Review Reports - Development and Psychometric Validation of a Knowledge, Attitudes, and Practices (KAP) Questionnaire on Sustainable Diets in Taiwan

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript entitled “nutrients-4161990_ Development and Psychometric Validation of a Knowledge, Attitudes, and Practices (KAP) Questionnaire on Sustainable Diets in Taiwan” is submitted for consideration in the “Nutrition Methodology & Assessment” section of the journal. The content of this work is appropriate for this section.

The study aimed to develop and validate a Knowledge, Attitudes, and Practices (KAP) questionnaire on sustainable diets, as well as to explore sociodemographic variations and behavioral correlates among Taiwanese adults. A two-phase mixed-methods design was employed: instrument development with expert validation, followed by a cross-sectional online survey. Content validity was high (S-CVI/Ave = 0.95–0.98). Exploratory factor analysis supported a three-domain structure, and internal consistency was good for the attitude (α = 0.853) and practice (λ = 0.834) domains. Test–retest analysis demonstrated moderate to good item-level reproducibility. Results indicated that females, older adults, and individuals with higher education levels had significantly higher attitude and practice scores. Attitude emerged as the strongest predictor of sustainable dietary practices, whereas knowledge showed weaker associations. Awareness of sustainability-related concepts in dietary guidelines was significantly associated with low-carbon food choices.

Strengths :

Robust methodological design: The use of a mixed-methods approach combined with expert validation ensures strong content validity.
Comprehensive psychometric evaluation: The study examines content validity, construct validity, internal consistency, and test–retest reliability.
Practical relevance: The questionnaire has potential utility for population-level monitoring and future research on sustainable diets.
Sociodemographic analysis: Provides valuable insights into gender, age, and education differences in sustainable diet attitudes and practices.

Weaknesses :

Limited internal consistency in the knowledge domain: This may indicate that items do not uniformly capture the construct, although the authors attribute it to the multidimensional nature of sustainable diet literacy.
Potential sample bias: As the survey was conducted online, participants may be biased towards those with greater technological access or interest in sustainability. This limitation should be acknowledged and discussed.
Causal relationships not assessed: The cross-sectional design prevents causal inferences regarding determinants of sustainable dietary practices; only associations can be inferred.
Limited generalisability outside Taiwan: The questionnaire is validated specifically in Taiwanese adults, which may restrict its applicability in other cultural or population contexts.

Recommendations for improvement :

Review the keywords to ensure alignment with MeSH terminology.
Reconsider the stated objective at the end of the introduction, which appears more focused on applicability rather than the specific research aim as pursued in the study.
Expand discussion on the knowledge domain: Consider confirmatory factor analysis or subdivision of the domain to address low internal consistency.
Provide detailed information on the sample: Include more specifics on representativeness (587 adults), potential selection biases in the online sample, sample size calculation, and, if convenience sampling was used, the statistical power, as this is essential for evaluating results and conclusions.
Consider external evidence: Compare findings with similar studies in other countries or cultural contexts to contextualize the validity and relevance of the questionnaire.
Longitudinal perspective: Suggest future studies employing longitudinal or interventional designs to assess changes in KAP over time and to explore causality.

Author Response

We sincerely thank Reviewer 1 for the thorough and constructive evaluation of our manuscript and for recognizing the methodological rigor and practical relevance of this study.

In response to the reviewer’s recommendations, we have refined the manuscript to enhance clarity, transparency, and international contextualization. Specifically, we have:

Revised the keywords to better align with MeSH terminology.
Clarified and more explicitly differentiated the study objectives in both the Abstract and Introduction.
Expanded the discussion of the knowledge domain, clarifying its multidimensional content-coverage conceptualization and aligning reliability interpretation accordingly.
Strengthened the description of sampling procedures and explicitly acknowledged limitations related to convenience sampling and demographic skew.
Clarified the cross-sectional nature of the study and emphasized that findings represent associations rather than causal relationships.
Incorporated additional international comparisons to situate the findings within broader sustainable diet research.
Added recommendations for future longitudinal and intervention-based research.

We believe these revisions have substantially strengthened the methodological transparency and interpretive precision of the manuscript. We sincerely appreciate the reviewer’s insightful suggestions, which have helped improve the clarity and robustness of this work.

Point-to-point reply：

Response to Reviewer 1 Comments

Summary

We sincerely thank the reviewer for the thorough and constructive evaluation of our manuscript entitled:

“Development and Psychometric Validation of a Knowledge, Attitudes, and Practices (KAP) Questionnaire on Sustainable Diets in Taiwan.”

We greatly appreciate the positive feedback regarding the methodological rigor, psychometric evaluation, and practical relevance of the study.

All comments have been carefully considered, and corresponding revisions have been made in the revised manuscript. All changes are highlighted in the resubmitted version.

Questions for General Evaluation

Reviewer’s Evaluation	Response and Revisions
Does the introduction provide sufficient background and include all relevant references?	Response: The introduction has been revised to clarify the study objective and strengthen alignment with international literature. Additional international comparisons have been incorporated (Section 4.1, Lines 368–375).
Are all the cited references relevant to the research?	Response: Yes. References were rechecked for relevance. Keywords were refined to better align with MeSH terminology (Lines 35–36).
Is the research design appropriate?	Response: Yes. The two-phase mixed-methods design (instrument development + cross-sectional validation) is appropriate for psychometric validation research. Clarifications were added where necessary.
Are the methods adequately described?	Response: The sampling procedure and eligibility criteria were further detailed in Section 2.6.1 (Lines 137–147).
Are the results clearly presented?	Response: Yes. Minor clarifications were added in Sections 4.2 and 4.3 to ensure associations are interpreted appropriately, given the cross-sectional design.
Are the conclusions supported by the results?	Response: Yes. The conclusion was slightly refined to emphasize associative rather than causal interpretation and to suggest longitudinal research directions (Lines 490–494).

Point-by-Point Response to Comments and Suggestions for Authors

Comment 1:

Review the keywords to ensure alignment with MeSH terminology.

Response 1:

Thank you for this helpful suggestion. We have revised the keywords to better align with Medical Subject Headings (MeSH) terminology to improve indexing and discoverability. The updated keywords appear in Lines 35–36 of the revised manuscript.

Comment 2:

Reconsider the stated objective at the end of the introduction, which appears more focused on applicability rather than the specific research aim as pursued in the study.

Response 2:

We agree that clearer differentiation of the study objectives improves conceptual clarity. We have revised the objective statement in both:

The Abstract (Lines 15–17), and
The final paragraph of the Introduction (Lines 64–68)

to explicitly distinguish between:

Psychometric development and validation of the instrument, and
Examination of sociodemographic variations and behavioral correlates.

Comment 3:

Expand discussion on the knowledge domain: Consider confirmatory factor analysis or subdivision of the domain to address low internal consistency.

Response 3:

We appreciate this important observation. In Section 4.1 (Lines 351–367), we have expanded the discussion to clarify that the knowledge domain was conceptualized as a multidimensional content-coverage construct rather than a strictly reflective latent factor.

Sustainability literacy encompasses heterogeneous factual domains (e.g., food miles, fisheries sustainability, environmental impact), which may not exhibit high inter-item homogeneity. Therefore, the comparatively lower internal consistency reflects conceptual breadth rather than measurement error.

To enhance transparency, we have added the inter-item correlation matrix in Supplementary Table S6.

Comment 4:

Provide detailed information on the sample: Include more specifics on representativeness (587 adults), potential selection biases in the online sample, sample size calculation, and, if convenience sampling was used, the statistical power, as this is essential for evaluating results and conclusions.

Response 4:

We have strengthened the description of the sampling strategy in Section 2.6.1 (Lines 137–147) and explicitly acknowledged potential selection bias in Section 4.6 (Lines 459–468).

We clarified that:

Online recruitment may have resulted in demographic skew,
No post-stratification weighting was applied, and
Findings should be interpreted as sample-specific rather than nationally representative.

Comment 5:

Consider external evidence: Compare findings with similar studies in other countries or cultural contexts to contextualize the validity and relevance of the questionnaire.

Response 5:

We incorporated comparisons with recently validated sustainable diet KAP instruments in other cultural contexts (e.g., Hilary et al., 2024) in Section 4.1 (Lines 368–375) to provide a broader international context.

Comment 6:

Longitudinal perspective: Suggest future studies employing longitudinal or interventional designs to assess changes in KAP over time and to explore causality.

Response 6

We have clarified in Sections 4.2 and 4.3 that the cross-sectional design precludes causal inference. All relationships are now consistently described as associations rather than predictive or causal effects. In addition, we have added statements in both the Discussion and Conclusion (Lines 490–494) encouraging longitudinal and intervention-based studies to evaluate temporal changes and potential behavioral pathways.

Response to Comments on the Quality of English Language

Point 1: Overall English quality.

Response 1:

The manuscript has undergone careful language editing to improve clarity, consistency, and readability. Minor grammatical and stylistic refinements were made throughout the revised version.

Additional Clarifications

We would like to emphasize that the primary contribution of this study lies in establishing a culturally adapted and psychometrically evaluated KAP instrument for sustainable diets in Taiwan. The revisions strengthen methodological transparency, theoretical interpretation, and international contextualization while maintaining alignment with the study’s original scope.

We sincerely appreciate the reviewer’s insightful and constructive recommendations.

Reviewer 2 Report

Comments and Suggestions for Authors

This manuscript presents the development and psychometric evaluation of a Knowledge, Attitudes, and Practices (KAP) questionnaire on sustainable diets in Taiwan. The topic is timely and relevant, particularly given the increasing policy emphasis on sustainable dietary transitions in Asia. The integration of instrument development with behavioral modeling is commendable, and the sample size is adequate for exploratory psychometric analyses.

The manuscript would benefit, however, from further clarification and expansion of several methodological and statistical aspects to enhance transparency and strengthen the robustness of the conclusions. The comments below are offered in a constructive spirit.

1)Content validity was assessed by two experts, and the reported S-CVI values are high. While this provides initial support for content adequacy, the use of a small expert panel may limit the stability of CVI estimates. The authors may wish to briefly justify this choice and acknowledge it as a potential limitation. Expanding the expert panel in future validation work could further strengthen the instrument’s content foundation.

2)The manuscript would benefit from more detailed reporting of the EFA procedures. Specifically, it would be helpful to include:

KMO measure of sampling adequacy
Bartlett’s test of sphericity
Criteria used for factor retention (e.g., eigenvalues, scree plot, parallel analysis)
Extraction and rotation methods

In addition, the proportion of explained variance for each domain appears relatively low. Providing additional justification for the retained factor structure, and possibly clarifying whether the knowledge domain is conceptualized as reflective or formative, would help readers better interpret the psychometric findings.

3)The knowledge domain shows lower internal consistency compared to the attitude and practice domains. The discussion suggests that this reflects multidimensionality and conceptual breadth, which is plausible. However, it would be helpful for the authors to explicitly clarify the intended measurement model (reflective vs. formative) and align the interpretation of reliability indices accordingly.

Consistency in reporting reliability coefficients (e.g., Cronbach’s alpha, Guttman’s lambda, McDonald’s omega) and a brief rationale for their selection would also improve clarity.

4)Test–retest reliability was evaluated in a subsample (n = 40), which provides useful preliminary evidence of temporal stability. It would be helpful to briefly justify the adequacy of this subsample size and clarify whether overall domain reliability was recalculated after excluding items with low stability or ceiling effects.

Providing a clear summary of the final number of items included in each domain score would also improve transparency regarding the scoring algorithm used in subsequent analyses.

5)The operational definition of “low adherence to a low-carbon diet” (practice score ≤ 2) would benefit from further clarification. Readers may appreciate additional explanation regarding:

The score range of the practice domain
The rationale for selecting this threshold
Whether the cut-off is theory-driven or empirically derived

If feasible, the authors might consider discussing the advantages and limitations of dichotomizing the practice score versus modeling it as a continuous outcome.

6)The logistic regression analysis identifies meaningful associations, particularly for attitudes and policy awareness. To further strengthen the robustness of the findings, it would be helpful to include:

Assessment of multicollinearity (e.g., VIF values)
Basic model fit indicators (e.g., Hosmer–Lemeshow test, pseudo-R², AUC)
Clarification that assumptions for continuous predictors in logistic regression were examined

Including these elements would enhance methodological transparency.

Author Response

We sincerely thank Reviewer 2 for the thoughtful and methodologically focused evaluation of our manuscript. We greatly appreciate the detailed and constructive recommendations aimed at strengthening statistical transparency and psychometric clarity.

In response to the reviewer’s suggestions, we have substantially expanded the methodological reporting and aligned the interpretation of findings with measurement theory. Specifically, we have:

Provided detailed reporting of exploratory factor analysis procedures, including KMO, Bartlett’s test, extraction and rotation methods, factor retention criteria, and total explained variance.
Clarified the conceptual treatment of the knowledge domain as a multidimensional content-coverage construct and aligned the interpretation of internal consistency indices accordingly.
Standardized the reporting and justification of reliability coefficients (Cronbach’s alpha, Guttman’s lambda, ICC, Cohen’s kappa).
Corrected and justified the test–retest subsample size and clarified scoring procedures and final item counts.
Expanded the explanation of the operational definition of low-carbon diet adherence and discussed implications of score dichotomization.
Included comprehensive logistic regression diagnostics (VIF, Hosmer–Lemeshow test, pseudo-R², AUC, linearity assessment) to enhance model robustness and transparency.

We believe these revisions have substantially strengthened the statistical rigor, conceptual alignment, and reproducibility of the study. We sincerely appreciate the reviewer’s insightful and technically rigorous comments, which have significantly improved the clarity and methodological robustness of the manuscript.

Point-to-point reply：

Response to Reviewer 2 Comments

Summary

We sincerely thank the reviewer for the thoughtful and constructive evaluation of our manuscript entitled:

“Development and Psychometric Validation of a Knowledge, Attitudes, and Practices (KAP) Questionnaire on Sustainable Diets in Taiwan.”

We greatly appreciate the recognition of the study’s relevance, adequate sample size, and integration of psychometric validation with behavioral modeling.

All comments have been carefully considered, and corresponding revisions have been made in the revised manuscript. All changes are highlighted in the resubmitted version.

Questions for General Evaluation

Reviewer’s Evaluation	Response and Revisions
Is the work scientifically sound and methodologically rigorous?	Response: Yes. Additional methodological clarifications have been incorporated to strengthen further transparency, including expanded EFA reporting, reliability justifications, and regression diagnostics.
Are the statistical procedures adequately described?	Response: Reporting of EFA procedures, reliability indices, and logistic regression diagnostics has been expanded (Sections 2.4, 3.1, 3.6, and Supplementary Tables).
Are conclusions appropriately supported by the data?	Response: Yes. Clarifications were added regarding cross-sectional interpretation and measurement assumptions to ensure alignment between results and conclusions.
Is the manuscript sufficiently transparent and reproducible?	Response: Yes. Additional methodological details and scoring clarifications have been included to enhance reproducibility and interpretive clarity.

Point-by-Point Response to Comments and Suggestions for Authors

Comment 1

Content validity was assessed by two experts, and the reported S-CVI values are high. While this provides initial support for content adequacy, the use of a small expert panel may limit the stability of CVI estimates. The authors may wish to briefly justify this choice and acknowledge it as a potential limitation. Expanding the expert panel in future validation work could further strengthen the instrument’s content foundation.

Response 1:

Thank you for this important observation. We acknowledge that the expert panel size (n = 2) was relatively small and may limit the stability of content validity index (CVI) estimates.

We have now:

Provided explicit justification of this approach in Section 2.3 (Lines 84–90), and
Acknowledged it as a limitation in Section 4.6 (Lines 459–468).

We clarified that while recommended agreement thresholds were met, future validation studies may incorporate a larger and more diverse expert panel to further strengthen content validity.

Comment 2

The manuscript would benefit from more detailed reporting of the EFA procedures. Specifically, it would be helpful to include:

KMO measure of sampling adequacy

Bartlett’s test of sphericity

Criteria used for factor retention (e.g., eigenvalues, scree plot, parallel analysis)

Extraction and rotation methods

Response 2:

We appreciate this valuable suggestion. We have expanded reporting in the following ways:

Section 2.4.1 (Lines 93–99):

Extraction method: Principal axis factoring
Rotation method: Orthogonal varimax
Retention criteria: Eigenvalues > 1.0, scree plot inspection, and theoretical interpretability

Section 3.1 (Lines 190–197):

Kaiser–Meyer–Olkin (KMO) measure
Bartlett’s test of sphericity
Total variance explained

Regarding explained variance and construct structure, we clarified in Section 4.1 that the knowledge domain was conceptualized as a multidimensional content-coverage construct rather than a strictly reflective latent factor.

These additions enhance conceptual and statistical transparency.

Comment 3

The knowledge domain shows lower internal consistency compared to the attitude and practice domains. The discussion suggests that this reflects multidimensionality and conceptual breadth, which is plausible. However, it would be helpful for the authors to explicitly clarify the intended measurement model (reflective vs. formative) and align the interpretation of reliability indices accordingly.

Consistency in reporting reliability coefficients (e.g., Cronbach’s alpha, Guttman’s lambda, McDonald’s omega) and a brief rationale for their selection would also improve clarity.

Response 3:

Thank you for this insightful comment.

We have clarified in Section 4.1 (Lines 351–367) that the knowledge domain was conceptualized as a content-based, multidimensional construct. Accordingly, internal consistency indices are interpreted as indicators of item homogeneity rather than strict unidimensionality.

We also ensured consistency in reporting reliability coefficients:

Cronbach’s alpha: Attitude and practice domains
Guttman’s lambda: Knowledge domain
ICC / Cohen’s kappa: Test–retest reliability

A brief rationale for these selections has been added in Sections 2.4.2 and 2.4.3.

Comment 4

Test–retest reliability was evaluated in a subsample (n = 40), which provides useful preliminary evidence of temporal stability. It would be helpful to briefly justify the adequacy of this subsample size and clarify whether overall domain reliability was recalculated after excluding items with low stability or ceiling effects.

Providing a clear summary of the final number of items included in each domain score would also improve transparency regarding the scoring algorithm used in subsequent analyses.

Response 4:

Thank you for raising this point.

We have corrected and clarified that the test–retest subsample consisted of 63 participants (Section 2.4.3, Lines 108–113), which exceeds commonly recommended thresholds (30–50 participants) for stability estimation in psychometric validation studies.

We further clarified in Section 2.5:

Items demonstrating low stability or substantial ceiling effects were excluded from composite domain scores.
A clear summary of excluded items and final item counts per domain has been provided.

These revisions improve scoring transparency.

Comment 5

The operational definition of “low adherence to a low-carbon diet” (practice score ≤ 2) would benefit from further clarification. Readers may appreciate additional explanation regarding:

The score range of the practice domain

The rationale for selecting this threshold

Whether the cut-off is theory-driven or empirically derived

If feasible, the authors might consider discussing the advantages and limitations of dichotomizing the practice score versus modeling it as a continuous outcome.

Response 5:

We have clarified in Section 2.7.1 (Lines 165–169):

The practice domain uses a five-point Likert scale (“never” to “always”).
The threshold of ≤ 2 corresponds to an average response between “never” and “rarely,” indicating consistently low engagement.

We further clarified that the cut-off is conceptually driven rather than empirically derived.

In Section 4.3, we added a discussion of:

Advantages of dichotomization (interpretability, policy relevance), and
Limitations (information loss, reduced variability),
As well as acknowledgment that future studies may model practice scores as continuous outcomes.

Comment 6

The logistic regression analysis identifies meaningful associations, particularly for attitudes and policy awareness. To further strengthen the robustness of the findings, it would be helpful to include:

Assessment of multicollinearity (e.g., VIF values)

Basic model fit indicators (e.g., Hosmer–Lemeshow test, pseudo-R², AUC)

Clarification that assumptions for continuous predictors in logistic regression were examined

Including these elements would enhance methodological transparency.

Response 6:

We thank the reviewer for this helpful recommendation.

We have now included comprehensive regression diagnostics in Section 3.6 and Supplementary Table S5:

Variance Inflation Factors (VIF) for multicollinearity
Hosmer–Lemeshow goodness-of-fit test
Nagelkerke pseudo-R²
ROC curve analysis (AUC = 0.778)
Box–Tidwell test for linearity-in-the-logit assumption

These additions strengthen methodological transparency and confirm the robustness of the regression findings.

Response to Comments on the Quality of English Language

Point 1: English clarity and readability

Response 1:

The manuscript has undergone careful language refinement to improve clarity, precision, and consistency in statistical reporting. Minor stylistic adjustments were made throughout the revised manuscript.

Additional Clarifications

We would like to emphasize that the primary contribution of this study lies in integrating psychometric validation with behavioral modeling in a culturally specific context. The revisions improve statistical transparency, clarify measurement assumptions, and strengthen interpretive rigor while maintaining alignment with the study’s original scope.

We sincerely appreciate the reviewer’s constructive and insightful comments.

Reviewer 3 Report

Comments and Suggestions for Authors

This study developed and evaluated a sustainable diet KAP questionnaire and examined how knowledge, attitudes, and practices are associated with adherence to a low-carbon diet among Taiwanese adults. The topic of the manuscript is timely, and considering the authors’ statement that validated instruments are lacking across a large and significant Asian context, the effort to develop and validate such a questionnaire is justified and valuable from the perspective of advancing scientific knowledge. The authors approached this task systematically. The manuscript is conceptually well designed, and the validation procedure is methodologically sound, with a clear and transparent presentation of the results.

Nevertheless, during the evaluation of the manuscript, several issues emerged that require clarification and further elaboration. For better understanding and easier follow-up of the study, I strongly recommend that the full questionnaire be included in the Supplementary Materials. I note that, as a reviewer, the Supplementary Materials were not accessible in a readable format (the zip file could not be opened in a clear and usable form).

Although the validation of a newly developed instrument was conducted, it would be desirable for the authors to specify more precisely in the methodology section (Subsection 2.2) the sources used for item development.

A major limitation of the study concerns the sampling method (convenience sampling), which, as the authors themselves acknowledge in the limitations section, resulted in a sociodemographic skew. It is important to more clearly emphasize how this may have influenced the psychometric results and to clarify whether any weighting procedures were considered. In addition, the manuscript should explicitly state when the survey was conducted.

With regard to the validation procedure, the primary concern relates to the Knowledge dimension of the questionnaire. The authors are requested to clarify the conceptual justification for assessing internal consistency (e.g., Guttman’s λ) if knowledge is treated as a formative construct, since internal consistency indices may not be conceptually appropriate in such cases. It is also unclear why items with factor loadings below 0.40 were retained (Table 1; K3 = 0.139).

In several sections of the manuscript, the authors mitigate the measurement weakness of the knowledge dimension by attributing the low internal consistency (λ = 0.576) to multidimensionality. While this explanation is plausible, the argument would be considerably strengthened by additional empirical support (e.g., providing an inter-item correlation matrix in the Supplementary Materials) and by explicitly acknowledging that lower reliability may limit the use of this dimension in inferential modeling.

Furthermore, please provide methodological references supporting the adequacy of the sample size used for test–retest reliability (n = 40), as this sample size may be considered methodologically debatable and could limit the interpretation of reliability estimates.

Finally, the authors are asked to clarify how the cut-off point for “low-carbon diet adherence” was determined.

After these clarifications and revisions, I believe the manuscript would be suitable for publication.

Author Response

We sincerely thank Reviewer 3 for the thoughtful, detailed, and conceptually insightful evaluation of our manuscript. We greatly appreciate the recognition of the study’s design, methodological rigor, and relevance within the broader Asian context.

In response to the reviewer’s comments, we have strengthened the manuscript with particular attention to construct specification, measurement transparency, and interpretive clarity. Specifically, we have:

Reformatted and verified the accessibility of the full questionnaire and included it as a clearly readable Supplementary File.
Expanded the description of theoretical and empirical sources guiding item development.
Clarified the timing of data collection and further elaborated on the implications of convenience sampling and the absence of weighting procedures.
Explicitly articulated the conceptual treatment of the knowledge domain as a multidimensional content-coverage construct and aligned the interpretation of reliability indices accordingly.
Provided additional empirical support for multidimensionality through inclusion of the inter-item correlation matrix.
Clarified the rationale for retaining theoretically relevant but lower-loading items while excluding them from composite scores when necessary.
Added methodological references supporting the adequacy of the test–retest subsample size.
Clarified the conceptual basis for the operational cut-off used to define low-carbon diet adherence and discussed the implications of dichotomization.

We believe these revisions have substantially enhanced the conceptual precision, measurement transparency, and interpretive boundaries of the study. We sincerely appreciate the reviewer’s constructive and methodologically thoughtful feedback, which has significantly strengthened the manuscript.

Point-to-point reply：

Response to Reviewer 3 Comments

Summary

We sincerely thank the reviewer for the thoughtful and encouraging evaluation of our manuscript entitled:

“Development and Psychometric Validation of a Knowledge, Attitudes, and Practices (KAP) Questionnaire on Sustainable Diets in Taiwan.”

We greatly appreciate the recognition of the study’s conceptual design, methodological rigor, and relevance within the broader Asian context. We have carefully addressed all comments and revised the manuscript accordingly. All corresponding changes have been incorporated into the revised manuscript and highlighted.

Questions for General Evaluation

Reviewer’s Evaluation	Response and Revisions
Is the work a significant contribution to the field?	Response: Yes. The study provides a culturally adapted and psychometrically evaluated sustainable diet KAP instrument for an Asian population, addressing an identified methodological gap.
Is the work well organized and comprehensively described?	Response: Yes. Additional clarifications regarding item development, sampling, and measurement assumptions have been incorporated to enhance transparency.
Is the work scientifically sound and not misleading?	Response: Yes. Further clarification has been added regarding the knowledge construct, reliability interpretation, and regression thresholds to ensure conceptual alignment.
Are there appropriate and adequate references?	Response: Yes. Methodological references have been added to justify test–retest sample size and measurement choices.
Is the English correct and readable?	Response: Yes. Minor stylistic refinements have been made to improve clarity and precision.

Point-by-Point Response to Comments and Suggestions for Authors

Comment 1

The full questionnaire should be included in the Supplementary Materials. The previous file was not accessible.

Response 1:

Thank you for this important suggestion. We have reformatted the full questionnaire to ensure clarity and readability and have now included it as Supplementary File 1. We verified that the file is fully accessible and properly formatted. A reference to the supplementary questionnaire has been added to the manuscript.

Comment 2

Response 2:

We appreciate this recommendation. Section 2.2 (Lines 76–82) has been revised to more precisely describe the theoretical and empirical foundations guiding item development. Specifically, items were informed by:

International sustainable diet frameworks (e.g., FAO/WHO guidance),
Environmental literacy constructs, and
Previously validated sustainability-related KAP instruments.

Items were subsequently adapted to reflect Taiwanese dietary practices and policy terminology. These revisions enhance conceptual transparency.

Comment 3

Response 3:

We have strengthened Section 2.6.1 (Lines 137–147) to explicitly describe the convenience sampling strategy and to state that data collection occurred between December 2024 and February 2025.

We also clarified that:

No post-stratification weighting procedures were applied,
Sociodemographic skew may influence psychometric stability and generalizability.

This issue is further elaborated in Section 4.6 (Lines 459–468), where potential implications for representativeness and external validity are discussed.

Comment 4

Response 4:

We appreciate this insightful comment.

In Section 4.1 (Lines 351–367), we clarified that the knowledge domain was conceptualized as a multidimensional content-coverage construct, rather than a strictly reflective latent factor. Accordingly, internal consistency indices (Guttman’s λ) were interpreted cautiously as indicators of item homogeneity rather than strict unidimensionality.

Regarding items with lower factor loadings (e.g., K3 = 0.139), we clarified that such items were retained due to their theoretical and policy relevance. However, items demonstrating insufficient psychometric performance were excluded from composite score calculations where appropriate to preserve inferential validity. This clarification has been added to Sections 2.5 and 4.1.

Comment 5

Response 5:

We have now included the inter-item correlation matrix in Supplementary Table S6 to provide empirical support for the multidimensional interpretation of the knowledge domain.

Additionally, in Section 4.6, we explicitly acknowledge that the comparatively lower reliability of the knowledge dimension may limit its sensitivity in inferential modeling. Findings related to this domain are therefore interpreted cautiously.

Comment 6

Please provide methodological references supporting the adequacy of the sample size used for test–retest reliability (n = 40), as this sample size may be considered methodologically debatable and could limit the interpretation of reliability estimates.

Response 6:

We have clarified that the correct test–retest subsample size was 63 participants (Section 2.4.3, Lines 108–113). Methodological references have been added to indicate that sample sizes between 30 and 50 participants are generally considered adequate for estimating stability coefficients in psychometric validation studies.

This clarification strengthens the justification for the temporal stability assessment.

Comment 7

Clarify how the cut-off for “low-carbon diet adherence” was determined.

Response 7:

In Section 2.7.1 (Lines 165–169), we now explicitly state that the practice domain is measured on a five-point Likert scale (“never” to “always”). The threshold of ≤ 2 corresponds to an average response between “never” and “rarely,” representing consistently low engagement.

In Section 4.3, we further clarified that this cut-off is conceptually driven rather than empirically derived. We also briefly discuss the advantages (interpretability, policy relevance) and limitations (information loss) of dichotomizing the practice score.

Response to Comments on the Quality of English Language

Point 1: Overall clarity and readability

Response 1:

The manuscript has undergone careful language refinement to improve clarity, precision, and consistency in terminology and statistical reporting.

Additional Clarifications

We would like to reiterate that the study’s primary contribution lies in developing and validating a culturally adapted sustainable diet KAP instrument within an underrepresented regional context. The revisions enhance transparency in measurement assumptions, strengthen methodological justification, and clarify interpretive boundaries while maintaining the original study scope.

We sincerely appreciate the reviewer’s constructive and detailed feedback.

Reviewer 4 Report

Comments and Suggestions for Authors

Dear authors,

I would like to thank the Editor for entrusting me with reviewing this manuscript. Reviewed manuscript details the development and psychometric validation of a Knowledge, Attitudes, and Practices (KAP) questionnaire focused on sustainable diets among adults in Taiwan. Utilizing a two-phase mixed-methods design, the authors conducted expert validation followed by a cross-sectional online survey with 587 participants. The study concludes that the instrument demonstrates robust psychometric properties for the attitude and practice domains, and identifies that attitudes are the strongest predictor of sustainable dietary practices. Furthermore, awareness of sustainability in national dietary guidelines was significantly associated with low-carbon food choices.

Below are my suggestions/comments:

Convenience sampling with a relatively large overrepresentation of women limits the possibility of generalizing the obtained results.

Several knowledge items exhibited ceiling effects or limited response variability. Specifically, Item K6 showed no response variability across test and retest administrations, making Cohen's kappa inestimable. Consequently, items K3, K9, A2, and A7 were excluded from the calculation of composite domain scores due to unsatisfactory reliability. While retaining these items for descriptive reporting has policy relevance, the authors should briefly elaborate on how the exclusion of these items from inferential analyses might impact the overall assessment of the KAP constructs.

Author Response

We sincerely thank Reviewer 4 for the careful evaluation of our manuscript and for the constructive comments regarding sampling considerations and scoring decisions. We greatly appreciate the recognition of the study design and psychometric validation process.

In response to the reviewer’s concerns, we have strengthened the manuscript with particular attention to external validity and construct interpretation. Specifically, we have:

Expanded the discussion of convenience sampling and gender imbalance, explicitly acknowledging limitations in generalizability and clarifying that findings should be interpreted as sample-specific rather than nationally representative.
Clarified that gender was statistically adjusted for in multivariable analyses, while noting that adjustment does not fully eliminate sampling imbalance.
Provided a more detailed explanation of the rationale for excluding items with ceiling effects or limited variability from composite scores.
Clarified that these items were retained for descriptive and policy relevance but excluded from inferential modeling to preserve variance and analytic stability.
Explicitly acknowledged that exclusion of such items may slightly narrow construct coverage and influence variance estimates, while emphasizing the trade-off between conceptual breadth and psychometric precision.

We believe these revisions enhance transparency regarding scoring decisions and strengthen the interpretive boundaries of the study. We sincerely appreciate the reviewer’s thoughtful and constructive feedback, which has improved the clarity and robustness of the manuscript.

Point-to-point reply：

Response to Reviewer 4 Comments

Summary

We sincerely thank the reviewer for the careful evaluation of our manuscript entitled:

“Development and Psychometric Validation of a Knowledge, Attitudes, and Practices (KAP) Questionnaire on Sustainable Diets in Taiwan.”

We appreciate the positive recognition of the study design, psychometric validation process, and key findings regarding the role of attitudes in sustainable dietary practices. All comments have been carefully considered, and corresponding revisions have been made in the revised manuscript. All changes are highlighted in the resubmitted version.

Questions for General Evaluation

Reviewer’s Evaluation	Response and Revisions
Is the work scientifically sound and methodologically appropriate?	Response: Yes. Additional clarifications regarding sampling limitations and scoring decisions have been incorporated to enhance transparency.
Are the psychometric procedures adequately reported?	Response: Yes. Further explanation has been added regarding the treatment of items with ceiling effects and their exclusion from composite scores.
Are conclusions supported by the findings?	Response: Yes. Clarifications have been added regarding external validity and construct coverage to ensure appropriate interpretation of results.

Point-by-Point Response to Comments and Suggestions for Authors

Comment 1

Convenience sampling with a relatively large overrepresentation of women limits the possibility of generalizing the obtained results.

Response 1:

Thank you for highlighting this important limitation.

We have strengthened Section 4.6 (Lines 459–468) to more explicitly acknowledge that the convenience sampling strategy and the overrepresentation of female participants may limit generalizability to the broader Taiwanese adult population.

We further clarified that:

Gender was adjusted for in multivariable regression analyses.
However, sampling imbalance may still influence external validity.
Findings should therefore be interpreted as sample-specific rather than nationally representative.

These revisions enhance transparency regarding population inference.

Comment 2

Response 2:

We appreciate this insightful comment.

We have expanded Sections 2.5 and 4.6 to clarify the rationale and implications of excluding items (K3, K9, A2, and A7) from composite domain scores due to unsatisfactory reliability or ceiling effects.

Specifically, we clarified that:

These items were retained for descriptive reporting due to their theoretical and policy relevance.
They were excluded from inferential modeling to preserve score variability and statistical stability.
This decision aimed to prevent attenuation of associations caused by restricted variance.

We also acknowledged that exclusion of these items may slightly narrow construct coverage within composite scores and may influence domain variance estimates. However, this trade-off was made to enhance psychometric precision and analytic robustness.

These additions improve transparency regarding the scoring algorithm and its implications for construct interpretation.

Response to Comments on the Quality of English Language

Point 1: Overall clarity

Response 1:

The manuscript has undergone minor language refinement to improve clarity, precision, and consistency in statistical terminology.

Additional Clarifications

We would like to reiterate that while certain items demonstrated ceiling effects or limited variability, their descriptive inclusion preserves policy relevance, whereas their exclusion from inferential analyses ensures statistical stability. The revisions further clarify the balance between construct coverage and psychometric rigor.

We sincerely appreciate the reviewer’s constructive and thoughtful feedback.

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors have substantially improved the manuscript. Most recommendations have been fully implemented, resulting in enhanced methodological transparency, interpretive precision, and international contextualization.

Reviewer 2 Report

Comments and Suggestions for Authors

Αfter the authors' final improvements, the article is accepted.

Reviewer 3 Report

Comments and Suggestions for Authors

The authors have addressed all comments, provided relevant and detailed explanations, and significantly improved the manuscript. In my opinion, the manuscript is now suitable for publication.