arxiv: 2605.00343 · v1 · submitted 2026-05-01 · 💻 cs.CY · cs.AI

Recognition: unknown

AI Adoption Among Teachers: Insights on Concerns, Support, Confidence, and Attitudes

Vanessa B. Sibug , Maria Anna D. Cruz , Vicky P. Vital , Juvy C. Grume , Almer B. Gamboa , Emerson Q. Fernando , Lloyd D. Feliciano , Jordan L. Salenga

show 1 more author

John Paul P. Miranda

Authors on Pith no claims yet

Pith reviewed 2026-05-09 19:09 UTC · model grok-4.3

classification 💻 cs.CY cs.AI

keywords AI adoptioninstitutional supportteacher confidenceteacher attitudesmediation analysisAI in educationteacher concernseducational technology

0 comments

The pith

Institutional support improves teachers' attitudes toward AI by building their confidence, while concerns do not moderate the relationship.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper examines how institutional support, teacher confidence, and concerns shape AI tool adoption in education settings. Analysis of survey responses from 260 teachers shows that support predicts higher confidence and more positive attitudes. Concerns fail to moderate those connections in the regression models. Mediation testing then reveals that confidence accounts for the entire link from support to attitudes. A sympathetic reader cares because the results point to concrete institutional actions that can raise educator readiness for AI without relying on concern reduction alone.

Core claim

Moderated multiple regression on data from 260 Philippine teachers found that institutional support significantly predicted both teacher confidence and attitudes toward AI adoption, yet teacher concerns did not significantly moderate either relationship. A follow-up mediation analysis tested whether confidence carries the effect of support onto attitudes. The indirect effect was significant by Sobel test, and the direct effect of support on attitudes dropped to non-significance once confidence entered the model, establishing full mediation. This shows that institutional support improves teacher attitudes by increasing their confidence.

What carries the argument

Mediation analysis with the Sobel test establishing full mediation by teacher confidence between institutional support and attitudes toward AI.

If this is right

Institutions should deliver structured and ongoing support to strengthen teacher confidence and thereby improve attitudes toward AI.
Professional development programs, mentoring, and AI integration into teacher education can raise readiness for effective adoption.
Reducing teacher concerns in isolation is unlikely to shift attitudes without parallel efforts to build confidence.
Support mechanisms that target confidence directly will produce fuller gains in AI adoption outcomes.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

Interventions could be designed to measure and boost specific confidence-building activities rather than broad concern-reduction campaigns.
The same mediation pattern may appear in teacher populations outside the Philippines, suggesting value in testing the model in additional countries.
Longitudinal tracking of teachers who receive increased support could confirm whether confidence gains produce sustained attitude and behavior changes over time.

Load-bearing premise

The survey scales accurately capture institutional support, teacher confidence, concerns, and attitudes, and the 260-teacher sample is representative enough for the regression and mediation results to generalize.

What would settle it

A replication with an independent sample or alternative validated scales that finds either significant moderation by concerns or a non-significant indirect effect through confidence via the Sobel test would falsify the mediation claim.

read the original abstract

The study examines the adoption of artificial intelligence (AI) tools in education by analyzing the roles of institutional support, teacher confidence, and teacher concerns. It aims to determine whether teacher concerns moderate the relationship between institutional support and two outcomes: teacher confidence and attitudes toward AI adoption. The sample included 260 teachers from the Philippines. Composite scores were calculated for institutional support, confidence, concerns, and attitudes. Moderated multiple regression analysis showed that institutional support significantly predicted both teacher confidence and attitudes toward AI. However, teacher concerns did not significantly moderate these relationships. A follow-up mediation analysis tested whether confidence explains the effect of institutional support on attitudes. Results showed full mediation. The indirect effect was significant based on the Sobel test, and the direct effect became non-significant when confidence was included in the model. This shows that institutional support improves teacher attitudes by increasing their confidence. The study recommends that institutions provide structured and ongoing support to strengthen teacher confidence. Professional development, mentoring, and AI integration in teacher education programs can increase readiness and support effective AI adoption.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

Standard cross-sectional survey of 260 Philippine teachers finds full mediation of institutional support on AI attitudes via confidence, but the causal wording overreaches the design.

read the letter

This paper reports that institutional support relates to better teacher attitudes toward AI entirely through higher confidence levels, based on a survey of 260 teachers in the Philippines. Concerns did not moderate the links. The mediation held via Sobel test and the direct path dropped to non-significant once confidence was added. That is the core empirical result. The work adds a data point from a non-Western context, which is useful since most AI adoption studies cluster in high-income settings. The authors apply ordinary moderated regression and mediation steps to composite scores, keep the analysis simple, and draw straightforward policy suggestions around professional development and mentoring. Those elements are fine as far as they go. The main limitation is the leap to causal language. The abstract states that support improves attitudes by raising confidence, yet the data are single-timepoint self-reports with no longitudinal follow-up, randomization, or instruments. This shows statistical accounting, not that support causes the change. Sample details, scale validation, and assumption checks are thin in the abstract, though the full text may fill some gaps. Generalizability beyond the Philippine sample is also unclear. The paper suits education-technology researchers who track adoption factors in specific regions and might inform local training programs. It is not methodologically novel, but the finding is concrete enough to warrant checking. I would send it to peer review with the expectation that referees will ask for tighter wording on causation and more transparency on measures.

Referee Report

2 major / 1 minor

Summary. The paper reports results from a survey of 260 teachers in the Philippines examining relationships among institutional support, teacher concerns, confidence, and attitudes toward AI adoption. Moderated multiple regression indicates that institutional support significantly predicts both confidence and attitudes, but concerns do not moderate these links. A follow-up mediation analysis using the Sobel test finds full mediation, with confidence accounting for the support-attitudes association (direct effect becomes non-significant). The abstract concludes that institutional support improves attitudes by increasing confidence and recommends structured institutional support and professional development.

Significance. If the statistical patterns are robust, the study supplies context-specific empirical data on AI adoption factors in Philippine education, highlighting a potential pathway through teacher confidence. This could usefully inform institutional policies on support mechanisms, even if limited to associational evidence.

major comments (2)

[Abstract] Abstract: The claim that the results 'show that institutional support improves teacher attitudes by increasing their confidence' uses causal language unsupported by the design. The data are cross-sectional composite scores from a single-timepoint survey analyzed with standard moderated regression and Sobel mediation; no longitudinal, experimental, or instrumental-variable evidence establishes temporal precedence or rules out reverse causation or confounding.
[Methods/Results] Methods/Results (inferred from abstract description): The manuscript omits essential details required to assess the central claims, including sampling procedure and response rate for the 260 teachers, item content and derivation of the four composite scores, scale reliabilities, regression assumption diagnostics, and effect sizes or confidence intervals for the moderated and mediated effects. These gaps directly affect evaluation of the full-mediation result and the non-significant moderation findings.

minor comments (1)

[Abstract] Abstract: Consider adding brief mention of the specific mediation procedure steps or software used and the magnitude of the indirect effect to strengthen the summary of the Sobel-test result.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the detailed and constructive feedback. We agree that the abstract employs causal language that exceeds what the cross-sectional design can support, and we will revise the wording throughout the manuscript to reflect associational findings only. We also acknowledge the need for greater transparency in the Methods and Results sections and will add all requested details in the revision.

read point-by-point responses

Referee: [Abstract] Abstract: The claim that the results 'show that institutional support improves teacher attitudes by increasing their confidence' uses causal language unsupported by the design. The data are cross-sectional composite scores from a single-timepoint survey analyzed with standard moderated regression and Sobel mediation; no longitudinal, experimental, or instrumental-variable evidence establishes temporal precedence or rules out reverse causation or confounding.

Authors: We fully agree with this observation. The study is based on cross-sectional survey data, so causal claims are not warranted. In the revised manuscript we will change the abstract and all interpretive statements to use associational language (e.g., “is positively associated with” and “the association between institutional support and attitudes is fully accounted for by confidence”). We will also add an explicit limitations paragraph noting that the mediation analysis is exploratory, that reverse causation and confounding cannot be ruled out, and that longitudinal or experimental designs are required to test directional hypotheses. revision: yes
Referee: [Methods/Results] Methods/Results (inferred from abstract description): The manuscript omits essential details required to assess the central claims, including sampling procedure and response rate for the 260 teachers, item content and derivation of the four composite scores, scale reliabilities, regression assumption diagnostics, and effect sizes or confidence intervals for the moderated and mediated effects. These gaps directly affect evaluation of the full-mediation result and the non-significant moderation findings.

Authors: We apologize for these omissions. The revised manuscript will expand the Methods section to report: (1) sampling procedure (convenience sampling via online teacher networks in the Philippines) and response rate; (2) the exact items or validated scales used to form each composite score, with citations; (3) internal consistency reliabilities (Cronbach’s α) for all four measures; (4) regression diagnostics (VIF for multicollinearity, Shapiro–Wilk or Q–Q plots for normality, Breusch–Pagan for homoscedasticity); and (5) complete effect-size reporting including unstandardized and standardized coefficients, R² values, and 95 % confidence intervals for all paths in the moderated regression and Sobel mediation models. These additions will allow readers to evaluate the robustness of the full-mediation result and the non-significant moderation findings. revision: yes

Circularity Check

0 steps flagged

No circularity in empirical mediation analysis

full rationale

The paper reports statistical findings from moderated regression and mediation analysis on composite scores from a single-timepoint survey of 260 teachers. The key result (full mediation where confidence accounts for the support-attitudes link) is a direct output of the Sobel test and regression coefficients applied to the observed data. No equations, predictions, or claims reduce to definitional equivalence with inputs, no self-citations serve as load-bearing premises, and no ansatz or uniqueness theorems are invoked. The derivation chain is self-contained against the collected survey responses and standard statistical procedures.

Axiom & Free-Parameter Ledger

1 free parameters · 2 axioms · 0 invented entities

Central claims depend on fitted statistical parameters and standard assumptions of linear regression and mediation analysis applied to self-reported survey data.

free parameters (1)

beta coefficients in regression models
Estimated from the survey data of 260 teachers using moderated multiple regression and mediation models.

axioms (2)

domain assumption The relationships between institutional support, confidence, concerns, and attitudes are linear
Required for the validity of multiple regression analysis.
domain assumption The sample is sufficiently representative of the target population of teachers
Needed to generalize findings beyond the 260 respondents.

pith-pipeline@v0.9.0 · 5525 in / 1286 out tokens · 54886 ms · 2026-05-09T19:09:41.649614+00:00 · methodology

discussion (0)

Reference graph

Works this paper leans on

15 extracted references · 12 canonical work pages

[1]

Abeer S Almogren, Waleed Mugahed Al-Rahmi, and Nisar Ahmed Dahri. 2024. Exploring factors influencing the acceptance of ChatGPT in higher education: A smart education perspective. Heliyon 10, 11 (2024), e31887. https://doi.org/https: //doi.org/10.1016/j.heliyon.2024.e31887

work page doi:10.1016/j.heliyon.2024.e31887 2024
[2]

Mukhammadfoik Bakhadirov, Rena Alasgarova, and Jeyhun Rzayev. 2024. Factors Influencing Teachers’ Use of Artificial Intelligence for Instructional Purposes. IAFOR J. Educ. Technol. Educ. 12, 2 (2024), 9–32. Retrieved from https://files.eric. ed.gov/fulltext/EJ1440077.pdf

2024
[3]

Annika Boldt, Anne-Marike Schiffer, Florian Waszak, and Nick Yeung. 2019. Confidence Predictions Affect Performance Confidence and Neural Preparation in Perceptual Decision Making. Sci. Rep. 9, 1 (2019), 4031. https://doi.org/10.1038/ s41598-019-40681-9

2019
[4]

Cecilia Ka Yuk Chan and Louisa H Y Tsi. 2024. Will generative AI replace teachers in higher education? A study of teacher and student perceptions. Stud. Educ. Eval. 83, (2024), 101395. https://doi.org/https://doi.org/10.1016/j.stueduc.2024.101395

work page doi:10.1016/j.stueduc.2024.101395 2024
[5]

Chima Abimbola Eden, Onyebuchi Nneamaka Chisom, and Idowu Sulaimon Adeniyi. 2024. Integrating AI in education: Opportunities, challenges, and ethical considerations. Magna Sci. Adv. Res. Rev. 10, 2 (2024), 6–13. https://doi.org/https: //doi.org/10.30574/msarr.2024.10.2.0039

work page doi:10.30574/msarr.2024.10.2.0039 2024
[6]

Iuliana Petronela Gârdan, Magdalena Beatrice Manu, Daniel Adrian Gârdan, Lucian Drago s, Leonard Negoi ță, Carmen Adina Pa s,tiu, Ecaterina Ghi ță, and Alina Zaharia. 2025. Adopting AI in education: optimizing human resource management considering teacher perceptions. Front. Educ. 10, (2025). Retrieved from https://www.frontiersin.org/journals/education/a...

work page doi:10.3389/feduc.2025 2025
[7]

Matúš Grežo. 2021. Overconfidence and financial decision-making: a meta- analysis. Rev. Behav. Financ. 13, 3 (January 2021), 276–296. https://doi.org/10. 1108/RBF-01-2020-0020

2021
[8]

Amrita Rajkumar Harjani. 2024. Reimagining Education – Exploring the Factors Influencing Perception Towards Artificial Intelligence and Its Educational Out- come. J. Informatics Educ. Res. 4, 1 (2024), 417–425. https://doi.org/10.52783/jier. v4i1.579

work page doi:10.52783/jier 2024
[9]

Nannan Liu. 2025. Exploring the factors influencing the adoption of artificial intelligence technology by university teachers: the mediating role of confidence and AI readiness. BMC Psychol. 13, 1 (2025), 311. https://doi.org/10.1186/s40359- 025-02620-4

work page doi:10.1186/s40359- 2025
[10]

Asmahan Masry Herzallah and Rania Makaldy. 2025. Technological self-efficacy and sense of coherence: Key drivers in teachers’ AI acceptance and adoption. Comput. Educ. Artif. Intell. 8, (2025), 100377. https://doi.org/https://doi.org/10. 1016/j.caeai.2025.100377

work page arXiv 2025
[11]

Davy Tsz Kit Ng, Eagle Kai Chi Chan, and Chung Kwan Lo. 2025. Opportunities, challenges and school strategies for integrating generative AI in education. Com- put. Educ. Artif. Intell. 8, (2025), 100373. https://doi.org/https://doi.org/10.1016/j. caeai.2025.100373

work page doi:10.1016/j 2025
[12]

Dholina Inang Pambudi, Suyatno Suyatno, Wantini Wantini, Lilis Patimah, Asih Mardati, and Nunu Nurfirdaus. 2024. The mediating role of meaning in work in promoting teachers’ technology integration. Front. Educ. 9, (2024). Retrieved from https://www.frontiersin.org/journals/education/articles/10.3389/feduc.2024. 1455669

work page doi:10.3389/feduc.2024 2024
[13]

Ram Singh. 2024. Navigating Through Education 5.0 Era: Imperative Compe- tencies for Success. In Preconceptions of Policies, Strategies, and Challenges in Education 5.0 , Ali Sorayyaei Azar, Ahmad Albattat, Marco Valeri and Viana Hassan (eds.). IGI Global, 33–50. https://doi.org/10.4018/979-8-3693-3041-8.ch003

work page doi:10.4018/979-8-3693-3041-8.ch003 2024
[14]

Nga Tran, Thanh Le, Lan Nguyen, Linh Hoang, and Thuy Nguyen. 2020. Factors Influencing Technology Adoption in Vietnam’s Educational System. J. Asian Financ. Econ. Bus. 7, 10 (2020), 347–357. https://doi.org/10.13106/jafeb.2020.vol7. n10.347

work page doi:10.13106/jafeb.2020.vol7 2020
[15]

James Wyant and Jun-Hyung Baek. 2019. Re-thinking technology adoption in physical education. Curric. Stud. Heal. Phys. Educ. 10, 1 (2019), 3–17. https://doi. org/10.1080/25742981.2018.1514983 269

work page doi:10.1080/25742981.2018.1514983 2019