Implementation of the E-Learning Model for Sustainability of Driver Rehabilitation Program

Round 1

Reviewer 1 Report

1. Please illustrate your research framework with figures.
2. Author(s) need to well- illustrate your viewpoint by relating your own methodology.
3. Figures can help you to explain how the research work and provide international readership to get a better understanding of your academic work.
4. Make it more clear, please. What model exactly? TAM 1? UTAUT model? ASSURE model? SIP model? or ??????
5. Update your references, please (from 2010-2021)

Suggested references:

1. The Optimal Setting of A/B Exam Papers without Item Pools: A Hybrid Approach of IRT and BGP
2.An Empirical Study of How the Learning Attitudes of College Students toward English E-Tutoring Websites Affect Site Sustainability
3. Application of Information Technology in Preschool Aesthetic Teaching from the Perspective of Sustainable Management
4.English e-learning in the virtual classroom and the factors that influence ESL (English as a Second Language): Taiwanese citizens’ acceptance and use of the Modular Object- Oriented Dynamic Learning Environment

Author Response

Dear

Thank you for suggestion. 

1. Please illustrate your research framework with figures. - Author(s) need to well- illustrate your viewpoint by relating your own methodology. accepted
2. Figures can help you to explain how the research work and provide international readership to get a better understanding of your academic work. accepted
3. Make it more clear, please. What model exactly? TAM 1? UTAUT model? ASSURE model? SIP model? or ?????? – partially accepted
4. Update your references, please (from 2010-2021) - accepted

Best regards

Djordje Vranjes

Reviewer 2 Report

The subject material is presented in a much better way and in a comprehensive manner, however, authors could improve some aspects that I suggest.

Abstract:

“For the needs of making a statistical comparison of tested groups before and after the training, statistical technique of paired sample t-test of independent samples has been used in this work,... ”: it is clear that the statistical test used was the “t-test”, however the problem here is that this statistical test is used for comparison between two dependent (or paired) samples or for two independent samples... not both simultaneously, as the sentence implies...

Introduction:

In the number reference system, the bibliography used is arranged by the order in which the citations appear in the text. The first source you cite in your text receives the number 1, the second number 2, and so on... . In the actual version of the paper the first reference in the text is identified by [1] however the second is identified by [8], and the third by [4]... instead of consecutively.

Works resulting of research can be found and giving a broad and emerging perspectives regarding the topic under study in the paper. So, it is expected an inclusion of more papers describing experiences carried out even in different countries and in different sectors, but that show the relevance and timeliness of the subject. Also, the papers that authors included are rather old (number of papers: 14, of which only 8 are (or considered as) papers and only two are considered new (2019 and 2021), the remaining from 2010, and 1997, 1992... knowing that there is no rule about the age of citations but authors could include the importance of these cited works; a few (relevant!) citations from recent years might be included in order to show that authors read not just at the beginning of the work and then ignored everything afterwards.

Also, some references must be completed and in the correct format: for example [2] is not correct (Kiwelekar A.W., Laddha M.D., Netak L.D., Gandhi S. (2020) An Architectural Perspective of Learning Analytics. In: Virvou M., Alepis E., Tsihrintzis G., Jain L. (eds) Machine Learning Paradigms. Intelligent Systems Reference Library, vol 158. Springer, Cham. https://doi.org/10.1007/978-3-030-13743-4_7); [13]...

Some sentences could have bibliography support (for example: page 2, line 50-52, Introduction: “... the great potential of the rehabilitation system have been recognized by the most developed countries....”; paragraph lines 102-113; lines 138-140 “... almost all projects in the countries worldwide...”; lines 204-206 “... With the development of technology, it is necessary to change the teaching methods in such a way that they use the new technology and teach how to use it. ...”; ...);

Material and Methods:

Lines 77-91: authors could identify the webpage of each European project;

Line 98 “The report... “: identify the report mentioned...

Line 103 “... can, based in this study, ...”: identify the study mentioned...

In my point of view, the name of this section (“Material and Methods”) is not the most appropriate; authors could find a more appropriate title reflecting its content. The first part of this section (lines 79-140), authors describe “Driver rehabilitation programs”: this could be an example of title...

In the second part of this section (lines 142-199), a subsection (2.1) only exist if and only if, a subsection 2.2 exist (at least)...

Lines 142-145, include reference citation...

Theoretical Model for E-Learning Based on Mistakes:

Lines 233-234: “The model in Figure 1 in the structure represents the following steps which will be presented by the algorithm:” -> what is the algorithm that structure is represented in figure 1?

Material and Methods

Section 2 and section 4, both have the same name “Material and Methods” ....?

Lines 318-324; authors state that “There are various other simulators meant to facilitate the same, ...” please identify some or at least one of these “other simulators..”... with references.

Lines 343-351, authors could include a set of figures that demonstrate the application and experimental (simulation drive) part of the project.

Authors mentioned “The questionnaire involving all respondents... “, without any description regarding the development of the questionnaire used to measure and test groups before and after training... authors must validate the instrument or refer how it was developed: new questionnaire or adapted from other works; since based on the data collected based on the questionnaire conclusions decisions will be drawn.

At the end of this section, authors must include the statistical test performed in order to compare behaviors before and after the driving simulation (and not only referring that “and the results were processed by the method of comparison.” (lines 374-375)

Also, authors must refers why they choose to perform the t-test. This test is a parametric test so, a set of assumptions must be verified (authors must include a comment regarding this): t-Test Assumptions: scale of measurement: the data collected follows a continuous or ordinal scale; the data is collected from a representative, randomly selected portion of the total population; data follow a normal distribution; homogeneity of variance of samples.

Lines 368-369: authors state “These tests have 3 levels, each of which represents a different scenario, ...”, and then identify four bullets... or the first bullet do not correspond to a level of drunk glasses? Please check. Also, why was it not considered an alcohol-free level? Or lower than 0.3% of alcohol? Corresponding to a fourth level... ?!

Results

Line 377: “All candidates from both the first and the second group, ...” authors could identify what comprises the first and second group to not to leave any doubts: first group means the drivers participant before the simulation drive and the second group, drives participants after the simulation drive? Or, first and second group correspond to the first two of the different level regarding the three “drunk glasses” scenarios?

Or simply use the same terminology as before: first group is “group I” (age 18-30) and second group is “group II” (age 30-40).

Tables 1 and 2 no identified in the text. Tables 1 and 2: since the information is the same, that is, the set of 10 questions, is the same, can be placed in a single table thus facilitating comparison. The questions can be coded thus reducing the space to occupy in the table. Since the dimension of each group (group I, nI=50, and group II nII=24) is different, the results in the table could be in percentage (a value of 18 in group I give different information as a 18 obtain in group II).

Lines 389-390, authors state that “The results of the questionnaire and the comparison after conducting the experiment for all three levels of driving... ”-> how the three levels were evaluated? The results only show the number of “Yes/No” answers obtained for group I and II before and after driving simulations...  

Line 391, “The changes are obvious in almost all questions.” -> how authors considered as “changes”? how a “change” could be considered positive? From which value it is considered positive change?

Authors could considered a new variable to measure “the change”... difference or other...

According to the nature of the variables (yes/no) the statistical test used do not seems the most appropriate; the t-test is for the parameter mean; but mean of what? Number of yes? The mean of the number of yes obtained on the ten questions? It is not clear which variable is used and under analysis.

For example, using the results available on Table 3, regarding the 95% Confidence Interval of the Difference, what is the meaning of -6,511 (lower limit) and 16,911 (higher limit)? This interval contains the zero value, meaning that the two means could be equal, however the decision identified by authors is that the two means are different... explain?... also, the No answers are the complement of Yes answers, so why study both?

In my opinion this statistical methods is not the correct one to be used. Since the data corresponds to the number of “Yes or No”, the test and to compare two groups, a chi-square test allows to verify relationship.

Author Response

Dear

Please see the attachment

Thank you for suggestions. 

Best regards

Djordje Vranjes

Author Response File: Author Response.pdf

Reviewer 3 Report

Authors aimed to develop an e-learning sustainability model based on errors, which will eliminate the  errors made by a candidate while driving in everyday life. It is expected to contribute to the improvement of quality of drivers rehabilitation training program and learning model for candidates.

Broad comments:

The subject taken is important nowadays. Due to the growing number of road accidents, there is a need to develop such models. The paper structure is problematic. The main strengths of the paper are: i) the importance of the problem taken, ii) methodological correctness, and iii) good discussion of results. The main weaknesses of the paper are: i) poor literature study,  ii) problematic structure of the paper (please see Broad comments);  iii) lack of the clear definition of the aim of this work, contribution to the body of knowledge, hypothesis and scope in the Introduction, iv)  unclear conclusions with references to other works (please see Broad comments). Therefore I recommend to reconsider a paper after major revision.

Specific comments:

INTRODUCTION

The clear definition of the aim of this work, contribution to the body of knowledge and scope are missing. There is also lack of the hypothesis in the introduction (it is only mentioned in the Abstract) Please add the missing information. The presented literature study is poor (only 14 positions). Please add more literature related to the subject taken.

The structure of the paper is problematic: There are two paragraphs 2 and 4 f the same title “Material and Methods”. Please change and clarify the titles, so that they reflect their content.  

5. RESULTS

Due to the fact that this paragraph also contains discussion on results, it is advised to name it as “Results and discussion”

CONCLUSIONS

Please avoid references to other works in the concluding section. It should summarize the authors’ contribution. Moreover, there is no comment on the stated hypothesis in the conclusions. Please add the missing information and correct.

Lines 450-451: “The results in this study also show that older drivers show differences in correlation with the younger drivers [11]”. Why is there a reference to literature in the description of the results of this work? Were they previously published?

Author Response

Dear 

Please see the attachment

Thank you for suggestions.

Best regards

Djordje Vranjes

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

The subject material is presented in a much better way and in a comprehensive manner but authors could improve even more some aspects that I suggest. The topic described is important... but the way the information is presented needs improvement and scientific correction.

Notice that, the authors in the cover letter prepared as response to previous revisions, nothing or little is said about the changes made. “Accepted” or “partially accepted” without any further comment or brief description or justification (acceptance or not) of what was changed/added, it is not enough.

• in section 4. (Research Methodology), (page 8) authors included the hypotheses considered for the research described. Authors stated “The hypothesis of this research is:” identified as “H”, however it is followed by three more sentences identified by “H-3”, “H-2”, “H-3” (check the numbering). Since only one hypothesis “H” is defined, what is the meaning of this three “hypotheses”? also, as it is presented, the first “H-3” (“... there is significant statistical difference.”, it does not appear to make sense, please check... also verify if you want to state only one hypothesis or more than one. Or, use the identification considered at page 16, lines 686: “The auxiliary hypotheses set...”
• “The structure of questionnaire has been prepared based on the objectives of research after reviewing the questions to be answered, as well as the relevant literature.” -> identify the “relevant literature”;
• “In the statistical processing of the results, the Chi square test was used to prove the null hypothesis.” -> this is true for all the statistics tests: to verify if the null hypothesis is true (or not)... so, what the Chi square test will prove? If in the following sentence authors mentioned used “... the T-test for independent samples has been used, as a procedure for assessing the statistical significance of the difference between the two arithmetic means.”
• Based on this paragraph (page 8, lines 396-405), it is a little confusing to realize what statistical test or tests were done and which hypotheses tested. Remember that the hypotheses of the statistical tests have nothing to do with the hypotheses defined by the authors and that the article intends to verify (same name but different concepts). The independent t-test compares the means between two unrelated groups on the same continuous, dependent variable -> so, I do not understand the meaning of the sentence “The research draft in which this method is applied implies the existence of at least one quantitative variable as dependent and one dichotomous variable as independent.” Or rewrite the sentence to make it more understandable, for example identifying the variables (dependent and independent) in the context of the problem.
• begin the section 4.2 with a short introduction to the topic to be presented: for example: “In this section the E-Learning research model is described. In Figure 1 the structure of the model is presented based on the following steps:”
• line 486, page 10, for better reading, put the reference ID [27] right after the author name instead of at the end of the sentence.
• Regarding the age in the two groups considered (end page 11 beginning page 12), candidate with 30 years of age, belong to which group? -> second group: 31-40?!
• Lines 561-566 (page 12): authors state “These tests have 3 levels, each of which represents a different scenario, ...”, and then identify FOUR bullets: “• Under the influence of alcohol with "drunk glasses", as follows: • From 0.3-0.5% of alcohol – I scenario • From 0.5-1.0% of alcohol – II scenario • Over 1.5% of alcohol – III scenario” -> it seems that the “three levels” are related to the three different levels when alcohol is considered and that somehow could interfere with driving (so, because of that, scenario with lower 0.3% of alcohol is not considered?!) -> confusing, check the best way to write what is intended... -> do not confuse: level/scenario.
• Page 14 (lines 604-609): authors mentioned “... used the paired sample t-test in statistical program...”, and mentioned that the statistical value obtained is based on the chi square test -> it seems that authors did implemented the chi-square test for independence (also called Pearson's chi-square test or the chi-square test of association), and used to discover if there is a relationship between two categorical variables... and if the chi square value is lower than the critical value means that the p-value is higher than the significance level considered of 0.05. So, the hypothesis would not be reject, meaning that the behaviour is similar independently of the age group. The chi-square do not compare means... but the strength of association between two categorical variables (it could be: the age group (two levels) and answer (Yes/No)).
• Sincerely, this part of the results and based on the available information, I do not understand...
• H-1 hypothesis do not exist or it is not identified in the text: authors refers to H hypothesis (see line 371 page 8: “the implementation of E-learning model for sustainability of driver rehabilitation programs will influence the change of driver's attitude and the reduction of traffic errors”)?!
• “... by the application of T test of independent samples ...” do not repeat “independent samples t test”... lines 618-619
• Table 3: what means a “Mean = 27,8000” and “=18,7000”? -> the mean of Yes obtained in the 10 questions? And the std. deviation? What would be the expected value of the mean? Do not forget that the two sample are of different size: group I with 50 and group II with 24. So, it would be expected that the mean of Yes in the first group to be higher than in the second one. The sample size is not N=10! -> Analysis based on the values obtained in Table 3 and 4, are not correct.
• Figure 2. caption: Small mistake...
• The values of the correlation coefficient could be lower however the important issue here is to verify if it is significant (that is, the significance level for that particular correlation coefficient).
• Line 679-680: “The value of the determination coefficient between these variables is 3.515831.” -> what this means in the context of the problem? (do not present this value with so many decimal places, two, three at most, are enough). This value corresponds to a percentage!

Author Response

Please see the attachment. Thank you very much.

Author Response File: Author Response.pdf

Reviewer 3 Report

The Authors addressed to many of the Reviewer’s comments and improved the quality of the manuscript. I am really sorry to say that the Authors did not address to the following comment:

INTRODUCTION

The clear definition of the aim of this work, contribution to the body of knowledge and scope are missing. The potential reader may not known what the authors' own contribution is, as there are many references to other works given in the Introduction and it is not clear. Please clearly indicate what is the aim of this work, what is the author's own contribution and describe the structure of the work (what will be analyzed and presented in the following paragraphs). Please add the missing information.

Therefore I recommend paper acceptance after minor revision.

Author Response

INTRODUCTION

The clear definition of the aim of this work, contribution to the body of knowledge and scope are missing. The potential reader may not known what the authors' own contribution is, as there are many references to other works given in the Introduction and it is not clear. Please clearly indicate what is the aim of this work, what is the author's own contribution and describe the structure of the work (what will be analyzed and presented in the following paragraphs). Please add the missing information.

We agree with the comment, and it is unclear to us how we missed it. A correction was made according to the suggestions in the lines of 87-141.

We are grateful for the useful suggestions, with a mutual desire to achieve the highest possible level of quality of work.

Round 3

Reviewer 2 Report

Authors show attention in improving the manuscript, more contextualization, different analysis and better conclusions based on the objective of the paper, but there are still points to be taken into consideration and improvement.

• Abstract: has undergone a lot of change... please verify the words: “short ciklus”; use “E-learning” or “e-learning”; authors mentioned that the “Spearman’s rank correlation coefficient will be used to determine...“, however in the text, see section... still is mentioned t-test, chi-square test (see subsection 4.1.1. Data Processing And Analysis) so, which statistical tests authors used? Also, authors mentioned have used the Spearman test, that is a no parametric test, and Pearson test, that is a parametric test, measuring almost the same: correlation between two variables... it is true that these two tests were used based on different variables, but why using them? What the characteristics of the variables under study? Which one follow a normal distribution and which one do not? Explain and identify (it could be in the text). In the Abstract, authors could only identify that a correlation analysis was performed, not identifying the test (Pearson or Spearman) and in the text, yes, exploring the reason on the use of each one. If possible, indicate the tendency or direction of the correlation observed for each pair of variables. Knowing that “.. small number of observations...” it is not enough to justify, and when you use the Spearman test the size of sample appears to have more data or at least in the same order of magnitude than the sample size when you used the Pearson test.
• Typos: Line 114, check a “.”;
• Lines 115-142: it seems that authors included a description related to how the paper is divided, but it can be resumed, as a summary... to not have repeated information, also, each section could also be identified by the corresponding number.
• Lines 459- : subsection 4.1.1. Data Processing and Analysis: this subsection seems that and everything indicates that this section has not undergone any change after the reformulation of the methodology chosen and adopted by the authors in the manuscript new version (chi-square, t-test... missing to refers the Spearman correlation and also the difference between Pearson and Spearman correlation regarding the characteristics of the data under analysis); in this subsection authors also could include how and it way the use of these two methods will help to verify and to validate the hypotheses stated.
• Section 5. Research Results and Discussion: first sentence (lines 711-712), reformulate (first and second group, repeated... may be referred as the 10 participants of both groups...
• Table 1 and Table 2: verify information in lines 721-728 and 745-751: unformatting information; it is not necessary to include r coefficient and t statistics’ formulas; the ten points in each sample, could be displayed in a graphical form (better than the table form) since, visually, it gives more information, and also can help in explain the results obtained for the coefficient;
• Spearman’ coefficient: no need to include the formula;
• Figure 2 and Figure 3: ok, authors represented the data in a graphical form helping to understand the tendency of the correlation between the two variables under study... however, both could have a different and simple and with all the information needed for a better understanding of the information that the figure must give: title in both axes; also independently of the correlation test used (in this case the non-parametric Spearman test) the figure could have the variable values (and not the ranks)... for the reader the most important is the relation between the variables under study (not the values based on the statistical methodology used).

Author Response

Authors show attention in improving the manuscript, more contextualization, different analysis and better conclusions based on the objective of the paper, but there are still points to be taken into consideration and improvement.

• Abstract: has undergone a lot of change... please verify the words: “short ciklus”; use “E-learning” or “e-learning”; authors mentioned that the “Spearman’s rank correlation coefficient will be used to determine...“, however in the text, see section... still is mentioned t-test, chi-square test (see subsection 4.1.1. Data Processing And Analysis) so, which statistical tests authors used? Also, authors mentioned have used the Spearman test, that is a no parametric test, and Pearson test, that is a parametric test, measuring almost the same: correlation between two variables... it is true that these two tests were used based on different variables, but why using them? What the characteristics of the variables under study? Which one follow a normal distribution and which one do not? Explain and identify (it could be in the text). In the Abstract, authors could only identify that a correlation analysis was performed, not identifying the test (Pearson or Spearman) and in the text, yes, exploring the reason on the use of each one. If possible, indicate the tendency or direction of the correlation observed for each pair of variables. Knowing that “.. small number of observations...” it is not enough to justify, and when you use the Spearman test the size of sample appears to have more data or at least in the same order of magnitude than the sample size when you used the Pearson test.
• The above suggestions were made in the part of the Abstract, as well as in chapter 4.1.1. where corrections of the mentioned suggestions were made.
• In the Abstract, we identified that we performed a correlation analysis without identifying tests (Pearson or Spearman),
• Typos: Line 114, check a “.”;
• Lines 115-142: it seems that authors included a description related to how the paper is divided, but it can be resumed, as a summary... to not have repeated information, also, each section could also be identified by the corresponding number.
• In the introductory part at the end of the text we have given the structure of the work according to the recommendation of the reviewer 1, but we acted according to your recommendation and put in chapter 1.1. Work structures
• Lines 459- : subsection 4.1.1. Data Processing and Analysis: this subsection seems that and everything indicates that this section has not undergone any change after the reformulation of the methodology chosen and adopted by the authors in the manuscript new version (chi-square, t-test... missing to refers the Spearman correlation and also the difference between Pearson and Spearman correlation regarding the characteristics of the data under analysis); in this subsection authors also could include how and it way the use of these two methods will help to verify and to validate the hypotheses stated.
• In the mentioned chapter 4.1.1. Corrections were made according to the new concept of work with new explanations.
• Section 5. Research Results and Discussion: first sentence (lines 711-712), reformulate (first and second group, repeated... may be referred as the 10 participants of both groups...
• We made the corrections (of both groups).
• Table 1 and Table 2: verify information in lines 721-728 and 745-751: unformatting information; it is not necessary to include r coefficient and t statistics’ formulas; the ten points in each sample, could be displayed in a graphical form (better than the table form) since, visually, it gives more information, and also can help in explain the results obtained for the coefficient;

We exclude the formulas and use guidlines

• Spearman’ coefficient: no need to include the formula;

• We exclude the formulas and use guidlines

• Figure 2 and Figure 3: ok, authors represented the data in a graphical form helping to understand the tendency of the correlation between the two variables under study... however, both could have a different and simple and with all the information needed for a better understanding of the information that the figure must give: title in both axes; also independently of the correlation test used (in this case the non-parametric Spearman test) the figure could have the variable values (and not the ranks)... for the reader the most important is the relation between the variables under study (not the values based on the statistical methodology used).

Corrections were made to the previously suggested suggestions in addition to clarifying the diagram

This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.

Round 1

Reviewer 1 Report

What model?

Reviewer 2 Report

The subject material is presented in a somewhat confused manner, so, I suggest authors to improve or clarify some aspects:

The present work is contextualized with some current or recent bibliography, however authors could also mentioned works developed by different authors.

Abstract:

• What does the “ -* “ symbol mean? Sentence in lines 26-29, long and confusing;
• “... technique of paired T test of independence... ”: usually the test is identified as “... paired sample t-test...”; and if the test is for paired sample they are not independent, one thing or the other, or it is not clear what authors want to say;

Introduction:

• In the number reference system, a number is added in square brackets in the appropriate place in the text, starting the numbering from 1, and then numbered consecutively. The bibliography of the work is arranged by the order in which the citations appear in the text. The first source you cite in your text receives the number 1, the second number 2, and so on... . Also, some references must be completed and in the correct format;
• Works resulting of research can be found and giving a broad and emerging perspectives regarding the topic under study in the paper. So, it is expected an inclusion of more papers describing experiences carried out even in different countries and in different sectors, but that show the relevance and timeliness of the subject;
• The text is legible but the English must be improved: the writing syntax needs to be improved (for example: page 1, line 44, Introduction: “So, the system is this program is directed to the exactly...”);
• Some sentences could have bibliography support (for example: page 2, line 48, Introduction: “The most developed countries have recognizes the need...”; line 105; line 135; line 196-198; ...)
• Line 55, “From June, 2012, the Agency organizes the seminars on the improvement of knowledge... “ - what Agency?
• At the end of the Introduction, the objectives must be defined (in the form of a question or hypothesis); and then the methodology followed;

Methodology:

• Lines 77-91: authors could identify the webpage of each European project;
• Verify the meaning of sentences on lines 152-156; it seem to make no sense;
• Some of information given could be also in form of graphs (lines 162-166, for example), facilitating the analysis and description; and/or in Table,

Theoretical Model for E-Learning Based on Mistakes:

• Figure 1: information in lines 230-248 could be part of caption of the figure for example, on the right side, to facilitate the understanding;

Material and Methods:

• Rename “Methodology” or “Material and Methods” – check the name of these two sections (mainly the first one);
• Figure 2 is not identified in the text;
• Authors mentioned “The questionnaire involving all respondents... “ for the first time in line 356. And later, on line 374, authors present a summary of the results, without any description regarding the development of the questionnaire used to measure and test groups before and after training... if the questionnaire was and it is considered as an important instrument to measure and analyze the presented e-model for sustainability of the program for drivers rehabilitation, it must be described and authors must validate the instrument or refer how it was developed: new questionnaire or adapted from other works;

Results:

• Tables 1 and 2 no identified in the text; format first row of tables;
• as statistical test used is a parametric one, authors verified the normality of data?
• “The changes are obvious in almost all questions of the poll. The changes are obvious in almost all questions of the poll.” – what authors considered as “changes”? how a “change” could be considered a positive? From which value it is considered positive change? Are all considered as significant? ...
• In line 394, authors mentioned having used the t-test for independent samples... however, in abstract something different is identified;
• According to the nature of the variables (yes/no) the statistical test used do not seems the most appropriate; the t-test is for the means; but means of what? Number of yes? It is not clear which variable is used and under analysis.

Reviewer 3 Report

Authors mentioned about size of control and experimental groups at the following sentence “size of the groups must be large enough to be able to demonstrate a significant differences”. It is not a precise term, as the reader may infer from it that group size is decisive for proving significant differences. I propose to correct of this sentence.

In following sentence "General aim of the improvement process of drivers’ behaviors and drivers rehabil-59 itation system is for all employees to adapt to the scheme of civilized behaviors on the 60 road!" Authors use exclamation mark "!". I suggest resigning from the sign expressing the emotions of the authors.

Reviewer 4 Report

The paper appears very thorough, with plenty of detail.  The main problem I see is that the English is not great.  On line 46, ‘brake the law’ should be ‘break the law’.  The four categories defined in lines 66–71 under bullet points may be better enumerated.  The attributes of the groups starting on line 120 should ideally include randomly selected.  Line 168 onwards seems to be a sentence split over paragraphs.  A sentence is repeated on lines 379–380.  We normally write ‘t-test’, rather than ‘T test’.  The conclusions are derived from a poll.  Far better would be revealed (rather than stated) evidence, such as fewer convictions/accidents in the future.  Of course, in practice this data is likely not available, but it may be worth mentioning.