Early Dropout Prediction Model: A Case Study of University Leveling Course Students
Round 1
Reviewer 1 Report
In the attached file, please see my general remarks, manuscript-wide comments and specific inline comments/suggestions.
Comments for author File: 
Comments.pdf
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 2 Report
This paper is interesting and the topic is important. However, this paper consists of several shortcomings that need to be improved before this paper can be further considered for publication. Below, I describe my concerns and suggestions.
- The first section (Introduction Section) is quite a bit long. As I carefully read this section, this section indeed can be divided into two sections (Introduction section and Literature review section). In the introduction section, the authors can talk about the motivation of writing this paper and why this issue is so important to be raised for investigation and discussion, and develop research questions that need to be investigated and discussed. In the literature review section, the authors briefly review past studies that are related this paper and use these past studies to build up a theoretical background to frame the study, as well as to develop research hypothesizes that need to be tested in the study.
- An article, I believe, is related to the present paper, but the authors missed it, which is:
Lin, Tin-Chun; Yu, Wei-Choun; and Chen, Yi-Chi, “Determinants and Probability Prediction of College Student Retention: New Evidence from the Probit Model” International Journal of Education Economics and Development, 3(3), 2012, pp. 217–236.
The authors need to cite this article and include it in the literature review section to support the present study and build up a stronger theoretical background.
- There are only four predictors (Regime, Leveling Course Type, Application Grade, and Vulnerability index) in the model and only two predictors (Regime and Application grade) are statistically significant. I believe that there should be more predictors in addition to these four predictors. How about a student’s academic GPA (Grade Point Average)? After students enrolled in the university, students at least have the first semester’s GPA which would influence students’ decision whether or not to keep going. But the authors did not include students’ GPA. In addition, the situation of the economy, such as unemployment rate, economic growth, etc., and students’ employment hours (full time or part time employees) would also affect a student’s decision whether or not to keep going. Several economic and non-economic factors that could influence a student’s decision of coming back to the school were not included in the model. Moreover, although the authors created a vulnerability index, the authors did not explain how this index was created and what factors were included in the index.
- The first two paragraphs in the Conclusions section should be placed in the Literature Review section rather than in the Conclusions section, because in these two paragraphs the authors expressed some results of previous studies. In the conclusion section, the authors should summarize their research findings and make a short discussion and application.
- In the abstract section, the authors said “The school-dropout problem is a serious issue that affects both a country’s education system and its economy”. I completely agree with the authors’ viewpoint. However, the authors did not discuss how the school-dropout problem affects its economy in the discussion section at all. The authors should carefully discuss this issue – how the school-dropout problem affects the economy. In addition, the authors even need to briefly discuss how this issue is related to the COVID-19 pandemic. That is, how the COVID-19 pandemic affects the issue of school-dropout, positive or negative impact?
- The authors even need to discuss how the school authority, administrators, and educators face this issue and how to improve the issue/problem, and offer your policy suggestions or implications.
- In conclusion, a good paper requires (1) a clear research question and hypothesis that can convince readers why this topic is so important to raise for investigation and discussion, (2) a clear and detailed literature review to discuss what previous studies have done in the issue and build up a theoretical framework; (3) a convincible and scientific research approach (a correct experimental design, correct research models, and empirical analysis), and (4) a thoughtful discussion and valuable implications. Unfortunately, this paper does not reach all of these requirements.
Author Response
We appreciate the reviewer's comments. Our responses are detailed in the attached file.
Author Response File: Author Response.pdf
Reviewer 3 Report
The authors should improve the structure of the paper, organizing the sections in a different way. The Introduction should be limited the argument about the pertinence of the study, allowing the reader to get acquainted with the problematic at hand and also with the aim of the study. Following the Introduction section the reader should come across with the theoretical framework, and only after this the Methods.
In order to balance the extension of the sections, especially the ones related to the theoretical framework and the Methods, and also strengthen and better support the results, the literature should be further developed.
In my opinion the interest of the paper would increase if the authors included some discussion (from the existing literature) about the reasons behind the presented numbers concerning drop out.
The title uses the expression “Case Study”. Is this choice of words influenced by the methodological design. If so, it has to be further explained in the Methods section, of not, I think the title should be adaped.
In the conclusions, line 349, it would be an asset for the paper if the authors identified possible ideas to intervene.
Aspects to improve/clarify:
- Line 33, assuming that the higher education attrition has been extensively researched implies presenting results from several studies.
- Lines 47-49, what does the Norwegian study say about drop out rates and the respective reasons?
- Line 65, has a space before the comma and we should read “pointed”.
- How did the authors come up with the 2097 participants? Is this the total number of 1st year students? Were they chosen using specific criteria?
- The courses identified in table 1 are the only courses available?
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
In this revision, the authors mainly followed my comments and concerns to revise the paper, and have significantly improved the paper. However, I still have some concerns and require a minor revision before this paper can be accepted for publication.
- The literature review section should be created as an independent section rather than a sub-section in the introduction section.
- In regard to the vulnerability index, the authors said “This information is stored in 25 variables and, subsequently, based on the students’ responses, weighted scores over 1000 points are assigned to each of the 25 variables.” What are these 25 variables? The authors need list these 25 variables so that readers can see the whole picture. In addition, is each variable’s score weighted by students in the survey? Can the authors show a formula of how this index is created? The authors need to provide more detailed explanation for developing the index.
- In the last paragraph of the conclusion section, the authors said “The COVID-19 pandemic negatively affects school dropouts …” but then the authors said “… Dropouts will increase …” If dropouts will increase in the pandemic period, then the COVID-19 should positively affects school dropouts rather than negatively affects school dropouts. Or the authors can say that the COVID-19 pandemic negatively affects school retentions. I think that the authors made a mistake. In addition, the authors said “… many of these students will drop out of school forever.” I suggest that the authors may say “… many of these students will probably drop out of school forever.” The authors need to add one more word “probably” because we don’t know what it will happen in the future.
Author Response
Please see the attachment
Author Response File: Author Response.pdf
