The Interplay of Mindset, Feedback Perception, and Academic Emotion Regulation in Undergraduates’ Self-Regulated Writing Ability

Response to Reviewer X Comments

1. Summary

We sincerely appreciate the reviewers' valuable comments and have thoroughly revised the manuscript to address their feedback to the best of our ability. Responses to the reviewers' comments are provided, and the revised sections of the manuscript are highlighted in yellow.

3. Point-by-point response to Comments and Suggestions for Authors

Comments 1: Introduction:

· Good introduction to the topic.

· Figures 1 and 2: There is a red line under “mindsets,” most probably because it’s a screenshot from another program with language checking. Or does the red line have a meaning? Then please explain it in a figure note.

Response 1: Thank you for your valuable comment regarding Figures 1 and 2. The red underline beneath the word “mindsets” was unintentionally captured as part of a screenshot from a language-checking program and does not hold any specific meaning.

Comments 2: Literature Review:

· Some parts of the literature review need to be shortened, as it is repetitive in several parts. Otherwise, the literature review includes all relevant constructs as well as relevant literature.

Response 2: Thank you for your valuable comments regarding literature review. In response to the reviewer’s comment, the literature review has been revised to reduce repetition and improve conciseness. Redundant sections were shortened or removed, while maintaining a comprehensive coverage of the relevant constructs and literature.

The revised content can be found in Lines 100-127, 137-138, 141-142, and 149-152.

Comments 3: There are several error messages throughout the whole manuscript: „Error! Bookmark not defined.“

Response 3: Thank you for pointing out the error messages (“Error! Bookmark not defined.”) visible in the manuscript. After verifying with the editorial office, we confirmed that the submitted manuscript file is intact and does not contain any such bookmark errors. The message likely resulted from a temporary rendering issue in certain versions of the document viewer.

Comments 4: Methods:

· Again, sometimes repetitive. Please delete all information that is included double, e.g., “Measures such as time limits and mandatory questions were implemented to enhance the completeness and reliability of the data” (in a similar phrasing in section 3.1 and 3.1.1). Please note that this is only an example of a general issue. Please review your whole manuscript in this regard.

Response 4: Thank you for pointing out the issue of redundant phrasing across sections. In response, we have carefully reviewed the entire manuscript and removed repetitive content to enhance clarity and conciseness. Specifically, the redundant sentence regarding data collection procedures—“Measures such as time limits and mandatory questions were implemented to enhance the completeness and reliability of the data”—has been deleted from Section 3.1.1, as it was already described in Section 3.1. Similar overlaps throughout the manuscript have also been addressed to avoid unnecessary repetition and improve the overall readability.

The revised content can be found in Lines 266-294.

Comments 5: You describe the sample in the section “sampling strategy and exclusion criteria) but not in the section “participants and data”. Please reconsider your headings and the related content to help the reader find the information they are looking for.

Response 5: Thank you for your insightful suggestion regarding the organization of the participant-related information. In response, we revised the section heading from “Participants and Data” to “Participants and Sampling Procedures” and integrated key information about the sample characteristics, data collection procedures, and sampling strategy into a single coherent section.

The revised content can be found in Lines 266-294.

Comments 6: “Although this approach facilitated inclusive participation, it did not incorporate randomization or stratified sampling to systematically account for variations in demographic or academic characteristics. Thus, future studies should consider utilizing stratified sampling techniques to improve representativeness by accounting for factors such as academic performance, major, and language proficiency.” -> I think this would be better fitting in the limitations section in the discussion than in the methods section.

Response 6: Thank you for your thoughtful suggestion regarding the placement of the statement on sampling limitations. In response, we have relocated the sentence concerning the lack of randomization and the recommendation for stratified sampling from the Methods section to the Limitations subsection of the Discussion.

The revised content can be found in Lines 609-616.

Comments 7: “168 male students (53.7%) and 145 female students (46.3%)” -> Were there no students identifying as non-binary/diverse? Or did you not include this as an option in the survey?

Response 7: Thank you for pointing this out. The survey included three options for gender identification: male, female, and “prefer not to disclose / other.” However, none of the participants selected the third option. We have clarified this in the manuscript to enhance transparency regarding the gender-related data.

The revised content can be found in Lines 286-287.

Comments 8: “Second, writing feedback perception was assessed using five items selected from [Ekholm et al., 2015] and [Zumbrunn et al., 2016], which were adapted to align with the course objectives. The scale comprised three subfactors—perceptions of instructor feedback, peer feedback, and feed-back from acquaintances—with a total of five items.” -> please report the exact items in a supplementary file to enhance transparency and reproducibility. Also, please justify how you can measure 3 subfactors with only five items in total. 1 subfactor cannot be presented by only one item for a valid test.

Response 8: Thank you for your insightful comment. We have added all five items used to assess writing feedback perception to Appendix to enhance transparency and reproducibility. As you correctly noted, using only five items to assess three subfactors presents limitations in construct validity. In response, we clarified in the manuscript that two subfactors (instructor and peer feedback) were each assessed with two items, while the third subfactor (feedback from acquaintances) was assessed with a single item. We have acknowledged this limitation explicitly and advised caution in interpreting results related to this construct.

The revised content can be found in Lines 313-315, and 665-729.

Comments 9: If I understood it correctly, self-regulated writing ability was measured by using self-report Likert scale items? This feels inappropriate to me, as you cannot measure “abilities” with a self-report. If you do this, you more likely measure self-efficacy beliefs. Please explain and justify your method and report example items for each instrument; maybe I have just misunderstood. Readers will appreciate example items to help them understand what exactly you measured.

Response 9: Thank you for raising this important point. We agree that self-report measures may not directly assess actual writing performance, and that such instruments primarily capture learners’ perceived ability or confidence in managing their writing processes. However, as noted in prior research (e.g., Bandura, 2006; Schunk & Zimmerman, 2007), perceived self-regulatory competence is a meaningful construct in itself, especially in understanding learning behaviors in educational settings. We have revised the manuscript to clarify this distinction, and we have added example items in the methods section, as well as the complete item list in Appendix, to enhance clarity and transparency.

The revised content can be found in Lines 665-729.

Comments 10: Did you check the requirements for Cronbach’s alpha? In this case, please report it. Otherwise, McDonald’s omega may be a more appropriate choice.

Response 10: Thank you for this thoughtful suggestion. We confirm that Cronbach’s alpha coefficients were calculated and reported in the manuscript. The internal consistency values ranged from .73 to .886 across all main constructs: .74 for growth mindset, .78 for fixed mindset, .73 for writing feedback perception, .881 (overall) for academic emotion regulation (with subscale alphas ranging from .763 to .886), and .777 for self-regulated writing ability. In response to your comment, we have added a brief justification in the Methods section, clarifying that the assumptions for using Cronbach’s alpha—particularly the unidimensionality of each subscale—were reasonably met. While we acknowledge that McDonald’s omega is a valuable alternative, Cronbach’s alpha remains an appropriate reliability estimate for the unidimensional subscales employed in this study.

The revised content can be found in Lines 333-337, and Section 3.2.

Comments 11: Which type of correlations did you conduct? Pearson? Spearman?

Response 11: Thank you for your question. We confirm that Pearson’s correlation coefficients were used in the preliminary analyses. We have now clarified this in the manuscript. As the variables met the assumptions of normality, Pearson’s r was deemed appropriate for assessing linear relationships among the continuous variables.

The revised content can be found in Lines 345-367.

Comments 12: Results:

· „χ²/df = 1.85“ -> please report not only the ratio as this has been shown to be not an appropriate method, but please report the exact results of the chi square test, including significance

Response 12: Thank you for your valuable comment. In response to the reviewer’s suggestion, the exact results of the chi-square test—including the chi-square value, degrees of freedom, and significance level—have been reported, as relying solely on the χ²/df ratio is not considered an appropriate method.

The revised content can be found in Lines 370-375, and 382-390.

Comments 13: Section 4.2: Please do not report the same results in both the main text and the table. This is especially related to the reporting of M. Please delete from one place, preferably the main text.

Response 13: Thank you for your suggestion. To avoid unnecessary repetition, we have removed the detailed reporting of means and standard deviations from the main text in Section 4.2, as they are already presented in Table 4.

The revised content can be found in Lines 392-395.

Comments 14: Please justify why you interpret a kurtosis of “1.273” as a sign of normal distribution and consider using further criteria

Response 14: Thank you for your valuable comment. In response, we have clarified the basis for interpreting a kurtosis value of 1.273 as indicative of univariate normality. Specifically, we referenced the criteria proposed by West, Finch, and Curran (1995), which suggest that skewness values less than 2 and kurtosis values less than 7 are acceptable indicators of normality for the application of parametric analyses.

The revised content can be found in Lines 396-399.

Comments 15: “fixed mindset did not significantly contribute to the model (β = −0.006, p > .05)” -> please report exact p-values

Response 15: Thank you for pointing this out. We revised the manuscript to report the exact p-value instead of simply stating "p > .05." The sentence now reads: “Fixed mindset did not significantly contribute to the model (β = −0.006, p = .930).” This revision has been made in Section 4.3. of the manuscript.

The revised content can be found in Lines 417-418.

Comments 16: Please explain why you left fixed mindset in model 3 and 4, even though it was not a significant predictor in model 2. Please also explain the order of including the various variables into the models.

Response 16: Thank you for your insightful comment. In the revised manuscript, we clarified the rationale for retaining the fixed mindset variable in Models 3 and 4, despite its non-significance in Model 2. Specifically, we explained that fixed mindset was retained to ensure theoretical consistency, control for possible suppressor effects, and allow for the examination of potential interactions with subsequently added predictors. We also elaborated on the rationale behind the order of variable inclusion across the hierarchical regression models. The description has been added in Section X.X of the manuscript.

The revised content can be found in Lines 418-426.

Comments 17: "This finding is consistent with previous research indicating that a growth mindset, rather than a fixed mindset, is more closely associated with psychological and behavioral factors related to learners’ self-regulation (Dweck, 2006)." -> This reads more like a discussion than the results section.

Response 17: Thank you for your careful observation. In response, we have revised the sentence to remove interpretive language and ensure that the tone remains consistent with the purpose of the Results section. The original statement has been relocated to the Discussion section, where it is more appropriately situated within the context of theoretical interpretation.

The revised content can be found in Lines 496-541.

Comments 18: Table 4: p can never be 0. Please adjust to < .001. Also, “0.2153**“ should have three astericts, because p < .001, and “0.1663*” two asterics.

Response 18: Thank you for pointing out the inconsistency in p-value reporting. We revised Table 5 to reflect the correct statistical convention that p-values can never be exactly zero. All instances of “p = .000” were adjusted to “p < .001.” In addition, we corrected the asterisk annotations to match the corresponding significance levels: “0.2153*” was updated to “0.2153***”, and “0.1663*” was updated to “0.1663**.” These corrections have been applied in Table 4 and the associated note below the table.

The revised content can be found in Tables 4 and 5 in yellow.

Comments 19: Figure 2: Please state in the table note the meaning of the number of asterisks, i.e., different ranges of p-values.

Response 19: Thank you for your helpful suggestion. We have added a note below Figure 2 to clarify the meaning of the asterisks representing levels of statistical significance. The following note was included: “*p < .05, **p < .01, ***p < .001.”

Revisions can be found in Figure 2 as follows.

4. Response to Comments on the Quality of English Language

Point 1: The English is fine and does not require any improvement.

Response 1: Thank you for your positive evaluation regarding the quality of English. Although the language was deemed acceptable, we had the final version of the manuscript professionally proofread by a native English speaker to ensure clarity, fluency, and consistency throughout.

Response to Reviewer X Comments

1. Summary

We sincerely appreciate the reviewers' valuable comments and have thoroughly revised the manuscript to address their feedback to the best of our ability. Responses to the reviewers' comments are provided, and the revised sections of the manuscript are highlighted in yellow.

3. Point-by-point response to Comments and Suggestions for Authors

Comments 1: First, the abstract needs to be self-sufficient and clear: it currently contains a truncated sentence and fails to indicate the type of design and the main statistic of the effect, although it does mention the sample size. I ask you to rephrase the incomplete sentence, add in a few words that the study is cross-sectional, and specify the value of the impact coefficient accompanied by its confidence interval.

Response 1: Thank you for your valuable feedback. In response to your comment, we revised the abstract to ensure that it is self-contained and clearly communicates the study design and main statistical findings.

The revised content can be found in Lines 7-20.

Comments 2: The introduction requires a substantial update. The first two paragraphs are redundant and rely on literature prior to 2018. It is advisable to condense this section, incorporate works published between 2022 and 2025 directly related to growth mindset and emotional regulation in writing, and conclude with an explicit and numbered presentation of the hypotheses that guide the serial mediation model. This will help the reader understand the novelty and directionality of the research.

Response 2: Thank you for your valuable feedback regarding the Introduction. In response, we have substantially revised the first two paragraphs to eliminate redundancy and improve clarity. We have incorporated recent literature published between 2022 and 2025 that specifically addresses growth mindset and academic emotion regulation in the context of writing. Furthermore, we have added an explicit and clearly numbered list of hypotheses at the end of the Introduction to clarify the theoretical foundation and directionality of the serial mediation model.

The revised content can be found in Lines 25-78.

Comments 3: The method section should include essential data about the sample (inclusion and exclusion criteria, mean age and standard deviation, distribution by gender and career) and, above all, ethical information: name of the committee that approved the study, approval code, and description of the informed consent process. It is also essential to report the reliability and validity of the scales used in this sample, provide a representative example of an item, and justify the sample size using a priori power analysis. Also explain why you chose PROCESS Model 6 and the bootstrap parameters.

Response 3: Thank you for your detailed and constructive feedback. We have thoroughly revised the Method section to address all points raised. Specifically, we now include detailed information about the sample, including inclusion and exclusion criteria, mean age with standard deviation, and distribution by gender and academic discipline. We have also added ethical information, including the name of the approving institutional review board, the approval code, and a clear description of the informed consent process. In addition, we now report the reliability and validity of each measurement instrument as tested with this sample and provide representative example items for each scale to enhance clarity. The justification for the sample size is now supported by an a priori power analysis. Lastly, we have explained the rationale for choosing PROCESS Model 6 for serial mediation analysis and specified the bootstrap parameters used in the estimation procedure.

The revised content can be found in Lines 266-294, 333-367, and Section 3.2.

Comments 4: The results need to be clearer. I suggest renaming the variables in Table 1 with understandable acronyms, accompanying the CFA with a table that includes factor loadings and R², verifying the null value of adjusted R² in Table 2, and providing the standardized indirect effects with their confidence intervals. An SEM diagram would help the reader visualize the mediation.

Response 4: We sincerely thank the reviewer for their valuable comments. In response, we have renamed the variables in Table 2, 3, 4, 5 and 6, using more intuitive acronyms, added a new table presenting the CFA results with factor loadings and R² values, and verified the adjusted R² value in Table 3. Furthermore, we have included the standardized indirect effects along with their 95% confidence intervals (in Table 7). Thank you for your suggestion regarding the inclusion of an SEM diagram. To avoid redundancy, we chose not to add a separate SEM figure and instead revised and clarified Figure 2 to effectively present the serial mediation model.

The revised content can be found in Lines 370-375, 382-402, 417-426, and 464-494.

Comments 5: Next, the discussion should focus on interpreting the findings in light of recent literature, eliminating repetitions of results, and delving deeper into the pedagogical implications through concrete examples (e.g., feedforward strategies aimed at regulating emotions during writing). It is also important to qualify the universal conclusions by considering the Korean cultural context of the sample.

Response 5: Thank you for the insightful feedback. The discussion section has been revised to better interpret the findings in light of recent literature, eliminate redundant repetition of results, and explore pedagogical implications through specific examples such as feedforward strategies for emotion regulation during writing. Additionally, the generalizability of the findings has been appropriately qualified by acknowledging the Korean cultural context of the sample.

The revised content can be found in Lines 496-541, and 583-595.

Comments 6: Add an explicit section on limitations and future directions, acknowledging self-report bias, the cross-sectional nature of the design, and the need for longitudinal or experimental studies.

Response 6: We sincerely thank the reviewer for highlighting the need to explicitly address the limitations of the study. In response, we have added a section titled 6. Limitations and Future Directions to the revised manuscript. This section now clearly acknowledges potential self-report bias, the constraints of the cross-sectional design, and the necessity of future longitudinal or experimental studies. Additionally, we discuss issues of generalizability and the need for broader variable inclusion and mixed-methods approaches.

The revised content can be found in Lines 596-632.

Comments 7: The reference list should be updated with works from 2021 to 2025, include DOIs in each entry, follow the MDPI format, and eliminate duplicates, ensuring exact correspondence between citations in the text and the bibliography.

Response 7: Thank you for your suggestion. We have updated the reference list by including recent studies published between 2021 and 2025, added DOIs to all entries, reformatted citations to follow the MDPI style, and removed duplicates. We also ensured full consistency between in-text citations and the reference list.

The revised content can be found in Lines 730-938.

Comments 8: Finally, the manuscript contains awkward word breaks and phrasing; a professional English review will improve clarity and overall presentation. In addition, the messages “Error! Bookmark not defined” should be removed.

Response 8: Thank you for pointing out the error messages (“Error! Bookmark not defined.”) visible in the manuscript. After verifying with the editorial office, we confirmed that the submitted manuscript file is intact and does not contain any such bookmark errors. The message likely resulted from a temporary rendering issue in certain versions of the document viewer. Therefore, no revision was necessary in the manuscript content itself.

4. Response to Comments on the Quality of English Language

Point 1: The English could be improved to more clearly express the research.

Response 1: Thank you for your valuable feedback. In response, we thoroughly revised the manuscript to improve clarity and precision in English expression. The entire manuscript was carefully proofread by a native English speaker to ensure fluency and academic appropriateness.

Response to Reviewer X Comments

1. Summary

We sincerely appreciate the reviewers' valuable and careful comments and have thoroughly revised the manuscript to address their feedback to the best of our ability. Responses to the reviewers' comments are provided, and the revised sections of the manuscript are highlighted in green.

3. Point-by-point response to Comments and Suggestions for Authors

Comments 1: a) Explicit hypotheses: A confirmatory mediation model requires that direct and indirect paths be formulated explicitly as hypotheses (H1, H2, etc.) or, at a minimum, as clearly stated research questions. Doing so will help link each statistical test to a specific theoretical expectation.

Response 1: Thank you for your insightful comment. In response, we have ensured that each direct and indirect path in the proposed mediation model is explicitly formulated as a hypothesis (H1–H4). These hypotheses have been clearly stated in the Introduction section (Lines 87–95) and are consistently referenced in the Results section to establish a clear link between the theoretical expectations and the statistical tests conducted. Specifically, each result has been aligned with its corresponding hypothesis.

The revised content can be found in Lines 404-406, and 473-475.

Comments 2: b) Sample description and ethics: It is essential to report the mean and standard deviation of participants’ age, gender and discipline distribution, inclusion criteria, and, in addition, to specify approval by an ethics committee (including code and date) or, at minimum, the acquisition of informed consent.

Response 2: In response to the reviewer’s comment, we have thoroughly revised the sample description and ethical considerations. The mean age of participants (M = 20.84, SD = 1.77) and the distribution of gender (53.7% male, 46.3% female) have been included. Additionally, the breakdown of academic disciplines has been provided, detailing the students' enrollment in various engineering and arts-related majors. The inclusion criteria, which required participants to be enrolled in the course, complete the survey, and provide informed consent, have been clearly stated. The study received ethical approval from the university’s Institutional Review Board (IRB number: 7002340-202505-HR-011, approval date: May 13, 2025), and all procedures adhered to the Declaration of Helsinki and the university’s IRB guidelines.

The revised content can be found in Lines 274-277, and 278-294.

Comments 3: The Justification of sample size: Please include a priori power analysis or an equivalent justification demonstrating that N = 313 is sufficient for the serial mediation model used.

Response 3: Thank you for your insightful comment regarding the need for a priori power analysis. We have revised Section 3.3 to explicitly describe the a priori power analysis conducted using G*Power 3.1. This analysis assumed a medium effect size (f² = .15), an alpha level of .05, and a desired power of .80. The minimum required sample size was calculated to be approximately 91. We have also clarified that the final sample size of 313 not only meets but exceeds this threshold, thereby ensuring sufficient statistical power for both multiple regression and the serial mediation model (PROCESS Model 6) involving two mediators in a specified causal sequence. This revision aligns with your suggestion and confirms the adequacy of the sample size for the analyses conducted.

The revised content can be found in Lines 338, and 343-346.

Comments 4: Anomaly in Adjusted R² for Model 2: In Table 2, the Adjusted R² remains very low (0.008), which contrasts with the magnitude of the coefficients. This discrepancy should be explained or the value corrected if it is due to a transcription error.

Response 4: Thank you for your careful observation. We have reviewed the original output and confirmed that the adjusted R² value for Model 2 is 0.165, not 0.008. The discrepancy arose from a transcription error in the earlier version of the manuscript. As shown in the revised Table 4), we have corrected the adjusted R² value and also ensured that all F-statistics and significance indicators are accurate and consistent with the ANOVA output from SPSS. We appreciate your attention to this detail.

The revised content can be found in Table 4 in green.

Comments 5: Depth of the discussion: It is advisable to replace paragraphs that repeat numerical results with more analytical interpretation. Additionally, the transferability of findings should be nuanced by referencing the Korean cultural context of the sample.

Response 5: Thank you for the valuable feedback. We revised the discussion and conclusion sections to reduce redundancy in reporting numerical results and instead emphasized analytical interpretation. We also added a cultural contextualization of the findings, discussing how Korean educational norms may have influenced learners’ responses to feedback and emotional regulation, thereby addressing the issue of transferability.

The revised content can be found in Lines 591-603.

Comments 6: f) Limitations and future directions: A dedicated section is needed to address the cross-sectional nature of the study, self report bias, use of a single sample, and potential uncontrolled variables. This will help guide readers in assessing the generalisability of the findings.

Response 6: Thank you for your insightful comment regarding the limitations of the study. In response, a dedicated section (Section 6) has been added to explicitly address the cross-sectional nature of the research, potential biases stemming from self-report measures, the use of a single sample from one institution, and the possibility of unmeasured confounding variables. We have also suggested future research directions, including the use of longitudinal and mixed-method designs, stratified sampling, and inclusion of additional variables to strengthen generalizability and validity.

The revised content can be found in Lines 604-640.

Comments 7: g) References: While the reference list is relevant and up to date, it is important to remove duplicates, add missing DOIs, and ensure full compliance with MDPI style (alphabetical order, APA MDPI punctuation).

Response 7: Thank you for your valuable comment regarding the reference list. In response, we carefully reviewed all entries to ensure full compliance with MDPI's APA style guidelines. We have removed any duplicated references, added missing DOIs for all journal articles (excluding books, which do not have DOIs), and ensured proper alphabetical ordering based on the first author’s last name. Punctuation and formatting have also been corrected according to MDPI requirements. The updated reference list reflects these revisions.

The revised content can be found in Lines 736-944.

Comments 8: h) English language quality: The text is understandable, but word breaks and overly long sentences persist. A style revision would improve fluency and clarity.

Response 8: Thank you for your valuable comment regarding style and readability. In response, we revised the manuscript to improve fluency and clarity by addressing awkward word breaks and simplifying overly long sentences. The final version was also professionally proofread by a native English speaker to ensure linguistic accuracy. Revisions were made throughout, especially in sections that previously contained dense or complex expressions.