Reimagining Attendance: Faculty Perspectives on Student Attendance Systems Powered by Facial Recognition Technology

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

Reimagining Attendance: Faculty Perspectives on Student Attendance Systems Powered by Facial Recognition Technology

Overall, it addresses new technology regarding students' attendance. However, it has some comments that need minor revisions to be clarified and corrected before its publication in its current form.

Comments in the article which need clarifications, corrections, rewrites and/or additional information:

• Lines 116, 122: facial recognition technology (FRT). No need to rewrite the full sentence, only in the 1^st mention, after that use Acronym FRT only.
• Line 126-127: Authors mentioned “Hence, this study addresses this gap by investigating faculty perceptions toward FRT in Saudi Arabia”: this can't be generalized for Saudi Arabia, as the study was conducted in only a single private Saudi medical college without probability sampling of college selection: rewrite and correct.
• Stratification was based on academic rank and disciplinary affiliation (basic sciences vs. clinical sciences) to ensure representative participation across faculty subgroups: demonstrate faculty subgroups
• Line 202-205: add reference for TAM tool used in the study. Also, demonstrate details of the scoring system to be classified as good or poor acceptance
• Line 218-220: Revise statistical tests used, rewrite (for example, multiple regression modeling identified significant predictors of technology acceptance while controlling demographic variables). Also, demonstrate dependent and independent outcomes.
• Line 295: authors mentioned “poor acceptance based on the comprehensive attitude scale: where is this scale?

 

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors

Well done and nice work.

All the comments and suggestions are intended to increase the manuscript clarity and scientific soundness. Despite the fact that the topic is not new and there are no breakthroughs, the study context (the different cultures and traditions in Saudi Arabia) adds more importance to it. The study can benefit from discussing this context. As a suggestion, expand the study or conduct another one exploring students' perceptions of the use of FRT in the cultural context rather than technology use.

1. Brief Summary

This mixed-methods study examines faculty perceptions of Facial Recognition Technology (FRT) for tracking student attendance at a private Saudi medical college. The authors surveyed 112 faculty members (quantitative phase) and conducted focus groups/interviews with 26 participants (qualitative phase). Key findings reveal a nearly balanced divide: 51.8% of faculty demonstrated good acceptance of FRT, while 48.2% showed poor acceptance. Faculty rated the technology highly for perceived ease of use (85.7%) and effectiveness (75%), yet over half (55.3%) expressed significant privacy concerns. Qualitative analysis identified six major themes, including initial reactions, communication/training issues, efficiency versus challenges, privacy/security/ethical concerns, and suggestions for improvement.

2. Major Concerns

Concern 1: Study objective.

There is an inconsistency in the study objective.

In the abstract, it is:  Explored faculty perceptions of using Facial Recognition Technology (FRT) for tracking medical student attendance at a private Saudi medical college.

In the introduction, it is: "Investigating faculty perceptions toward FRT in Saudi Arabia."

Recommendation: unifies the objective across all parts of the study.

Concern 2: Use of abbreviations: Abbreviations should be reported at first appearance and then used.

Concern 3: Methodological Flaws in Sample Size Calculation and Reporting

Lines 186-190: The author's state: "Sample size estimation using G*Power 3.1.9.7 indicated that a minimum of 108 participants out of 164 faculties was required to detect a medium effect size (Cohen's d = 0.5) with 80% statistical power at a significance level of α = 0.05."

Even with using G*Power 3.1.9.7, the provided details do not match the number reported by the method. Please reconsider the sample size recalculation.

Issue: For a survey study measuring proportions/percentages, Cohen's d is inappropriate as it is designed for mean differences between two groups. A proper sample size calculation for a proportion should use formulas for margin of error or precision. This fundamental error undermines the validity of the sample size justification.

Recommendation: Recalculate using appropriate methods (e.g., sample size for proportion with desired precision) or justify why Cohen's d was applied to a single-group descriptive study.

Concern 4: In the study, convenience sampling is a non-probability, low-cost method that relies on easily accessible participants, resulting in high bias and limited generalizability. Conversely, calculating sample size for a specific population uses statistical formulas to ensure representative, randomized sampling, thereby minimizing bias and enabling accurate generalization to the target population.

Recommendation: Please clarify which method was used.

Concern 5: Major Discrepancies in Quantitative Data Reporting

Lines 274-287 (Table 1): Academic rank data contains serious inconsistencies:

• Row sums for academic background: clinical (n=60, 53.6%), Basic Science (n=38, 33.9). Total = 98, and percentages sum to 87.5%.
• The total number should be 122 and the percentage 100%
• Row sums for academic ranks: Instructor (n=16, 27.6%), Lecturer (n=16, 27.6%), Assistant professor (n=34, 30.6%), Associated professor (n=16, 27.6%), Professor (n=30, 26.8%). Total = 112, but percentages sum to >140%.
• The percentages (27.6% + 27.6% + 30.6% + 27.6% + 26.8% = 140.2%) suggest the denominator varies across rows, which is statistically invalid.
• Within-row percentages for "Good Acceptance" and "Poor Acceptance" under Instructor: "12 (20.4%)" and "4 (10.4%)" — these denominators appear to be different (one ~59, one ~38), which is not explained.
• The percentages should be: "Good Acceptance 12 (75%)" and "Poor Acceptance 4 (25%)".
• Please revise the other ranks.

Recommendation: Verify all percentages and present a single, consistent denominator for each row. Report actual n and column percentage with a clear indication of what constitutes 100%.

Concern 6: Statistical Analysis Errors

Line 297-298: "academic rank reveals a statistically significant association (p = 0.025)"

Issue: No test statistic is reported (e.g., χ² value, degrees of freedom). Additionally, with 5 academic ranks and 2 acceptance categories, Fisher's exact test or the chi-square test would be appropriate. However, the expected cell counts are likely below 5 given the small n in some rank categories (e.g., Associate Professor, with only 16 total, divided into 6 and 10). This violates chi-square assumptions.

Line 272: "p value" column appears in Table 1, but the statistical test used is never specified in methods or footnotes.

Recommendation: Report the exact statistical test used, test statistic with degrees of freedom, and confirm assumptions were met or use appropriate alternatives (Fisher's exact). Consider merging smaller rank categories.

Concern 7: Critical Missing Information in Methods

General Issue: The study claims a "mixed-methods approach" and "sequential explanatory design" but fails to provide essential methodological details:

• Quantitative instrument: No description of the survey instrument, item sources, validation process, pilot testing, or reliability coefficients (Cronbach's alpha for subscales).
• Qualitative methodology: No interview/focus group protocol, question guides, duration of sessions, or saturation assessment.
• Analysis procedures: No description of quantitative analysis plan beyond mentioning G*Power. The qualitative analysis cites Braun & Clarke but provides no coding framework, inter-coder reliability, or procedures for researcher triangulation.
• Missing data handling: Not addressed.

Recommendation: Add a detailed "Measures" subsection describing instrument development, validation, and reliability. Expand qualitative methods to include protocol details and rigor strategies.

Concern 8: Ethical Concern Regarding Dual Roles

Lines 570-572: "Informed consent was obtained from all participants, with particular attention to confidentiality protections... participants were assured that responses would not be shared with institutional administrators."

Issue: Given that the study was conducted at the authors' own institution (Ibn Sina National College), and some authors appear to be faculty members there, there is potential for coercion or perceived obligation to participate. The manuscript does not address how this conflict was managed, whether an independent researcher collected data, or whether participants were informed of the authors' institutional positions.

Recommendation: Add a statement addressing potential conflict of interest from conducting research at one's own institution and describe safeguards implemented (e.g., third-party data collection, anonymized response handling).

Concern 9: Poor Flow and Narration Issues

The captions of the tables and figures should be complete enough to stand alone and describe their contents.

Recommendation: Repair figure and table formatting. Restructure the results section for logical flow. Proofread thoroughly.

Concern 10: Overinterpretation of Findings

Line 484: "This aligns with the logical flow of Kubler Ross change management cycle, progressing from denial and resistance stages till reaching the final stage of acceptance."

Issue: The Kubler-Ross change curve (originally for grief) is inappropriately applied here without any longitudinal data tracking faculty through stages. Cross-sectional data cannot demonstrate progression through change stages.

Line 493-494: "lower than [22] 68% Australian public opposition [17] but higher than Guillén-Gamez et al. [34] 42% EU acceptance"

Issue: Comparing opposition rates across different populations (Australian public vs. Saudi faculty vs. EU general population) with different survey instruments, contexts, and timeframes is methodologically questionable.

Recommendation: Remove the Kubler-Ross reference or reframe as a speculative hypothesis. Acknowledge limitations of cross-contextual comparisons.

Concern 11: Data Reporting Completeness

Issue: The manuscript reports means and standard deviations for the attitude scale (mean 42.5 ± 8.27, range 11-55, median 44) but does not report:

• Skewness/kurtosis to justify parametric statistics
• Response rate (112 out of 164 eligible = 68.3% — acceptable but unreported)
• Comparison between respondents and non-respondents
• Missing item analysis

Recommendation: Report response rate and conduct non-response bias analysis if possible.

3. Minor Comments

1. Line 16: "Tayseer Mansourt" has a typo ("t" at the end) — should be "Mansour".
2. Line 56-58: Not related, rephrase or delete it.
3. Line 69-70: Please add reference (s).
4. Line 91-94: Better to be moved to line 78
5. Line 107-110: Not related, please rephrase or delete it.
6. Line 111-114: Please move it to line 126.
7. Line 115-120: Please move it to line 105.
8. Line 115-120: Please move it to 105.
9. Line 183: G*Power uses the minimum sample size needed to detect an effect, regardless of population size. It is the correct tool for power-based sample size. However, it is not sufficient on its own for small populations like 164. It is better to use population-based sample size formulas.  

Please justify using G*Power rather than population-based sample-size formulas or total coverage.

1. Line 184: Please reconsider the sample size recalculation, as the provided details do not match the number reported by the method.
2. Line 205: According to the narration, the survey was conducted in English, while the interviews were conducted in the language of choice or by preference. Please justify the difference in methodology.
3. Line 205: Some sociodemographic data. Please be specific

13. Line 207:Did the data from the pilot study include in the main analysis? Is there any modification applied after the piloting?
14. Line 226-28:According to the narration, the survey was conducted in English, while the interviews were conducted in the language of choice or by preference. Please justify the difference in methodology.

1. Line 272: Table 1 uses "P value," but statistical tests are not specified — add footnote.
2. Line 286: "Associated professor" should be "Associate professor".
3. Line 294: Better to start a sentence with the full name, not an abbreviation.
4. Line 295: "3.6% (n = 4) had not heard about FRT implementation" — these participants should likely have been excluded from analysis as they lack familiarity with the technology.
5. Lines 297-298: "professors showing markedly higher good face recognition rates" — unclear whether "face recognition rates" refers to technology acceptance or system performance.
6. Lines 309-310: Table 2 formatting is severely corrupted with nested row span tags, making interpretation difficult.

Suggestion: Use abbreviations instead of full names to reduce the table size.

1. Lines 346-352: The study uses staff members (112), and there is no reporting on included students. How and why does student perception exist?
2. Line 400: "overpass the system" should be "bypass the system".
3. Line 424: "Faculty reported students' manipulation to overpass" — grammar error.
4. Line 453: Table 3 appears to duplicate Table 4 content — consolidate or clarify distinction.
5. Line 463: explored the integration of AI-powered facial recognition (FR) technology into tracking. Please unify the use of names and terminology.
6. Line 482: Reference 21. The topic of facial recognition technology for attendance in medical school or colleges is not new; there are many references. Please refer to the studies conducted in the same context rather than companies.
7. Line 487-9: Missing/ unclear sentence.
8. Line 487-9: FR's. Please unify the use of names and terminology.
9. Line 509-11: The study uses staff members (112), and there is no reporting on the included students. How and why does student perception exist?
10. Line 522-24: The cultural and contextual dimensions of the current study add more to the knowledge about FRT. Please expand the discussion and interpretation of data. One major finding is a higher acceptance rate among females, along with concerns among faculty who cover their faces. Please discuss.
11. Line 560: Reference format inconsistency — some DOIs present, others missing.
12. Line 574: "BERA Guidelines" — spell out British Educational Research Association on first use.

4. Section-by-Section Feedback

Title

Assessment: Acceptable but generic. Suggests a broader scope than a single-institution study.
Suggestion: Consider "Faculty Perspectives on Facial Recognition Attendance Systems at a Saudi Medical College: A Mixed-Methods Study"

Abstract

Assessment: Adequately summarizes methods and key findings but overstates significance.
Suggestion: Remove "reimagining" framing; clarify that findings are from one institution.

Introduction

Assessment: Well-referenced but overly long. Some claims lack support (e.g., "40-60% workload reduction" — does the reference [6,12,5] include a non-peer-reviewed source?).
Suggestion: Shorten by 30%; ensure all substantive claims have peer-reviewed support.

Methods

Assessment: Major deficiencies as detailed in Concern 4. Missing instrument description, validation, reliability, and qualitative protocol.
Suggestion: Complete rewrite of sections 2.3-2.5 with full methodological detail.

Results

Assessment: Severely compromised by data reporting errors (Concerns 2, 3, 6). Tables are corrupted; percentages are inconsistent.
Suggestion: Complete revision with verified data and properly formatted tables. Remove or repair all figures.

Discussion

Assessment: Reasonably thoughtful interpretation but overreaches with Kubler-Ross analogy and cross-population comparisons. Missing key limitations section.
Suggestion: Add an explicit limitations subsection addressing single-site design, potential coercion bias, and cross-sectional nature.

Conclusion

Assessment: Appropriate length but overly definitive given methodological issues.
Suggestion: Soften claims; emphasize need for replication.

References

Assessment: 

• References are listed in different styles with missing data.
• Mix of peer-reviewed and non-peer-reviewed sources (e.g., reference 29 from the Daon commercial website).
• Reference 18 format error.

Suggestion: Replace commercial/non-academic sources with peer-reviewed equivalents where possible.

Minor

1. Proofread for grammar and typographical errors throughout.

1. Standardize reference format (check journal requirements for DOIs).
2. Spell out all acronyms on first use (BERA, etc.).

Comments on the Quality of English Language

Although some words and sentences are missing to complete the meanings, the quality of the English language is good.

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Reviewer 3 Report

Comments and Suggestions for Authors

The article Reimagining Attendance: Faculty Perspectives on Student Attendance Systems Powered by Facial Recognition Technology by Eltarhouny et. al describes a study in which faculty at a medical education institution underwent surveys and interviews regarding the acceptability and use of facial recognition technology to assess attendance for class. Overall, the study is well designed and well presented - showing the various opinions and themes illicited from faculty. However, the discussion and conclusion seem to deviate from the overall theme of the article - the authors make numerous mentions regarding theories of behavior and attitude change. This gives the idea that the authors inherently believe that FRT is the solution for attendance tracking at medical institutions and that faculty will 'come around' to accepting it eventually. I would strongly recommend restructuring it to focus on the results of the mixed methods study - commenting on thoughts/feelings surrounding FRT and not trying to push that FRT will eventually be accepted. See attached PDF for more comments. 

Comments for author File: Comments.pdf

Author Response

Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

Comments and Suggestions for Authors

Significant improvement from prior draft with focus on the initial perceptions as opposed to focusing on theories of change. Some minor edits to improve syntax and English. See attached

Comments for author File: Comments.pdf

Author Response

Dear Reviewer,

We sincerely thank you for your thorough and constructive review of our manuscript. We have carefully addressed each of your points and have revised the manuscript accordingly.

Point-by-Point Responses

Comment 1 (Page 2 — Introduction)

Reviewer: "Effective"

Response: We meant the Affective Domain of competencies, so no changes were required.

Comment 2 (Page 2 — Introduction)

Reviewer: "Need to spell out Facial Recognition Technology"

Response: We agree with the reviewer. Although the full term is introduced in the manuscript title and abstract, the abbreviation appeared without expansion at its first use in the body of the Introduction. We have corrected this by spelling out the full term at this point.

Comment 3 (Page 3 — Introduction)

Reviewer: "Start with 'AI and FRT' as you have already described them"

Response: The sentence has been revised.

Comment 4 (Page 3 — Introduction)

Reviewer: "TAM 'which' is..."

Response: The sentence has been revised.

Comment 5 (Page 4 — Setting and Participants)

Reviewer: "Pharm D --> Pharmacy"

Response: "Pharm D" has been changed to "Pharmacy" in the Setting and Participants section.

Comment 6 (Page 4 — Setting and Participants)

Reviewer: "I thought the study was only faculty? What data are you collecting from students?"

Response: The reviewer raises an important and valid concern. The study does indeed focus exclusively on faculty perceptions, and no data were collected from students. The reference to "1200 students" was intended solely to describe the institutional context and the scale of the FRT system's deployment not to imply student participation in data collection. We recognize that the original phrasing was misleading and have revised the sentence to eliminate any ambiguity.

Correction made: The sentence has been revised from: "Data was collected from December 2024 to April 2025 from a pool of a total of 1200 students and 164 staff."

To: "The institution serves a student body of approximately 1200 students. For this study, data were collected from December 2024 to April 2025 exclusively from the faculty pool of 164 staff members, all of whom interact directly with the FRT system in their teaching roles."

Comment 7 (Page 6 — Reflexivity)

Reviewer: "This should be two separate sentences with the second sentence starting with 'Authors SE & SA...' The second sentence should then be converted into the past tense, not the present tense as it currently is."

Response: We thank the reviewer for this grammatical and structural correction. The text has been revised.

Comment 8 (Page 6 — Reflexivity)

Reviewer: "This sentence is an important Limitation. While it is important, it belongs in the Limitation section, not here."

Response: We fully agree with the reviewer. This statement, which acknowledges the potential influence of researcher positionality on the study's findings, constitutes a methodological limitation rather than a reflexivity statement. We have removed it from the Reflexivity subsection and incorporated it into the Limitations section of the Discussion, where it is more appropriately situated.

Correction made: The sentence has been removed from the Reflexivity section and added to the Limitations section with the following revised wording: Furthermore, as insider-researchers conducting a study within their own institutional context, the authors acknowledge that personal and professional positions may have inevitably influenced data interpretation. While peer debriefing and negative case analysis were employed to mitigate this, residual positionality bias cannot be entirely excluded.

Comment 9 (Page 10 — Results, Section 3.2)

Reviewer: "Is there a reason these two paragraphs are in quotations?"

Response: We apologize for this formatting error. The quotation marks enclosing both paragraphs in Section 3.2 have been removed.

We trust that these revisions have satisfactorily addressed all the reviewer's concerns. We remain grateful for the opportunity to improve our manuscript and look forward to your further consideration.

Sincerely,

The Authors

Shereen Eltarhouny, Shayma Aljedaani, Rania Alkhadragy, Tayseer Mansour

Author Response File: Author Response.docx