Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 89 of the 89 references it cites were found in a registry or an index; 33 of the 91 sources its claims cite were checked for support.

AI teaching innovation behavior among college teachers: a structural equation modeling analysis based on the AI-TPACK framework, teaching self-efficacy, professional identity, and AI literacy

Open access
Bai, Xue, Hsieh, Tsung-Shun · 2026 · Frontiers in Psychology

Abstract

Background: As generative artificial intelligence (GenAI) increasingly enters higher education, understanding how teachers integrate GenAI into innovative teaching is critical. Prior research has mainly focused on intentions to use AI tools. Guided by the AI-Technological Pedagogical Content Knowledge (AI-TPACK) framework, Social Cognitive Theory, and professional identity theory, this study explored the relationships surrounding GenAI-supported teaching innovation behavior, with a focus on teaching self-efficacy, professional identity, and AI literacy. Methods: A total of 898 Chinese university teachers from eight comprehensive universities completed standardized scales assessing AI-TPACK dimensions, teaching self-efficacy, professional identity, AI literacy, and AI teaching innovation behavior. Data were analyzed via structural equation modeling and multi-group analysis. Results: The results showed that among the AI-TPACK dimensions, all except AI technological knowledge and integrative knowledge were significantly associated with AI teaching innovation behavior. Professional identity and AI literacy partially mediated the relationship between competence structures and innovative behavior, whereas the mediating effect of teaching self-efficacy was only partially supported. Further multi-group analysis showed that the relationship between professional identity and innovative behavior was stronger among teachers with lower reported frequencies of GenAI tool use. Discussion: These findings provide empirical evidence on the associations of competence- and psychology-related factors with GenAI-supported teaching innovation. The study highlights the roles of professional identity, AI literacy, and teaching self-efficacy in promoting innovative teaching and offers practical implications for faculty development and AI integration in higher education.

Overview

CoChat Summary

TL;DR

This survey study of 898 Chinese university teachers found that AI-TPACK dimensions were associated with AI teaching innovation behavior unevenly: AI technological knowledge (AI-TK) and overall AI-TPACK were not significant direct predictors, while the other dimensions were. Professional identity and AI literacy were important direct correlates and mediators of AI teaching innovation behavior; teaching self-efficacy was supported more modestly and only partially.

Key findings

  • The sample was 898 teachers from eight comprehensive universities in China; the model explained 62.50% of the variance in AI teaching innovation behavior (AI-TIB), with R² = 0.625.
  • Among AI-TPACK dimensions, all except AI-TK and AI-TPACK were significantly and positively associated with AI-TIB; AI-TK and AI-TPACK had no significant direct association with AI-TIB.
  • Professional identity (PI), AI literacy (AIL), and teaching self-efficacy (TSE) were all positively associated with AI-TIB, and the authors note the significant effects were generally small in magnitude.
  • Mediation results showed PI and AIL partially mediated the competence–behavior link, while TSE was only partially supported; for AI-TK → TSE → AI-TIB, the indirect effect was not supported.

Methods

  • Cross-sectional, self-report survey of 898 college teachers recruited by convenience sampling from eight Chinese universities between February and March 2026.
  • Measures: standardized Likert scales for AI-TPACK (42 items, 7 dimensions), TSE (8 retained items from a 12-item scale), PI (19 retained items from a 25-item scale), AIL (20 items), and AI-TIB (7 items).
  • Analysis: PLS-SEM with bootstrapping (5,000 resamples), mediation analysis, and multi-group analysis comparing high-frequency vs low-frequency AI tool users.

Limitations

  • The paper states the design was cross-sectional, limiting temporal or causal conclusions; reciprocal relationships among AIL, PI, and TSE are possible.
  • The study used convenience sampling from a limited number of Chinese universities, with a relatively high proportion of female and younger teachers, so generalizability is limited; the paper explicitly notes the findings should not be generalized directly to all higher education contexts.
  • The study relied primarily on self-report data, which may introduce social desirability or common method bias, although the authors report tests did not indicate severe common method bias.
  • Evident from the text: the AI-TIB scale used general innovation items with scenario instruction for GenAI, so some responses may not have been fully AI-specific, and the high/low AI-use grouping was relatively crude.

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

Integrity signals

Registry

Crossref · 10.3389/fpsyg.2026.1834827

Cites

89 of 89 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Verification

Claims the paper makes

25 Valid · 10 Not valid · 43 Not checked · 192 Own results and interpretations

137 more claims · 13 with their cited sources checked

Create a free account to see all 270 claims and every source verdict.

See all claims — free account
  • Probably not a substantial claim. 72
  • It may state two separate claims as one. 77
  • Borderline: one of the tests for a substantial claim is unclear. 1

Create a free account to see these claims.

Citations

Cites

Show all 87 references

Research works printed but not linked: 8

  1. Byun, 2012, The structural relations among hotel employees' job attitudes, customer orientation, and job performance, Int. J. Tour. Hosp. Res., 26, 357
  2. Hair, 2017, Advanced Issues in Partial Least Squares Structural Equation Modeling
  3. Joo, 2018, Factors influencing preservice teachers’ intention to use technology: TPACK, teacher self-efficacy, and technology acceptance model, J. Educ. Technol. Soc., 21, 48
  4. Kim, 2013, The impact of organization justice on social capital and knowledge sharing, Korean J. Bus. Adm., 26, 1123
  5. Leguina, 2015, Introduction to Mediation, Moderation, and Conditional Process Analysis. A Regression-Based Approach
  6. Lin, 2021, Structural relationship between organizational justice, organizational trust, and knowledge sharing and innovative behavior: focus on professors from Chinese sport universities, J. Phys. Educ. Sport, 21, 882, 10.7752/jpes.2021.02110 · DOI
  7. Liu-Schuppener, 2023, Artificial Intelligence and Digitalization in China's Education System. A Systematic Analysis of the Policy Framework and Underlying Strategies
  8. Rahman, 2022, Significant effect of the TPACK framework on teachers’ self-efficacy: a systematic literature review, J. Educ. Res., 40, 63

Metadata

Authors
Bai, X., & Hsieh, T.-S.
Affiliations
  • Department of Education, Xinzhou Normal University
  • Assumption University (TH)
Corresponding author
Bai, Xue
Venue
Frontiers in Psychology
Published
2026-09-07
Volume
17
Pages
1834827-1834827
ISSN
1664-1078 (ISSN-L)
OpenAlex
W7211909479
Cited by
0
References
87
FWCI
0.00
Type
journal article
Language
en
Open access
Locations
  • Frontiers in Psychology (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • PubMed Central (publishedVersion, cc-by)
Topics
  • Social Sciences › Education › Artificial Intelligence in Education
  • Social Sciences › Education › Teacher Education and Leadership Studies
  • Social Sciences › Education › Digital literacy in education
Keywords
generative AI, teacher self-efficacy, professional identity, AI literacy, college teachers, higher education, AI in teaching, faculty development, mediating 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.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

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).

Cited sources are checked with a language model’s help (gpt-6-sol).

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: AI teaching innovation behavior among college teachers: a structural equation modeling analysis based on the AI-TPACK framework, teaching self-efficacy, professional identity, and AI literacy. https://app.cochat.ai/papers/10.3389/fpsyg.2026.1834827
The paper
Bai, X., & Hsieh, T.-S. (2026). AI teaching innovation behavior among college teachers: a structural equation modeling analysis based on the AI-TPACK framework, teaching self-efficacy, professional identity, and AI literacy. Frontiers in Psychology. https://doi.org/10.3389/fpsyg.2026.1834827

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.