The Malleability of Higher Education Study Environment Factors and Their Influence on Humanities Student Dropout—Validating an Instrument
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThe study fills a research gap as there is a shortage of research on the effect of pedagogical factors and the study environment on the dropout rates of higher education students. As a novel perspective, the paper examines temporal aspects (changes in the study environment over time). As the analysis focuses only on humanities students at one Danish university, the effects of the field of study and country, explored in several previous papers, are not examined here.
Another goal of the paper is to validate an instrument and to test Tinto’s model of dropout. I suggest including a brief discussion of Astin’s theory in addition to Tinto’s. According to Astin, student persistence is primarily an action, while in Tinto’s model, it is predominantly an attitude. This could provide an alternative perspective on dropout. Additional clarification is required for Figure 1, where the dependent variable is persistence: students with a non-persistent attitude are not necessarily equivalent to those who will drop out later. Please also specify the dependent variable in the regression model, as there could be students who have neither graduated nor dropped out (i.e., those who are still studying) during the examined period.
The research presents several interesting empirical results, but the two research questions do not cover all of them. To address this, I suggest formulating more research questions and also hypotheses based on the findings of the literature (with references). The discussion contains assumptions, which should be formulated before conducting the empirical research.
There are 8+8+13 theoretical dimensions and 28+22+64 empirically measured items, but for the academic and social systems, not all theoretical factors are included. Why? Another question: is the register data only used in the regression analysis, or is it also used alongside the questionnaire data to create empirical factors? Please clarify this in the methodological section, possibly in the form of a table. For the regression analysis, please provide the descriptive statistics of the dependent and independent variables (gender and age are not enough). If the response rate is decreasing, please explain what speaks against applying weights based on population data. In what sense is the sample distorted due to low response rates?
In the factor analysis, are there communalities lower than 0.25? Is the total variance explained by the factors higher than 50%? Factor loadings are higher than 0.3 but are still quite low; why are they not higher than 0.4 or 0.5? In Tables 3-5, what is marked in black? (Could it perhaps mean that factor loadings are higher than 0.3?) Table 3 is not about the social system, so please replace it.
In Table 6, please replace the word “dependent” with “independent” when referring to the variables. Please clarify whether the model inputs were included stepwise with only the growth of Nagelkerke R-squared indicated, or whether these were independent models. (The stepwise inclusion of independent variables in logistic regression can be problematic.) Are the odds ratios of the final model presented with the inclusion of all independent variables, or are they from separate models? The former approach would be preferable.
In the discussion, based on the results, please provide answers to more research questions and test the hypotheses. There are several interesting results in the factor analysis, but the discussion only touches on a few interesting results. The temporal changes are well discussed in the regression analysis, but if the focus is on pedagogical factors, explore them in more depth. Then, formulate conclusions, emphasising the novelty of the research. The conclusion that early interventions are needed and that students at risk of dropping out should be identified early is well known from several papers. Please emphasise the novelty of this paper. If the validation of the instrument is also an important goal of the paper, please discuss this separately. Than provide policy recommendations based on the findings, discuss the limitations of the study, and present further research plans. Please address the limitations of generalising the results, as the study examines only one faculty at one university. What kind of distortion can this cause? Finally, do not mix these subjects in the discussion.
Author Response
We thank the two reviewers for the positive words about both the article’s value and its content and form. Furthermore, we thank for the good and thorough comments, which are very helpful in strengthening and completing the article. We attach a document with a table, where we systematically describe how we revised the article based on each of the reviewers’ comments.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Comments and Suggestions for AuthorsReview of education-3076432– The Malleability of Higher Education Study Environment Factors and Their Influence on Humanities Students' Drop-Out—Validating an Instrument.
July 9, 2024
General
This study is a well-written and much-needed addition to the scarce research on some elements of Tinto's Institutional Department Model.
The study is well executed and could be replicated across other universities, which makes this study even more important. The identified factors could also be converted into a framework to study environmental factors. These considerations could be considered in the final discussion.
I'm excited about this study and look forward to some improvements. Future work could study machine learning to build models over time, as prediction models show what factors become more or less critical between models (variable importance factors).
Abstract
Minor remarks
- The final sentence of the abstract is quite long and hard to read.
Introduction
The introduction is well-written and very informative. However, there are some major and minor points to improve upon.
Major remarks
- The categories in the table need additional explanation on the relationship to the model; they relate to institutional experiences. However, the model does not present 'Teaching' as a concept. Possibly, 'Teaching' relates to 'Classes, Labs, and Studios'?
- The differences between the original model of Tinto and the additional factors found in the literature should be more explicit in the text and Figure 1: what are new elements?
Minor remarks
- Intension should be Intention in Figure 1 (in the 7th column).
- The layout/typography of Table 1 is confusing: why is the first column in bold? I advise making the headers bold and the first column in a regular font face.
- Around the T2 Goal Commitments, there should be a box.
Materials and Methods
The Materials and Methods section is clear and concise.
Major remarks
· Please clarify if all items of the model were operationalized or not. If not, what elements were not operationalized?
Results
Major remarks
· The description of findings below the tables in paragraph 5.1 is hard to read. Maybe using some keywords in bold or italics relating to the tables could help.
Minor remarks
· The resolution of the tables is too low.
· Table 3: header: columns has a typo.
· Table 6: considerations
Discussion
Major remarks
· National regulations on progression, for instance, forced drop-out if credits in the first year are below a certain threshold, could influence the importance of certain factors. The same applies to national financial aid regulations. I would like to invite the authors to reflect upon their national regulations, if there are any.
· Van Gennep's theory of transition is an exciting addition to this study. However, a more straightforward explanation could be that assignments in the final phase of the study, such as internships and a thesis, require more academic skills and executive functions. This reasoning would explain many of the findings on support and the differences between males and females. I would advise considering this kind of reasoning as well.
· The implications for educational institutions to improve the educational context for students could benefit from some examples.
Minor remarks
· One reference needs improvement: {Olteanu, 2013 #7218}
Comments on the Quality of English Language
Author Response
We thank the two reviewers for the positive words about both the article’s value and its content and form. Furthermore, we thank for the good and thorough comments, which are very helpful in strengthening and completing the article. We attach a document with a table, where we systematically describe how we revised the article based on each of the reviewers’ comments.
Author Response File: Author Response.pdf
