Explanatory Factors of School Climate and School Identification: An Analysis of Multilevel Latent Profiles

Round 1

Reviewer 1 Report

The presented study deals with the school climate, which affects the academic results of students, but also their behavior. At the same time, it was confirmed that the behavior of students also affects the school climate. The study also showed that family problems affect students' integration into the education system. The same problems were also identified on other continents, e.g. in Europe. I recommend authors to supplement the list of literature with these titles as well. I recommend emphasizing and clearly specifying the goal of the research.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Very relevant article, addressing a very pertinent theme, both in its substance and implications for the development of quality education, capable of responding adequately to the needs of students and allowing to intervene in educational institutions effectively.

Well-founded, with a comprehensive and appropriate research design for the work presented to us, which we must understand as a prolegomenon for further work; in fact, the cross-sectional type of study is fixed in time and the longitudinal perspective is desirable in future research that the team may carry out, looking for causal links between the variables analyzed.

Ambitious work, mobilizing a relevant sample, which allows us to support the conclusions reached. Appropriate and rigorous instruments, full ethical care in the pursuit of research, proficiency in statistical analysis.

The results are relevant and meaningful, and the discussion is proficiently conducted. The limitations of the study are pointed out, but the future line of work that enriches and develops the acquired results is left pointed out.

Congratulations to the team, and best wishes for the article and future research.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

The research purpose is clearly stated and an appropriate metod of sudy is applied. The discussion is efficient and responds to the results and the conclusion. Taking in to account the following remarks in the Introductory section, I suggest specifying: ι)the gap in the reference Literature. ii) to include research hypothesis that are fundamental in a research paper. iii)the contribution of the work on a theoritical ana political level.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 4 Report

Review report of sustainability-2555496

This is a highly relevant and interesting piece of academic work. The authors run several multilevel latent profile analyses to identify individual and school-level profiles of school climate and school identification. I have reason to believe that the study is significant and original to merit publication. Nevertheless, I also have several reservations with regards to the theoretical framing and the analytic procedures, which need to be addressed before further processing of the manuscript.

Main comments

1.     Although the authors have done a good job in describing the importance of school climate, it is not clear why we need this person-centred study of school climate. The authors should describe the differences between variable-centred versus person-centred studies and why we specifically need this person-centred study.

2.     The authors should review other relevant person-centred studies on school climate, discuss their limitations, and how the current study adds to the body of knowledge. The following empirical studies need to be reviewed and critically evaluated to highlight gaps in evidence base: Gage, Prykanowski et al., 2014; Zhao & Kin, 2023; DiStefano, Leighton et al., 2016; Burns, Collie et al., 2022; Pas & Bradshaw, 2014; Capp, Sullivan et al., 2022.

3.     It is not clear what is the conceptualisation of school identification adopted in this study. Additionally, the links between school climate and identification are sparsely discussed and are backed by few relevant references. I definitely think that the authors should provide a greater discussion into the links between these two focal constructs.

4.     The covariates have not been discussed at all in the introduction and no hypotheses have been formulated regarding the possible impact of these. Additionally, the discussion section is devoted to the covariates’ effects on the latent profiles; however, since these covariates were not discussed in the introduction, it comes out of the blue. Please review relevant literature on the covariates and school climate.

5.     In general, it is crucial to create a “present study” section to reflect on the limitations of past studies. Additionally, please explain the aims and the hypotheses of the current study, as well as how the current study adds to existing knowledge of school climate and identification profiles.

6.     It is not clear to what the IFC index refers to? Please add the cut-off values for all fit indices and reliability coefficients utilised.

7.     All confirmatory factor analyses’ results should be transferred to a “preliminary results” section after the method section. Please also comment on the range of the factor loadings per scale.

8.     The sample size of ¬2k students at the level 1 is definitely acceptable; however, I am wondering if the sample size of 28 schools at level 2 is sufficient for a MLPA? Please provide justification by reviewing relevant simulation studies.

9.     The description of the LPA analyses’ parameterisation is not clear. For instance, when estimating the 1-level LPAs, did the authors use cluster-robust standard errors to account for the multilevel design? Furthermore, what is the reasoning behind the authors’ decision to estimate the LPAs separately for level 1 and level 2? The latter question is particularly important since we are talking about different types of parameterisations. Is it a parametric or non-parametric approach to MLPA? Did the authors permit random intercepts across level 2 units?

10.  The authors also mentioned drawing a stratified sample. Therefore, has this stratification been accounted for when estimating the LPA?

11.  How were the covariates included in the modelling? Contemporary modelling theory (e.g., Bakk et al., 2013; Bakk & Kuha, 2021) recommends the ML-3step or the BCH approach in order to avoid shifting the number of latent profiles. Have the authors checked whether including covariates shifted the number of participants per profile or the profile enumeration?

12.  I am also wondering why were the demographics presented in Table 1 not included as covariates? It is highly probable, for example, that different types of schools would display differences in school climate perceptions.

13.  The discussion’s section feels like a repetition of the results’ section since the authors explain the findings of the covariates. There is significant room for improvement here in terms of highlighting the novelty of this work in light of existing evidence. In my opinion, it is more relevant to contrast the findings regarding the four profiles with past profiling studies given the overall emphasis placed on the profiling.

Minor comments

1.     There is spelling error in page 2, whereby “form” should be “from”.

2.     The fourth paragraph of the introduction section 1 begins with a too long sentence that needs to be broken down.

3.     In Figure 1, the “cluster” labels should be placed on the left as a legend box. Alternatively, the authors could switch the labels of the variables with the cluster labels.

4.     A correlation matrix the variables should be presented to gauge how closely related are the variables included in the LPA.

5.     I assume that the authors used z-scores for the LPA given the range of scores in Figure 1. Please clarify.

6.     The authors used the term “profiles” earlier, but the term “cluster” in the results section. Please homogenise the terminology.

7.     Were the covariates grand mean centred?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 5 Report

Dear Authors,

The article is original research manuscripts. The title of the paper has 14 words. The abstract is 195 words.

The work includes the following chapters:

1. Introduction

The introduction chapter has briefly placed the study in a broad context and highlighted why it is important. 

The purpose of the research is formulated –

This study sought to integrate the theoretical approach of social identity and school climate.

 This study aimed at identifying school climate profiles at the individual and school levels, exploring how these profiles are associated with individual and school factors.

 The objectives of the research are not clearly formulated but can be extracted from the context.

Research hypotheses are formulated - The study of the heterogeneity of school climate groups is a relevant line of research. This research approach assumes that there may be different profiles of students who could eventually be interacting in the same school.

 2. Materials and Methods

2.1 Participants

2.2 Instruments

2.3 Procedures

2.4 Data analysis

3. Results

3.1. Identification of the number of clusters and classes of school climate

4. Discussion - ok

5. Conclusions - ok

Recommendation - The clearer formulation of the research objectives/ aims

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 4 Report

I thank the authors for providing point-to-point responses to my initial comments. The manuscript has been significantly improved and the authors have addressed all my main comments. However, I would like to propose some final revisions that could be easily tackled with.

1.      In page 2, the Kraft et al. reference appears twice in the same sentence. Please revise.

2.      Where you mention that parental warmth and income predicting externalizing and internalizing symptoms, I believe it would make the argument more robust to add a couple more references on the warmth dimension, the hostile parenting, and income from robust longitudinal studies using the following relevant works:

Bayer, J.K., Ukoumunne, O.C., Lucas, N., Wake, M., Scalzo, K., & Nicholson, J.M. (2011). Risk factors for childhood mental health symptoms: national longitudinal study of Australian children. Pediatrics 128, e865–e879.

Katsantonis, I., & Symonds, J. E. (2023). Population heterogeneity in developmental trajectories of internalising and externalising mental health symptoms in childhood: differential effects of parenting styles. Epidemiology and Psychiatric Sciences, 32, e16.

Parkes, A., Sweeting, H., & Wight, D. (2016). Early childhood precursors and school age correlates of different internalising problem trajectories among young children. Journal of Abnormal Child Psychology 44, 1333–1346.

3.      The added hypotheses should be stated in a future tense to be anticipatory rather than allude to the results. E.g., will be heterogeneity.

4.      The authors mention also the greatest least bound reliability coefficient but do not report that in Table 3. Please revise.

5.      As you mentioned in your cover letter, please explain that you used the BCH method in the “Data analysis” section.

6.      I assume that “intersections” should be “intercepts” (pg. 8).

7.      In Table 2, please add a note explaining the abbreviations in the first column. Similarly, with Table 3. “Escala” should be “scale” in English.

8.      There are some Spanish terms (e.g., Alfa de Cronbach) that should be translated into English.

9.      Please provide the SRMR index in Table 2. Additionally, add a note to explain the full names of the fit indices.

10.   Although the authors present the Wald tests for the covariates, it is not possible to determine from Table 6 what are the multinomial logistic regression coefficients. I assume that the latent class (i.e., the latent categorical variable) is regressed on the covariates. Therefore, the multinomial logistic regression coefficients should be presented to understand how the covariates predict membership to different latent classes/clusters. If I am mistaken, please clarify.

I thank the authors for the excellent work they have done and I hope to see the published manuscript in due course!

Please consult my comments above regarding the minor language editing.

Author Response

Please see the attachment

Author Response File: Author Response.pdf

Round 3

Reviewer 4 Report

The authors have addressed all my comments and have responded to my queries convincingly. I have no further concerns that need to be addressed. I wish the authors swift publication!