Review Reports

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript entitled From visions to transformation: Enabling safer learning environments in creative higher education following a capabilities informed, multimethod approach to misconduct prevention presents a timely and socially significant qualitative study addressing misconduct prevention and the development of safer learning environments within Creative Higher Education (CHE). By adopting a multimethod qualitative design and incorporating the perspectives of multiple stakeholder groups, the study contributes valuable empirical insights into an issue that has received increasing international scholarly attention. The manuscript is well structured, methodologically rigorous, and demonstrates a clear commitment to advancing institutional practices that support aspiring artists within higher education.

The study's principal strength lies in its comprehensive examination of stakeholders' perspectives concerning preventative measures against misconduct and the institutional conditions required to establish psychologically safe learning environments. The integration of students' visions with institutional perspectives represents an important contribution to current discussions surrounding organisational culture, safeguarding, and wellbeing within creative higher education.

Nevertheless, I believe that the manuscript would benefit considerably from greater conceptual clarity regarding the role of the capabilities approach throughout the study. This issue affects the coherence between the title, the theoretical framework, the research questions, the analytical structure, and the conclusions.

The title suggests that the manuscript primarily investigates capabilities, or more specifically the development of creative capabilities, within higher education. However, the empirical investigation itself focuses predominantly on the institutional, organisational, social, and environmental conditions that facilitate or hinder safe learning environments. The three research questions similarly emphasise preventative measures, students' perceptions of safety, and the integration of institutional and student perspectives rather than the development or assessment of capabilities themselves.

Throughout the manuscript the authors repeatedly state that "Our analysis is guided by the capabilities approach." However, the capability approach appears to function primarily as a broad normative framework for examining enabling environments rather than as the central analytical object of investigation. In practice, the empirical analysis is directed towards contextual conditions influencing individuals' opportunities rather than towards capabilities themselves.

This distinction becomes particularly evident in the presentation of the findings. The three principal analytical domains—community, institutions, and knowledge and norms—all represent external structural and cultural dimensions. Consequently, the manuscript appears to examine the environmental and organisational determinants of capability development rather than capabilities as such.

From this perspective, the theoretical framing would benefit from a more explicit acknowledgement that the manuscript adopts a sociological and environmental psychological interpretation of the capability approach. Such clarification would considerably strengthen the conceptual consistency of the paper and help align readers' expectations with the actual scope of the study.

Accordingly, I encourage the authors to consider one of two possible revisions.

First, the title could be simplified by removing the explicit reference to the capabilities approach:

From visions to transformation: Enabling safer learning environments in creative higher education to prevent misconduct

Alternatively, if the authors wish to retain the reference to the capabilities approach, its specific interpretation should be articulated more explicitly throughout the manuscript. In this regard, it may be beneficial to emphasise that the capability approach is employed from a sociological and environmental psychological perspective, focusing on the contextual conditions that enable human flourishing rather than on individual capabilities themselves.

This clarification could also be reflected in the organisation of the theoretical section. For example, Section 1.3 could adopt a title such as:

Analysis of Creative Higher Education from the Sociological and Environmental Psychological Perspective of the Capability Approach.

Similarly, explicit reference to this interpretative perspective at later points in the manuscript (for example around line 783) would further improve conceptual coherence.

Clarifying this issue is particularly important because the Discussion ultimately argues that improving institutional environments enables young artists to flourish. This conclusion is both convincing and well supported by the presented evidence; however, it also reinforces the impression that the manuscript primarily investigates enabling environments rather than capabilities themselves.

I also have one methodological observation regarding the temporal context of the study. The manuscript reports that ethical approval was obtained in 2021 and that data collection took place during 2022. Given that the manuscript is being submitted in 2026, it would be helpful if the authors briefly explained the interval between data collection and publication. While such delays are by no means unusual, particularly in qualitative research, a brief explanation would enhance the transparency of the research process and reassure readers regarding the contemporary relevance of the findings.

 

I would like to draw the authors' attention to apparent inconsistencies in the numerical data presented in the tables.

Table 3. The percentage values appear to be incorrect, as the totals should sum to 100%. However:

• Gender: 69.5 + 20.6 + 9.9 + 1.7 = 101.7%
• Degree program cluster: 34.0 + 5.6 + 21.9 + 18.9 + 9.0 + 10.8 = 100.2%

Table 4. The reported total frequencies do not correspond to the sums of the respective subcategories:

1. Open, empathetic communication: n = 103; however, the subcategory counts sum to 58 + 34 + 17 + 17 + 16 = 142.
2. Solid foundations for work and learning: n = 88; however, the subcategory counts sum to 42 + 40 + 21 + 12 = 115.
3. Inclusive and equitable institutions: n = 85; however, the subcategory counts sum to 64 + 32 + 3 = 99.
4. Institutional accountability and integrity: n = 65; however, the subcategory counts sum to 34 + 27 + 17 + 13 + 11 = 102.
5. Reshaped power dynamics: n = 46; however, the subcategory counts sum to 29 + 19 + 8 = 56.
6. Diversity in artistic institutions: n = 40; however, the subcategory counts sum to 23 + 20 = 43.
7. Strong artistic communities: n = 35; however, the subcategory counts sum to 12 + 12 + 10 + 8 = 42.

These numerical inconsistencies may substantially affect the interpretation of the findings. I therefore recommend that the authors carefully review all reported results and correct them where necessary. In addition, the data presented in Table 5 should also be thoroughly re-examined in light of these discrepancies.

 

Finally, the reference list requires careful technical revision. Several hyperlinks (for example References 1, 3, and 26) appear either to be inaccessible or to direct readers to sources different from those cited. A systematic verification of all electronic references is therefore recommended before publication.

Overall, I consider this manuscript to be a valuable contribution to the growing literature on safeguarding, organisational culture, and wellbeing in creative higher education. Its empirical findings are both relevant and potentially impactful. The manuscript would be substantially strengthened by improving the conceptual alignment between the title, the theoretical framing, and the empirical analysis through a more precise articulation of how the capability approach is understood and operationalised within the study. Addressing this issue would enhance the manuscript's theoretical contribution while increasing its conceptual coherence and accessibility for an international readership. Furthermore, a comprehensive review and correction of the numerical data presented throughout the manuscript are essential to ensure the accurate interpretation and reliability of the reported findings.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Thanks for the priviledge to be one of the first readers of your research results! Your paper has strong potential. The topic is important to dig into, the combination of stakeholder interviews and student accounts is original, and the findings are practically relevant. The main work now is to tighten the methodological logic and make sure the conclusions do not go further than the data allow.

While I have a very good impression of the paper as is, I would recommend the folllowing to strengthen it: 

1. At the moment, the interviews focus specifically on misconduct prevention in CHE. But the survey asks about a safer training and working culture, including artistic education, professional life and the wider industry. The two studies do not seem to investigate exactly the same phenomenon. Plz show this difference clearly and present the datasets as complementary rather than fully equivalent. The authors nevertheless treat the studies as a convergent design and classify findings as confirmatory or expanding.

2. You could better explain how the research team decided that a finding was “confirmatory” rather than merely similar, and what counted as an “expansion.” It would help to add a third category, divergent or not directly comparable, and provide one concrete example of each type in the Methods section or joint display.  The current statement that the survey could “quantify” interview themes needs softening as it is overstating now. The survey can show the frequency of coded mentions among the open-text respondents, but it cannot quantify the prevalence or importance of themes identified through the interviews.

3. Only 233 of the 611 survey participants answered the optional open-text question, which is 38% of the full survey sample. This is still a valuable and comparatively large qualitative dataset, but it cannot automatically provide “wider representativeness,” especially because respondents may differ systematically from those who skipped the question. A simple fix would be to replace “wider representativeness” with “broader coverage,” “greater breadth,” or “a wider range of student perspectives.” If the quantitative dataset allows it, please briefly compare open-text respondents with nonrespondents on key demographic or study variables. The current language about confirmatory overlap “strengthening validity” should likewise be replaced by more cautious wording about credibility or convergence.

4. Try better explaining why 16 interviews were sufficient for the stated aims. The sample can be appropriate for reflexive thematic analysis, but the paper needs a clearer rationale based on information power, disciplinary coverage or analytic sufficiency. Plz also report how many institutions were represented and make the sampling limitations more explicit. The interview sample is strongly dominated by music and music-research settings, while several other creative disciplines are absent or only minimally represented. PLz highlight that the themes may be especially reflective of music education and avoid making unrestricted claims about all CHE disciplines.

5. As well, show the differences between stakeholder groups more systematically. The first research question asks what different stakeholder groups consider necessary, but the findings mostly combine students, lecturers, equality experts and senior professionals into shared themes. Try adding a compact stakeholder-by-theme matrix showing areas of agreement, different priorities, tensions and minority positions. This would also strengthen statements such as “all participants,” “participants from all groups” and “several participants across stakeholder groups.” Now readers cannot always see how broadly each claim was supported.

6. The intercoder-agreement procedure needs to be fully. Your paper states that two external coders analysed a random 10% sample and discussed differences until at least 80% agreement was reached. Please specify:

• the exact coding unit: whole response, sentence or meaning segment;
• whether several codes could be applied to one unit;
• how the 10% sample was randomly selected;
• how the 80% figure was calculated;
• what the agreement level was before discussion;
• whether the entire dataset was checked again after the codebook was refined.

Agreement reached after discussion is useful consensus, but it is not the same as an independent reliability estimate. The wording should distinguish these two stages clearly.

7.  Please make clear that the percentages represent the proportion of open-text respondents whose answers were coded under each theme. They do not demonstrate population prevalence, intervention priority, or the relative importance of one theme over another. The themes are also unequal in breadth: some include several broad subthemes, while others contain only one narrow idea. Frequency rankings should therefore be interpreted descriptively rather than as a hierarchy of needs. The table note already explains multiple coding, but the Results and Discussion need the same caution.

8. The theoretical section identifies bodily integrity, emotional development, affiliation, imagination and control over one’s environment as relevant capabilities. However, it remains unclear whether these concepts informed coding, served as sensitising concepts, or were applied only after the inductive themes had been produced. State this explicitly in the Methods section and add a short table mapping the empirical themes onto relevant capabilities. For example, complaint mechanisms could relate to control over one’s environment, safe teaching relationships to bodily integrity, and community support to affiliation. This would make the “capabilities-informed” contribution much more visible.

9. The final analytic sample and correct the demographic table needs a revamp. Now it reports 233 open-text responses, then says one response was excluded and four others were unrelated meta-comments. Please state clearly whether the final thematic analysis included 232, 228 or all 233 cases, and which denominator was used for the percentages. Please also recheck the gender data. The listed percentages (69.5%, 20.6%, 9.9% and 1.7%) sum to 101.7%.  

10. Only four themes are explained in detail, while the other six are moved to the supplementary materials. Since all ten are later used in the integration and discussion, readers need at least a brief definition and one illustrative quotation for each theme in the main paper. A compact summary table with theme definition, one quotation and frequency would be enough.

11. Please explain who translated the German material, whether the coding was performed in German, and how the accuracy of translated quotations was checked. The phrase “lightly edited for clarity and readability” is too open-ended for sensitive material involving misconduct, power and boundaries.   

12. The studies identify what participants believe would make CHE safer; they do not test whether mentoring, training, complaint systems or codes of conduct actually prevent misconduct. Please consistently use formulations such as “participants proposed,” “the findings point to,” “candidate measures,” or “strategies requiring evaluation.” The paper itself correctly states that future research should evaluate the long-term effects of these measures - this means that the abstract, discussion and conclusion should reflect that same caution rather than describing the measures as already effective or essential.

13. Try adding the main limitations directly raised by the design: the broader survey framing, optional-response self-selection, disciplinary imbalance in the interviews, limited systematic comparison between stakeholder groups, and the fact that the same research team analysed and integrated both datasets. The current limitations section does not fully address the comparability and interpretation issues central to this paper.

14. The Methods section clearly reports ethical approval from the Charité – Universitätsmedizin Berlin Ethics Committee, approval number EA2/295/21. The final declaration nevertheless states: “Institutional Review Board Statement: Not applicable.” These two statements contradict each other and must be aligned.

15. The references are generally relevant and current, but a careful technical check is needed. For example, reference 12 appears twice in the same citation sequence “[1, 3, 12, 14, 15, 12]. in the introduction.
Strong statements (such as some DEI training being ineffective or harmful) should be closely matched to the cited evidence and not generalized beyond the scope of that literature.
Either soften this to “training perceived by participants as insufficient” or add direct evidence evaluating anti-harassment, bystander or safeguarding interventions in HE/CHE.
Either explain how the literature was searched or rewrite it as: “We identified limited published evaluation evidence specific to the creative industries.”
Reference [47] is an applied qualitative study about difficulties following psychedelic drug use, not a methodological authority establishing an 80% intercoder-agreement standard. Please cite a methodological source explaining how agreement was calculated and justify why 80% was considered adequate; reference [47] may remain only as an example of applying ST-TA
References [48] and [49] come from mixed-methods health research, and [49] specifically concerns the integration of quantitative and qualitative findings. Since this study integrates two qualitative datasets, please explain how these principles transfer to the present design or add a source addressing qualitative multimethod integration more directly.
Reference [14] appears to report the larger survey project from which Study 2 was drawn. Please identify it clearly as the parent or related study, rather than treating it as independent external confirmation that prevention in CHE is inadequate.
Claims about independent reporting structures, codes of conduct, mentoring, team teaching and mandatory training currently come mainly from participants’ proposals, please cite intervention studies, institutional evaluations or systematic reviews or clearly state that they are participant-informed proposals awaiting evaluation.
The reference list contains several grey literature as valuable reports, institutional surveys and toolkits, including UNESCO, ILO, UniSAFE, institutional reports and AEC recommendations, but central theoretical and effectiveness claims should, where possible, also be supported by peer-reviewed empirical studies or reviews.
Lots of strong language, therefore search for sentences containing words such as “essential,” “must,” “effective,” “harmful,” “demonstrates,” “confirms,” and “no evidence.” - each such claim should be softened or supported by previous literature, or a normative recommendation.
Reference [30] is listed as a 2024 advance online publication. Since the manuscript is dated 2026, please check whether final volume, issue and page details are now available
Please correct inconsistencies such as “Available from: URL:” in reference [32], inconsistent use of abbreviated versus full journal titles, DOI capitalization
Refresh the literature search before resubmission, now the paper seems to be 2 years old and raises concern of being rejected and revised somewhere else because the article makes claims about what is “current” and what evidence does or does not exist, please run a final search covering 2025–2026 work.  

16. Give the manuscript a focused language edit. The English is understandable, but there are recurring awkward or incorrect phrases, such as “mean interviews duration,” “quotes as the following,” and claims such as “putting our conclusions on a solid ground.” A careful language edit would make the argument sound more precise and academically confident.

17.  Figure 1 & Figure 2 need a re-design for better readability. 

18.  The title is too long and ineffective and the abstract needs closer alignment with what the study actually did and can claim.

These concerns are indeed significant, but they are also addressable. I would genuinely like to see this paper published. I therefore recommend major revision and resubmission.  With careful upgrade, it could make a meaningful contribution to research on safer learning environments in higher education. I also want to acknowledge the substantial work already completed and encourage the authors to take the manuscript through this important next stage.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

I would like to take a moment and congratulate the authors on the effort they invested in this revision, esp. in summer.

Now it is considerably stronger, more transparent methodologically, and better aligned with what the two datasets can support. Clearly, the authors engaged seriously with the previous review rather than making only surface-level changes.

Many of the issues that were central to my first review have now been addressed very well: the relationship between the two datasets is much clearer, the methodological choices are better explained, the interview sample and its limitations are more transparently presented, and the stakeholder comparison is much more visible.

I also particularly appreciated the much clearer reporting of the inter-analyst agreement procedure, the corrections to the survey sample and percentages, the stronger explanation of the capabilities approach, and the inclusion of all ten themes in the main paper.

The revised limitations section is much stronger, as well as the literature has been refreshed, and both the figures and title work much better. This has clearly been a major piece of revision work, and I want to acknowledge the care and effort behind it as the manuscript is now in a very different and much stronger place than in the previous round.

At this stage, I see a minor and quite manageable set of remaining points before the manuscript is ready to proceed to publication. These are no longer questions about the overall design - you've done major work here already; so let's focus on: 

• Aligning the three-category integration logic throughout as Figure 2 classifies Themes 9 and 10 as divergent/not directly comparable, but the Results discuss them under “Expanding” and conclude that “the survey findings expanded the interview-based insights.” Add a short separate paragraph for divergent findings and check the Abstract, Results and Discussion for consistency.
• The Discussion refers to a “noticeable lack of awareness regarding economic vulnerabilities and material dependencies” among lecturers and equality officers. The data show that these issues were largely absent or less visible in their interview accounts, which is not necessarily the same as lack of awareness. Likewise, “the two datasets partially confirmed one another” would be better as “were consistent with one another” or “converged on several concerns.”
• Revise the Abstract once more, now it says that the ten survey themes “resonate with these areas and expand them,” although two themes are now classified as divergent/not directly comparable. It also still contains clear errors such as “mean resonate” and “and and prioritize.”
• Do one complete language and technical proofread. There are still duplicated words, punctuation errors, malformed citation strings and awkward sentences. This should be a full final read-through, not only correction of examples already raised by reviewers. Particularly duplicated words, doubled punctuation, citation brackets, subject–verb agreement and sentences that became grammatically unstable during revision, etc.
• Check the final ethics declaration.  Please ensure the full approval statement and protocol number are restored in the final unblinded version.
• Make sure the final volume fits the journal requirements.
• Make the manuscript fully internally consistent and technically clean enough for publication.

Comments on the Quality of English Language

There are recurring editing errors, including duplicated words (“and and,” “the the,” “in in”), malformed citation strings such as “[1,[2]]” and “[3,4,, 5,6],” doubled punctuation, and several awkward or grammatically unstable sentences. The Abstract especially should be checked again, as it still contains “mean resonate” and “and and prioritize.” A full read-through by a fluent academic English editor would help ensure consistency, readability, and a publication-ready final version.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf