly states this claim. 1
- Probably not a substantial claim. 14
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
“ChatGPT was found to be beneficial across all time periods, with interventions lasting more than 8 weeks (g = 1.054) showing a substantially larger effect size than those lasting less than 1 week (g = 0.203), 1-4 weeks (g = 0.404), and 4-8 weeks (g = 0.493).”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Results
plaincorrelational- 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.
It may state two separate claims as one.
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
“The results indicate that ChatGPT has a substantial positive effect on students' learning performance, with an overall effect size of 0.867.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
hedgeddirect- 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.
It may state two separate claims as one.
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
“Notably, compared to its large positive impact on improving learning performance (g = 0.867), ChatGPT has only a moderately positive impact on learning perception.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
plaindirect- 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.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“However, the improvements in learning performance attributed to ChatGPT are quite strong (g = 0.867), while the improvements in higher-order thinking are only moderate.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
plaindirect- 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.
It may state two separate claims as one.
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
“The final sample consisted of 51 articles, published between November 2022 and February 2025.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Inclusion criteria
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.
It may state two separate claims as one.
The paper states this without a source or data behind it here, so there is nothing to check it against. The quote shows it is in the paper.
In the paper
“this study yielded a unique conclusion: in project-based learning, the effect of ChatGPT on students' learning performance is the weakest, showing only a small positive impact.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
plaindirect- 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.
It may state two separate claims as one.
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
“| Mixed | 2 | 0.313[0.024,0.602] | 2.122*** | 1 | | |”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Limitations and future research
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
The quote only partly states this claim.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“The results show that, for learning performance, after the removal of any single study, the range of the 95% confidence interval (CI) of the effect size still remained between 0.462 and 0.574 for the fixedeffects model and between 0.683 and 1.052 for the random-effects model.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Results
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.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“In this study, the fail-safe N for learning performance was calculated to be 5490 (5k + 10 = 230, k = 44), for learning perception was calculated to be 399 (5k + 10 = 105, k = 19), and for higher-order thinking was calculated to be 101 (5k + 10 = 55, k = 9), indicating that the possibility of publication bias is relatively small for this study.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Results
hedged- 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.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“In this study, the fail-safe N for learning performance was calculated to be 5490 (5k + 10 = 230, k = 44), for learning perception was calculated to be 399 (5k + 10 = 105, k = 19), and for higher-order thinking was calculated to be 101 (5k + 10 = 55, k = 9), indicating that the possibility of publication bias is relatively small for this study.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Results
hedged- 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.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“In this study, the fail-safe N for learning performance was calculated to be 5490 (5k + 10 = 230, k = 44), for learning perception was calculated to be 399 (5k + 10 = 105, k = 19), and for higher-order thinking was calculated to be 101 (5k + 10 = 55, k = 9), indicating that the possibility of publication bias is relatively small for this study.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Results
hedged- 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.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“The effect values for the impact of ChatGPT on learning performance, learning perception, and higher-order thinking are relatively evenly distributed on both sides of the funnel plot, indicating that the data have sufficiently high reliability to be used for meta-analysis.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Results
hedged- 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.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“All relevant studies were conducted at the college level, and thus no analysis was performed for grade level.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Results
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“Furthermore, the experimental and control groups had to adopt the same learning method.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Inclusion criteria
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.
Probably not a substantial claim.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“ChatGPT can make a course more interesting and encourage students to actively search for information, ultimately contributing to an increase in learning perception.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
plaindirect- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“The literature inclusion and exclusion criteria were developed by two researchers specializing in educational technology, and are presented in Table 1.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Method
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it.
In the paper
“This finding agrees with the results of Guan et al. (2024) and may be attributed to the consistent positive feedback students receive from ChatGPT during extended usage.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
What the sources say
- Not checked yetcites Guan et al. 2024 source
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“The various roles it can play and the areas in which it can be applied are essentially manifestations of its core function as a learning support tool.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
plain- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence.
None of the sources cited for this claim have been checked yet.
In the paper
“This may be because such courses typically involve well-defined task objectives and procedural steps; ChatGPT provides immediate feedback, targeted guidance, and problemsolving support (Rakap, 2023;Urban et al. 2024), and thus can enable students to complete such tasks more efficiently.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
What the sources say
- Not checked yetcites Rakap, 2023
- Not checked yetcites Urban et al. 2024
hedgedconditional- 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.
Probably not a substantial claim. It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“It should be noted that the primary school sample was excluded from the analysis because it consisted of only one study, which did not find a statistically significant effect.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Results
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
In the paper
“Moreover, ChatGPT-generated content is not always completely accurate, and this bias can hinder the development of students' higherorder thinking (Niloy et al. 2023), thus limiting the overall degree to which ChatGPT can promote students' higher-order thinking.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
What the sources say
- Not checked yetcites Niloy et al. 2023
plaindirect- 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.
It may state two separate claims as one.
In the paper
“It thus may lack critical analysis abilities and the ability to provide creative problem-solving solutions (Lu et al. 2024).”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
What the sources say
- Not checked yetcites Lu et al. 2024
hedgedconditional- 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.
It may state two separate claims as one.
- Rests on an assumption: ChatGPT is based on pretrained models that rely on existing data and patterns.
In the paper
“This may be because the prolonged use of ChatGPT can lead students to become overly reliant on the AI tool, and thus they may neglect to reinforce the knowledge they have learned (Agarwal, 2023), which, in turn, would result in a decline in learning performance.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
What the sources say
- Not checked yetcites Agarwal, 2023
hedgedconditional- 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.
Probably not a substantial claim. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“These features enable ChatGPT to support more diverse learning scenarios, which may, in turn, contribute to its larger effect size.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
hedgedconditional- 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.
Probably not a substantial claim. It may state two separate claims as one.
In the paper
“This may be because project-based learning emphasizes the completion of a comprehensive project in a realworld context (Botha, 2010).In such a scenario, ChatGPT's support for addressing complex problem-solving tasks may be limited, resulting in only a minor effect.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
What the sources say
- Not checked yetcites Botha, 2010
hedgedconditional- 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.
Probably not a substantial claim. It may state two separate claims as one.
In the paper
“Since differences in sample sizes between studies might introduce bias into the calculated effect sizes, Hedges's g was used to adjust the effect sizes to correct for small sample size biases (Hedges, 1981).”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Data analysis.
What the sources say
- Not checked yetcites Hedges, 1981 source
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.
Probably not a substantial claim.
None of the sources cited for this claim have been checked yet.
In the paper
“This may be attributed to its inherent strengths, in that ChatGPT excels at providing clear explanations for a wide range of problems, whether establishing simple correspondences between geometric structures and algebraic equations (Poola and Božić, 2023) or explaining complex scientific concepts (Ali et al. 2024).”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
What the sources say
- Not checked yetcites Poola and Božić, 2023
- Not checked yetcites Ali et al. 2024 source
hedgedconditional- 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.
It may state two separate claims as one.
In the paper
“Meta-analysis can be used to comprehensively analyze the results of multiple independent research studies that are focused on the same research purpose, reveal the differences and commonalities among them, and thereby draw universal and regular conclusions (Lipsey and Wilson, 2001).”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Introduction C
What the sources say
- Not checked yetcites Lipsey and Wilson, 2001
plain- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence. It may state two separate claims as one.
None of the sources cited for this claim have been checked yet.
In the paper
“In STEM and related courses, students are often required to collaborate with ChatGPT to complete complex projects (Li, 2023;Li et al. 2024), and the process of completing such projects inherently promotes the development of higher-order thinking skills such as creative thinking, problem-solving, and critical thinking.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
What the sources say
- Not checked yetcites Li, 2023
- Not checked yetcites Li et al. 2024
plaindirect- 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.
Probably not a substantial claim. It may state two separate claims as one.
A conclusion the paper draws from its own results — it cites no other source for it, so there is none to check. The quote shows it is in the paper.
In the paper
“This may be due to the fact that when students first started using ChatGPT, they lacked the questioning skills needed to use it effectively.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
hedgedconditional- Asserted: passes — The authors put it forward as true.
- Checkable: fails — It may be too vague to check against evidence.
- Load-bearing: passes — The paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. It may be too vague to check against evidence.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“Articles that met the coding criteria were independently selected by two researchers, and acceptable interrater reliability was achieved (Cronbach's α = 0.96).”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Inclusion criteria
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
In the paper
“When using ChatGPT, students continuously reflect on their learning processes and adjust their learning strategies in real-time.This process effectively fosters the development of students' higherorder thinking skills (Lu et al. 2024).”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > How effective is ChatGPT in promoting learning performance?
What the sources say
- Not checked yetcites Lu et al. 2024
plaindirect- 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.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“Finally, this study only selected experimental and quasi-experimental studies and did not include relevant qualitative or mixed-methods research.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Limitations and future research
plain- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: unclear — It is unclear whether the paper’s conclusions depend on it.
- Based: passes — The paper gives a reason why it holds.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“First, this study sample size was limited, comprising only 51 documents, of which only one addressed primary school and none addressed kindergarten.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Limitations and future research
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.
It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“Second, this study only explored seven moderating variables, but there are many more variables that could potentially affect the impact of ChatGPT on student learning, such as students' cultural backgrounds, parents' occupations, and relevant national policies.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Limitations and future research
hedged- Asserted: passes — The authors put it forward as true.
- Checkable: passes — It can be checked against evidence.
- Load-bearing: fails — The paper’s conclusions may not depend on it.
- Based: passes — The paper gives a reason why it holds.
Probably not a substantial claim. The paper’s conclusions may not depend on it. It may state two separate claims as one.
What the paper did — it cites no source for this, so there is none to check. The quote shows it is in the paper.
In the paper
“Moreover, the number of included samples that examined learning perception and higher-order thinking was relatively small, with only 19 and 9 studies, respectively, compared to a significantly larger sample for learning performance.This imbalance may limit our ability to comprehensively understand ChatGPT's impact on learning perception and higher-order thinking.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Limitations and future research
hedgedconditional- 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.
It may state two separate claims as one.
The paper relies on this without support.
ChatGPT is based on pretrained models that rely on existing data and patterns.
“ChatGPT is based on pretrained models that rely on existing data and patterns.”
The paper relies on this without support.
The meta-analysis by Chen et al. (2025) on other AI assessment tools is comparable to this meta-analysis on ChatGPT.
“The results indicate that the effect size for the impact of ChatGPT on students' learning performance is significantly larger than that of other AI assessment tools (g = 0.390).”
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?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Their study also found a large effect on learning performance, particularly when structured frameworks like Bloom's Taxonomy are used to guide AI-supported tasks.”
From Enhancing service-learning through Generative AI: a mixed-methods study on educational Game design in a finance course (2025) · cites this paper as Wang and Fan (2025)
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Wang and Fan (2025) found that ChatGPT has a moderate effect on higher-order thinking, but only when appropriate scaffolding is present, such as when instructors position ChatGPT as a learning partner.”
From Enhancing service-learning through Generative AI: a mixed-methods study on educational Game design in a finance course (2025) · cites this paper as Wang and Fan (2025)
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“much of the literature on AI-supported feedback has focused on general writing tasks or broad comparisons among feedback sources, whereas disciplinary reasoning in university science courses has received less sustained attention”
From Human-centered GenAI feedback design in higher education: a multisite experiment on direct, reflective, and hybrid approaches to scientific argumentation (2026) · cites this paper as Wang & Fan, 2025
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“Wang and Fan (2025) analysed studies from a similar period and reported comparable findings to those of Deng and colleagues: a substantial positive effect of ChatGPT on learning.”
From ChatGPT in Education: An Effect in Search of a Cause (2025) · cites this paper as Wang and Fan (2025)
The cited source says what the claim attributes to it.
In the citing paper
“which broadens forms of expression, strengthens creativity [4,5], and supports higher-order and critical thinking [6,7].”
From A Systematic Review of Generative AI in K–12: Mapping Goals, Activities, Roles, and Outcomes via the 3P Model (2025) · cites this paper as [7]
The cited source says what the claim attributes to it.
In the citing paper
“Existing evidence suggests that GenAI can support learning performance, learning perceptions, and higher-order thinking, but these benefits vary substantially depending on the role assigned to the tool, the instructional design, and the characteristics of the task”
From Human-centered GenAI feedback design in higher education: a multisite experiment on direct, reflective, and hybrid approaches to scientific argumentation (2026) · cites this paper as Wang & Fan, 2025
The cited source says what the claim attributes to it.
In the citing paper
“More broadly, the study adds nuance to recent syntheses of GenAI in higher education by suggesting that apparent benefits are highly design-sensitive and that concerns about authorship, evidence of learning, and learner responsibility are not only matters of policy or integrity, but also matters of feedback design(Wang & Fan, 2025;Xia et al., 2024).”
From Human-centered GenAI feedback design in higher education: a multisite experiment on direct, reflective, and hybrid approaches to scientific argumentation (2026) · cites this paper as Wang & Fan, 2025
The check wasn’t confident either way, so it gives no verdict — never disputed.
In the citing paper
“immediate product improvement is often examined without simultaneously investigating feedback uptake, self-regulated learning, and later performance without AI support, making it difficult to determine whether observed gains reflect durable learning, stronger revision processes, or temporary performance support”
From Human-centered GenAI feedback design in higher education: a multisite experiment on direct, reflective, and hybrid approaches to scientific argumentation (2026) · cites this paper as Wang & Fan, 2025
Experimental — accuracy not yet measured
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 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
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Coding of study characteristics.
“A total of six moderating variables were selected (see Table 2).”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Limitations and future research
“Second, this study only explored seven moderating variables, but there are many more variables that could potentially affect the impact of ChatGPT on student learning”
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
“| >8 week | 13 | 0.531[0.490,0.684] | 11.912*** | 12 | | |”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Limitations and future research
Reported 0.531 · recomputed 0.587
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
“| Language learning and academic writing | 12 | 0.531[0.310,0.531] | 7.454*** | 11 | | |”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Limitations and future research
Reported 0.531 · recomputed 0.4205
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
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Data availability
“No datasets were generated or analysed during the current study.”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Coding of study characteristics.
“Ultimately, the coding process resulted in the extraction of 72 independent effect sizes from the selected 51 articles.”
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
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Results
“According to Table 4, the Qvalue for learning performance is 339.725”
§ The effect of ChatGPT on students' learning performance, learning perception, and higher-order thinking: insights from a meta-analysis > Limitations and future research
“| Learning performance | 44 | 0.867 | 0.683 | 1.052 | 8.458 | <0.001 | 399.725 | 43 | 89.243 | <0.001 |”
Citations
Cites
The PRISMA 2020 statement: An updated guideline for reporting systematic reviews
Page, M. J., McKenzie, J. E., Bossuyt, P. M., Boutron, I., Hoffmann, T. C., Mulrow, C. D., Shamseer, L., Tetzlaff, J. M., Akl, E. A., Brennan, S. E., et al. (2021) · BMJ · doi:10.1136/bmj.n71 · Cited by 108,283
Measuring inconsistency in meta-analyses
Higgins, J. P. T., Thompson, S. G., Deeks, J. J., & Altman, D. G. (2003) · BMJ · doi:10.1136/bmj.327.7414.557 · Cited by 63,139
Preferred reporting items for systematic reviews and meta-analyses: The PRISMA statement
Moher, D., Liberati, A., Tetzlaff, J., Altman, D. G., & the PRISMA Group* (2009) · Annals of Internal Medicine · doi:10.7326/0003-4819-151-4-200908180-00135 · Cited by 37,554
Borenstein, M., Hedges, L. V., Higgins, J. P. T., & Rothstein, H. R. (2009) · doi:10.1002/9780470743386 · Cited by 14,225
Technology Acceptance Model 3 and a Research Agenda on Interventions
Venkatesh, V., & Bala, H. (2008) · Decision Sciences · doi:10.1111/j.1540-5915.2008.00192.x · Cited by 8,159 · cited 1×
“The technology acceptance model posits that perceived usefulness and perceived ease of use in fl uence an individual ’ s intention or attitude toward using new technology (Venkatesh and Bala, 2008).”
ChatGPT for Good? On Opportunities and Challenges of Large Language Models for Education
Kasneci, E., Seßler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., Gasser, U., Groh, G., Günnemann, S., Hüllermeier, E., et al. (2023) · doi:10.35542/osf.io/5er8f · Cited by 7,552
ChatGPT for good? On opportunities and challenges of large language models for education
Kasneci, E., Sessler, K., Küchemann, S., Bannert, M., Dementieva, D., Fischer, F., Gasser, U., Groh, G., Günnemann, S., Hüllermeier, E., et al. (2023) · Learning and Individual Differences · doi:10.1016/j.lindif.2023.102274 · Cited by 7,326 · cited 3×
“…studies have claimed that ChatGPT can improve students ’ learning performance, learning perception, and higher-order thinking (Jalil et al. 2023; Kasneci et al. 2023; Lu et al. 2024), but other studies have found that it does not signi fi cantly improve students ’ learning performance or…”
Distribution Theory for Glass's Estimator of Effect Size and Related Estimators
Hedges, L. V. (1981) · Journal of Educational Statistics · doi:10.2307/1164588 · Cited by 5,695
Schulz (1995) · doi:10.1001/jama.1995.03520290060030 · Cited by 5,634
New Effect Size Rules of Thumb
Sawilowsky, S. S. (2009) · Journal of Modern Applied Statistical Methods · doi:10.22237/jmasm/1257035100 · Cited by 3,254
Research works printed but not linked: 5
- Alsharif MHK, Elamin AY, Almasaad JM, Bakhit NM, Alarifi A, Taha KM, Hassan WA, Zumrawi E (2024) Using ChatGPT to create engaging problem-based learning scenarios in anatomy: a step-by-step guide. New Armen Med J 18(4). https://doi.org/10.56936/18290825-4.v18.2024-98 · DOI
- Lipsey MW, Wilson DB (2001) Practical meta-analysis. Sage Publications, Inc
- Luo H, Liao X, Ru Q, Wang Z (2024) Generative AI-supported teacher comments: an empirical study based on junior high school mathematics classrooms. e-Educ Res 45(5):58–66. https://doi.org/10.13811/j.cnki.eer.2024.05.008 · DOI
- Poola I, Božic V (2023) Guiding AI with human intuition for solving mathematical problems in ChatGPT. Int J Eng Sci Res 11(07):31–40. https://www.researchgate.net/publication/373447147
- Vygotskij LS, Hanfmann E (1966) Thought and language. MIT Press
Cited by
Sorted by citations
Items 201–300 of 463
Chu, C. H., Jibb, L., MacInnes, N., Shan, C. C., Mohammed, S., Nanos, S. M., O'Sullivan, M., Muntaner, C., Thomson, H., & Widger, K. A. (2026) · Journal of Clinical Nursing · doi:10.1111/jocn.70376 · Cited by 1
Lara, M. A. B., Cazales, V. J. R., Rivas, N. E., & Mendiola, M. S. (2026) · Revista Mexicana de Investigación Educativa · doi:10.66708/rmie.v31i110.2042 · Cited by 1
Chirinos, R. A. G. (2026) · Journal of Intelligent Decision Making and Information Science · doi:10.59543/jidmis.v3.1042 · Cited by 1
Human-Centered Approaches to Generative AI in Education
Diordiev, E.-A., & Makransky, G. (2026) · Springer eBooks · doi:10.1007/978-3-032-24819-0_6 · Cited by 1
van Broekhoven, K., Ouwehand, K., de Koning, B. B., Nederhand, M. L., & Bocanegra, B. R. (2026) · Springer eBooks · doi:10.1007/978-3-032-24819-0_12 · Cited by 1
Bremner, C. B., Brooks, D. D., Campbell, M., Carroll, C. C., Russell, D. M., & Swartz, A. M. (2026) · Kinesiology Review · doi:10.1123/kr.2026-0071 · Cited by 1
Banquez-Maturana, A.-G. (2026) · Journal of Trends and Challenges in Artificial Intelligence · doi:10.61552/jai.2026.03.002 · Cited by 1
Generative AI in English Sixth Form Education: Student Use, Perceptions, and Literacy GapsPreprint
Ennion, M. (2025) · Research Square · doi:10.21203/rs.3.rs-7122270/v1 · Cited by 1
Maheux, A. J., & Maes, C. (2026) · Perspectives on Psychological Science · doi:10.1177/17456916261460771 · Cited by 2
From Topic to Thesis: Critical Engagement with Generative AI in Thesis Preparation
Virkki, O. T., Naeem, U., & Styve, A. (2026) · 2026 IEEE Global Engineering Education Conference (EDUCON) · doi:10.1109/educon67543.2026.11574181
Huot, S., & Loch, S. (2025) · International Journal of Latest Technology in Engineering Management & Applied Science · doi:10.51583/ijltemas.2025.140500058
Flip the script: leveraging blockchain and generative AI for a bottom-up learning paradigmPreprint
Franceschet, M. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5296754
Ingebrigtsen, M., & Lukic, M. (2025) · doi:10.31234/osf.io/wgu56_v1
The AI Revolution in Education: From Cognitive Enhancement to Hybrid IntelligencePreprint
Rodrigues, D. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5327624
Preciado, F. A. M., & Pinargote, F. M. M. (2025) · Arandu-UTIC. · doi:10.69639/arandu.v12i3.1353
Wu, Y., Meng, J., Liu, P., Xiong, Y., Chen, S., Yin, H., Wang, S., & Gao, S. (2025) · doi:10.2196/preprints.83291
Zhang, T. (2025) · International Conference on Education, Knowledge and Information Management (ICEKIM) · doi:10.1145/3756580.3756622
Effect of Artificial Intelligence on Mathematics Performance: A Meta-Analysis
Maanu, V., Boateng, F. O., & Larbi, E. (2025) · INTERNATIONAL JOURNAL OF MATHEMATICS AND COMPUTER RESEARCH · doi:10.47191/ijmcr/v13i9.09
Hegde, S., & Chandunavar, D. J. M. (2025) · International Journal of Social Science and Human Research · doi:10.47191/ijsshr/v8-i9-85
Yee, J., Lim, F. V., & Quek, J. (2026) · Journal of Applied Learning & Teaching · doi:10.37074/jalt.2026.9.2.6 · Cited by 1
Einleitung: Intelligente Maschinen verändern die Welt
Lang, F. U. (2025) · Springer eBooks · doi:10.1007/978-3-662-71542-0_1
Phúc, V. L. H. (2025) · Tạp chí Khoa học Trường Đại học Mở Hà Nội · doi:10.59266/houjs.2025.778
ChatGPT gains impact in education and research: potential, benefits, and challenges
(2025) · Scientific Papers of Silesian University of Technology Organization and Management Series · doi:10.29119/1641-3466.2025.231.2
Revisión bibliométrica sobre la adopción de inteligencia artificial en la educación superior
Jiménez, M. U., López-Mallama, O. M., & Espinosa, A. A. (2025) · Revista Guillermo de Ockham · doi:10.21500/22563202.7780
Impact of Generative AI Characteristics on Continuous Use Intention: Mediating Role of Utilization
Ko, Y.-J. (2025) · Journal of Digital Contents Society · doi:10.9728/dcs.2025.26.10.2961
Kim, K. T., & Bae, H.-Y. (2025) · Journal of Digital Contents Society · doi:10.9728/dcs.2025.26.10.2859
Lernende und ihre emotionalen Anliegen in der Interaktion mit Chatbots
Schulz, L., & Schmid-Meier, C. (2025) · merz | medien + erziehung · doi:10.21240/merz/2025.6.09
Chamba, J. A. T., & Palacios, B. E. C. (2025) · Arandu-UTIC. · doi:10.69639/arandu.v12i4.1694
Tran, T. T. P. (2025) · International journal of AI in language education. · doi:10.54855/ijaile.25237
AI usage strategies for college students: the necessity of establishing AI usage standards
Duan, Z. (2025) · Advances in Education, Humanities and Social Science Research · doi:10.56028/aehssr.15.1.932.2025
ОСВІТНІ ТРЕНДИ: ЗАСТОСУВАННЯ ЕЛЕМЕНТІВ ШІ ДЛЯ ФОРМУВАННЯ ФАХОВИХ КОМПЕТЕНЦІЙ ЗДОБУВАЧІВ
університет, О. Д. А., Березовський, С., Koliada-Berezovska, T., & університет, О. Д. А. (2025) · Notes of modern science · doi:10.61718/nsn31.01
Prihatmoko, Y., Praherdhiono, H., Nindigraha, N., Firdaus, K. H. C., & Utami, D. D. (2025) · International Conference on Emerging Technologies · doi:10.1109/icet67257.2025.11290897
Tahir, M. S., Qurat-Ul-Ain, & Latif, A. (2025) · Social science review archives. · doi:10.70670/sra.v3i3.1084
ChatGPT as a tool for Enhancing Explanation Letter Writing Skills
O’Brien-Melford, M. A. P., & Gador, S. C. (2025) · International Journal of English Literature and Social Sciences · doi:10.22161/ijels.106.16
Khairani, S., Kurniawati, Y., Permana, D., & Fitri, F. (2025) · UNP Journal of Statistics and Data Science · doi:10.24036/ujsds/vol3-iss4/430
AI-Enhanced English Language Learning and Critical Thinking Development by EFL Omani Students
Jomaa, N. J., Attamimi, R. A., & Al-Yafaei, Y. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-3561-2.ch002
Tasić, N., Vecštejn, I., Amižić, V., & Milić, T. (2025) · International Conference on Industrial Engineering and Environmental Protection (IIZS) · doi:10.5937/iizs25060t
CHATBOT SEBAGAI INOVASI PEMBELAJARAN DIGITAL: SLR TREN GLOBAL DAN NASIONAL 2020–2025
Yafianto, I. F., Pradana, R. T. A., & Winahyu, R. S. (2025) · Jurnal Riset Teknik Komputer · doi:10.69714/z4nq9733
THE USE OF WEB TOOLS IN EDUCATION: A CASE STUDY IN A SECONDARY SCHOOL IN MONTENEGRO
Ilic, B. (2025) · International Journal of Management Trends Key Concepts and Research · doi:10.58898/ijmt.v4i2.95-106
Kokubun, K., Yamakawa, Y., Yoshida, A., & Sanji, S. (2025) · Brain Sciences · doi:10.3390/brainsci16010058
Ohno, S. (2025) · IEEE International Conference on Smart Internet of Things (SmartIoT) · doi:10.1109/smartiot66867.2025.00074
Alhokail, L. (2025) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-3451-6.ch007
STUDENT ENGAGEMENT, DEVELOPING CRITICAL THINKING SKILLS AND THE USE OF AI
Hatcher, J. W., Davis, J. M., & Russell, B. D. A. S. (2025) · Cognition and Exploratory Learning in Digital Age · doi:10.33965/celda2025_202509c061
DeepRead: A Personalized Literacy Enhancement Service for Adults Using Generative AI
Seo, B., Kim, D., Park, H., & Kim, S. (2025) · Journal of Digital Contents Society · doi:10.9728/dcs.2025.26.12.3477
GenAI and Ancient Greek Natural Philosophy
Sotiropoulos, D., Xenakis, C., Xenakis, A., & Kalogiannakis, M. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-2913-0.ch003
Qinjie, S., & Arunrugstichai, W. (2026) · Environment and Social Psychology · doi:10.59429/esp.v11i1.4348
The influence of Generative AI on the transfer of algorithmic skills in students
Kerimbayev, N., Menlibay, Z., Shadiev, R., Mambetakunov, E. M., Kolumbayeva, S., Nurymova, A., & Adamova, K. (2025) · International Conference Automatics and Informatics (ICAI) · doi:10.1109/icai67591.2025.11324568
Practical Application and Future Prospects of Generative AI in Education
Yoshida, L. (2026) · Journal of JSEE · doi:10.4307/jsee.74.1_8
Amjad, H. S., & Alfalig, H. A. (2026) · Arab World English Journal · doi:10.24093/awej/ai3.15 · Cited by 1
The Brat Brain: ChatGPT and the Crisis of Intellectual Effort in Language Education
Wardak, M. (2026) · Arab World English Journal · doi:10.24093/awej/ai3.4
Βαλασκατζής, Η. (2026) · Συνέδρια της Ελληνικής Επιστημονικής Ένωσης Τεχνολογιών Πληροφορίας & Επικοινωνιών στην Εκπαίδευση · doi:10.12681/cetpe.9369
Elgammal, N., Fawzy, H. E., Elballat, B. A.-E., Braimoh, J. J., & eita, A. M. A. (2026) · Frontiers in Education · doi:10.3389/feduc.2026.1737363
Oluwagbenga, T. T., & Itanyi, M. F. (2026) · International Journal of Research · doi:10.26643/ijr/2026/18
Two Years to Madness: A 30-Month Journal of Human–Machine Communication and Second-Hand Reality
Prahl, A. (2026) · Human-Machine Communication · doi:10.30658/hmc.12.4
Trang, T., Tran, P., Le, T. M., Le, L., Maheshwari, G., Do, N., & Dao, S. V. T. (2025) · RIVF International Conference on Computing and Communication Technologies (RIVF) · doi:10.1109/rivf68649.2025.11365126
Aristawidya, A. P., & Ramadhana, M. R. (2026) · INJECT (Interdisciplinary Journal of Communication) · doi:10.18326/inject.v11i1.6235
Perreau, É., Pinquié, R., Drouin, R., Huchard, J., & Piques, J.-D. (2025) · IEEE International Symposium on Systems Engineering · doi:10.1109/isse65546.2025.11370089
Laaksonen, A. (2026) · Undergraduate topics in computer science · doi:10.1007/978-3-032-07453-9_10
Exploring Communication Literacy Levels and ChatGPT Usage Among Senior High School Students
Fidel, D. M. C., Pangilinan, J. C., Dator, M. E. M., Coronel, M. R. D., & Tutor, B. O. D. (2025) · International Workshop on Artificial Intelligence and Education (WAIE) · doi:10.1109/waie67422.2025.11381146
Analyzing University Students’ Perceptions of ChatGPT: A Case Study at Ajman University, UAE
Alqawasmi, A., Alsalhi, N. R., Eltahir, M. E., Al-Qatawneh, S. S., Aljarrah, K. H., & Al-Taani, A. A. K. (2025) · International Conference on Social Networks Analysis, Management and Security (SNAMS) · doi:10.1109/snams67467.2025.11391238
Cognitive and Cultural Dimensions of AI-Generated Language
Ismail, N. M. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-4397-6.ch008
Nguyen, P. D. A. (2025) · Journal of Thu Dau Mot University · doi:10.37550/tdmu.ejs/2025.04.686
An Explication and Classroom Field Study of the Virtual Human Interaction Lab’s Expert (VHIL-E) LLM
Bailenson, J. N., You, J., Markowitz, D. M., Petersen, G. B., Ratan, R., Santoso, M., & Wang, P. (2026) · Cyberpsychology Behavior and Social Networking · doi:10.1177/21522715261423752
ІНТЕЛЕКТУАЛЬНІ ТЬЮТОРСЬКІ СИСТЕМИ В ОСВІТНЬОМУ ПРОЦЕСІ ДЛЯ ПІДГОТОВКИ МАЙБУТНІХ ІТ-ФАХІВЦІВ
Мельник, С. (2026) · Education Innovation Practice · doi:10.31110/2616-650x-vol14i2-011
Contreras, A. U., & Alonso, R. F. (2026) · doi:10.36006/16458-1
AI-Powered Virtual Teachers and Their Role in Developing Critical Thinking Skills
Aksin, A. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-2637-5.ch008
The Impact of AI Tools on Cross-Border E-Commerce English Learning Effectiveness
Liang, M., Liang, L. Y., Tan, Z., & Xing, L. (2025) · International Conference on Educational Technology Management (ICETM) · doi:10.1109/icetm67477.2025.11413566
Modelling Prompting Behaviours in LLM-Mediated Task Solving
Ammar, A. E. C. (2026) · ACM International Conference on Intelligent User Interfaces (IUI) · doi:10.1145/3742414.3789230
Salazar, N. A. B., -Anchundia, B. J. D., & Campoverde-Moscol, A. I. (2026) · MQRInvestigar · doi:10.56048/mqr.2026.e164
Uporaba ChatGPT-ja kot podpornega orodja pri pouku psihologije
Slatenšek, B. (2026) · Revija Inovativna pedagogika · doi:10.63069/hfaqzr98
Ojeda, A. D., Álvarez, D. D. O., Cañavera, A. M., & Altahona, O. R. C. (2026) · Formación universitaria · doi:10.4067/s0718-50062026000100115
Generative AI and Redefining Scope of Language Curriculum
Ahmed, R. A., Al-kadi, A. M. T., & Namusiginyi, B. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-4521-5.ch009
Re-Tooling Research Pedagogy in Higher Education Contexts
Al-Sowaidi, B. S. A., Kamalie, M., & Saif, H. (2026) · Advances in computational intelligence and robotics book series · doi:10.4018/979-8-3373-4521-5.ch013
Menyiapkan Generasi Cerdas Digital : Pengenalan Konsep AI di SMPN 3 Gondang Mojokerto
Fahmi, M. N., Izzudin, H., Sari, N. L., Rosyida, Y. F., Najih, A., & Hafzy, L. A. (2026) · Insani Journal of Social Development and Humanities · doi:10.52620/ijsdh.v1i1.16
Urbina, A. A. N., & Castillo, H. S. (2026) · Editorial Transdigital eBooks · doi:10.56162/transdigitalbc13.14
Design, Implementation, and Early Engagement Outcomes of a Student-Led AI Tutor in Higher Education
Alsharif, A. (2026) · IAFOR International Conference on Education, official conference proceedings · doi:10.22492/issn.2189-1036.2026.15
Liu, Y., & Bonilla, M. (2026) · IAFOR International Conference on Education, official conference proceedings · doi:10.22492/issn.2189-1036.2026.7
ПОДВОРНЮК, О. О., & ПОЛІЩУК, Н. В. (2025) · АКАДЕМІЧНІ СТУДІЇ СЕРІЯ «ПЕДАГОГІКА» · doi:10.52726/as.pedagogy/2025.4.10
Generatív mesterségesintelligencia-eszközök az egyetemi kommunikációoktatásban
Kuttner, Á., & Kárpáti, A. (2026) · Médiakutató · doi:10.55395/mk.2026.1.2
Lehren und Lernen mit ChatGPT im Statistikpraktikum
Ebner, B., Höfer, S., & Pabst, D. (2026) · Mitteilungen der Deutschen Mathematiker-Vereinigung · doi:10.1515/dmvm-2026-0009
Wu, G., & Qiuxiang, H. (2026) · SABIR INTERNATIONAL BULLETIN OF APPLIED LINGUISTICS · doi:10.25115/ibal.v10.10922
ÇELİKTÜRK, H., & Başar, S. (2026) · Anadolu Üniversitesi Sosyal Bilimler Dergisi · doi:10.18037/ausbd.1749050
Mualip, N., Jeplin, R. G., & Hashim, J. (2026) · Research Square · doi:10.21203/rs.3.rs-9181082/v1
Kanwal, N., Rodrigues, R. B., & Khan, P. D. N. (2026) · Voyage Journal of Educational Studies · doi:10.58622/e391v673
Ordeñana, A. I. G., & Sánchez, V. H. R. (2026) · Revista Científica Multidisciplinar G-nerando · doi:10.66473/rcmg.v7i1.738
Derlina, D., Prahanı, B. K., Ormancı, Ü., Saphıra, H. V., Ramadhani, I., & Riani, R. (2026) · International Conference on Innovation in Education, Science and Culture (ICIESC) · doi:10.4108/eai.16-9-2025.2361132
Dennisson, J., & Ross, G. (2025) · JALT PanSIG Journal · doi:10.37546/jaltpansigj11.1-6
Yoo, S.-B., & Bae, H.-Y. (2026) · Journal of Digital Contents Society · doi:10.9728/dcs.2026.27.3.751
AI literacy beyond tools: Safeguarding human judgment in research and higher education
Weng, Y.-S. (2026) · AI & Future Society · doi:10.63802/afs.v2.i1.207
Yundayani, A., Mukminin, A., & Alghadari, F. (2026) · Высшее образование в России · doi:10.31992/0869-3617-2026-35-3-149-168
ChatGPT as a Training Coach: Pattern-Specific Outcomes in Build America Buy America Act Training
Shaheen, Q. H., Erfani, A., Mohammed, M. Y., Zhen, D., & Cui, Q. (2026) · Journal of Management in Engineering · doi:10.1061/jmenea.meeng-7190
Bekor, T. F., & Pandapotan, D. (2026) · Journal of Community Service and Engagement · doi:10.65310/spy8t590
Awashreh, R. A., Ghunaimi, H. A., & Ghenaimi, S. A. (2026) · Social Sciences & Humanities Open · doi:10.1016/j.ssaho.2026.102788
Metadata
- Authors
- Wang, J., & Fan, W.
- Affiliations
- Hangzhou Normal University (CN)
- Corresponding author
- Fan, Wenxiang
- Venue
- Humanities and Social Sciences Communications
- Published
- 2025-05-06
- Volume
- 12
- Issue
- 1
- ISSN
- 2055-1045 (ISSN-L)
- 2662-9992
- OpenAlex
- W4410129808
- Semantic Scholar
- 45606f671465c499d9299234e8c5a3f444556f1b
- Cited by
- 322
- References
- 134
- Citations by year
- 2026: 217 · 2025: 105
- FWCI
- 48.98
- Type
- journal article
- Language
- en
- Status
- Retracted
- Update notices
- retraction (2026-04-22)
- Open access
- gold
- doi.org
- Licence
- Locations
- Humanities and Social Sciences Communications (publishedVersion, cc-by-nc-nd)
- RePEc: Research Papers in Economics (submittedVersion)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Funders
- Major Project of Humanities and Social Sciences Key Research Base of the Ministry of Educatio (22JJD880023)
- University-Industry Collaborative Education Program (240904171111322)
- Hangzhou Social Science Routine Planning Project (Z24JC014)
- Hangzhou Normal University’s 2023 Teaching Construction and Reform Project “Exploration and Practice of Generative Artificial Intelligence to Promote University Teaching Reform”
- Topics
- Social Sciences › Education › Artificial Intelligence in Education
- Physical Sciences › Computer Science › Artificial Intelligence › Intelligent Tutoring Systems and Adaptive Learning
- Health Sciences › Medicine › Health Informatics › Artificial Intelligence in Healthcare and Education
- Keywords
- ChatGPT, generative AI in education, learning performance, perceptions of learning, higher-order thinking skills, AI-assisted learning, problem-based learning, Bloom's taxonomy, instructional scaffolding, intelligent tutoring systems, moderating effect
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 2026-10-06, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).
A second model labelled each claim on 2026-10-06 (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
