Exploring Cyclists’ Behavior, Traffic Safety Literacy, and Crash Occurrence in Latvia
Round 1
Reviewer 1 Report
Comments and Suggestions for AuthorsThis study assesses the relationships among self-repotted cyclists’ behavior, traffic safety literacy, and their cycling crash involvement rates. A total of 299 cyclists aged M=32.8 from across Latvia (42% females) participated in an online survey, which comprised questions regarding respondents’ demographics, frequency of riding, and the number of crashes in the past five years. The Cycling Behavior Questionnaire (CBQ) and the Cyclist Risk Perception and Regulation Scale (RPRS) were applied to assess cyclists’ behavior patterns and traffic safety literacy. Results: It is noticed that cyclists frequently engage in riding errors and traffic violations while cycling. Those who exhibit more antisocial behavior patterns are also more likely to be involved in road crashes. Conversely, cyclists with greater positive behavior rates more often also tend to possess better knowledge of traffic rules and exhibit a heightened risk perception, indicating a greater awareness of road traffic safety.
On page 2, it is written 'However, 45 it must be noted that in recent years, the decrease in the number of accidents has not been 46 sufficiently positive.' what is meant by this? Please explain.
Please cite reference uniformly eiyher using numbers or years.
What is the median age in your survey?
Were there any missing values? If yes, how are they treated? Are these discarded?
From th emANOVA table, it looks that riding errors, traffic violations are only significant. Have you further investigated these two factors?
Should not the statment ' traffic violations have 1.421 times higher odds' be as ' traffic violations have 42.1% higher odds'? The same is true for 'cyclists are 1.031 times more'.
In Figure 1, I would suggest to remove the OR circles which are less than one.
Please discuss future work in the conclusion section.
Author Response
Dear Reviewer,
Thank you very much for your thoughtful and detailed feedback on the manuscript. I greatly appreciate the time and effort you have invested in providing your comments and suggestions, all of which have helped to improve the quality and clarity of the work. In this response, I have addressed each of your questions, comments, and suggestions in detail and in the order provided. I hope my revisions and clarifications meet your expectations.
Q: On page 2, it is written 'However, 45 it must be noted that in recent years, the decrease in the number of accidents has not been 46 sufficiently positive.' what is meant by this? Please explain.
A: Thank you for your question regarding the statement on page 2. What I intended to convey is that, while there has been some reduction in cycling accidents in recent years, the progress has not been as significant or substantial as one might have hoped or expected. In other words, although the number of accidents has decreased, the rate of reduction has been slower or smaller than anticipated, and further improvement is still necessary. Additionally, on the same page, in lines 49 to 55, I have included statistics that illustrate the troubling situation in Latvia regarding accidents involving cyclists, further emphasizing the need for more substantial progress.
Q: Please cite reference uniformly either using numbers or years.
A: Thank you for noticing this. I have revised the citations to ensure they are now uniformly formatted, as per your recommendation.
Q: What is the median age in your survey?
A: I have added the median age of the participants, which is 31 years, and this information can now be found on line 153.
Q: Were there any missing values? If yes, how are they treated? Are these discarded?
A: Thank you for your question. Yes, there were a small number of missing values in the surveys, and a total of 11 surveys were deemed invalid and subsequently discarded. We took great care in designing the survey to minimize missing data and aimed to gather as many valid responses as possible. The surveys were mostly considered invalid because the respondents were either under 18 years of age and completed the survey accidentally, or they were not permanent residents of Latvia.
Q: From the MANOVA table, it looks that riding errors, traffic violations are only significant. Have you further investigated these two factors?
A: Thank you for your insightful question. This research represents the initial stage of our study and the first publication of our data. While the MANOVA table shows that riding errors and traffic violations are significant factors, we have not yet delved deeper into these specific aspects. However, we do recognize their importance and plan to investigate them further in future research. These factors are indeed intriguing and could offer valuable insights into cyclist safety and behavior, which we look forward to exploring in more depth as our work progresses.
Q: Should not the statment ' traffic violations have 1.421 times higher odds' be as ' traffic violations have 42.1% higher odds'? The same is true for 'cyclists are 1.031 times more'.
A: Thank you for your valuable remark. Regarding your suggestion on the phrasing, you are absolutely correct, and your suggestion is much appreciated. After thoroughly reviewing the analysis and considering the reviewers' feedback, we decided to remove the regression analysis from the manuscript. The sample size was too small, as indicated by the G*Power analysis, and the results from the regression were not statistically significant enough to draw meaningful conclusions. Instead, we opted to proceed with a pathway analysis to explore the individual indicators more effectively and gain deeper insights into the relationships between the variables. Instead, we replaced it with pathway analysis and SEM model (see line 441-471), which we believe is more suitable for this research context. Based on it we have added more paragraph in introduction part, discussion, conclusions and limitation (highlighted in yellow).
Q: Please discuss future work in the conclusion section.
A: Thank you for your note regarding the discussion of future work in the conclusion section. While I have addressed future work in the discussion section, I have also incorporated additional points in the conclusion as you recommended. These can be found in lines 726 to 734. Your feedback has been invaluable in enhancing the clarity and comprehensiveness of this section, and I am grateful for your insights.
All the mentioned additions have been highlighted in yellow in the text. Thank you so much for your valuable contribution to enhancing the quality of the publication!
Kind regards,
Katrina
Author Response File: 
Author Response.docx
Reviewer 2 Report
Comments and Suggestions for AuthorsThe article entitled "Exploring Cyclists’ Behavior, Traffic Safety Literacy, and Crash Occurrence in Latvia" aimed to assess the relationships among self-reported cyclists’ behaviour, traffic safety literacy, and their cycling crash involvement rates. The manuscript presents an interesting topic, but, independently, below there are a few suggestions that could be taken into account by the Authors to complete the paper:
1. After reading the first part of the paper and taking into account the aim of it, it is not clear to me which road traffic incidents in relation to cyclists’ behaviour the paper is concerned with, i.e., whether these are incidents involving other cyclists (and only cyclists) or road traffic accidents involving both cyclists and car drivers ("At a practical level, identifying cyclists’ behavior and its correlation with road traffic crashes could raise awareness …", lines 116-117). This is especially true since the Authors provide such a distinction later in the work, i.e., "… older cyclists are more likely to crash without a motor vehicle involvement …" (lines 541-542).
In view of the above, in the Materials and Methods section it would be worthwhile to specify which road incidents are referred to, i.e., those involving only cyclists or also car drivers ("Over the past five years, 156 cyclists (52.2%) reported no accidents while riding a bicycle, while 142 participants (47.5%) indicated they had been involved in at least one crash", lines 149-151).
I suppose this is obvious to the Authors, whereas it may not be entirely clear to the reader.
2. In the Limitations of the study section, it would be useful to pick up a theme related to the use of methods based on self-reporting to study selected variables, as well as the sample size and the possibilities to generalise the conclusions of the research to the whole population of cyclists.
3. In the Practical implications of the study section, it is crucial to extend the theme only mentioned by the Authors and to describe not only that (i.e., "… it can be expected that promoting cyclists’ risk perception (i.e., increasing their awareness of the potential risks and consequences of different cycling behaviors) could enhance both safety and positive behavior among cyclists", lines 604-606), but also how the perception of risk in cyclists should be increased, through which exact actions, also with reference to existing activities and efforts undertaken towards this group of road users in other countries.
Also, the second paragraph in this section (lines 608-612) definitely needs to be elaborated and detailed, i.e., in what exact ways and through what activities "address positive riding habits".
4. There are some minor linguistic deficiencies in the paper that require correction (e.g., "self-repotted", line 14; "… am age-based perspective", line 526).
I hope that the above-mentioned suggestions will be helpful in improving the paper, which, in my opinion, is both interesting and carefully prepared.
Best regards,
Reviewer
Comments on the Quality of English LanguageThe quality of the English language is good, in my opinion. However, there are a few linguistic errors throughout the paper that need improvement (e.g., "self-repotted", line 14; "… am age-based perspective", line 526).
Best regards,
Reviewer
Author Response
Dear Reviewer,
Thank you very much for your thoughtful and detailed feedback on the manuscript. I greatly appreciate the time and effort you have invested in providing your comments and suggestions, all of which have helped to improve the quality and clarity of the work. In this response, I have addressed each of your questions, comments, and suggestions in detail and in the order provided. I hope my revisions and clarifications meet your expectations.
Q: In the Materials and Methods section it would be worthwhile to specify which road incidents are referred to, i.e., those involving only cyclists or also car drivers ("Over the past five years, 156 cyclists (52.2%) reported no accidents while riding a bicycle, while 142 participants (47.5%) indicated they had been involved in at least one crash", lines 149-151). I suppose this is obvious to the Authors, whereas it may not be entirely clear to the reader.
A: Thank you for your comment regarding the clarity of the road incidents discussed in the Materials and Methods section. To address your concern, I have clarified that the analysis encompasses various types of crashes, including those involving only cyclists, incidents with other vehicles, and crashes involving stationary objects. Specifically, the data indicate that over the past five years, 142 participants (47.5%) indicated they had been involved in at least one crash, regardless of whether these incidents involved other vehicles or objects. I appreciate your suggestion to enhance the clarity of this section. I have added a statement in lines 169-170 to further specify the types of incidents considered in our analysis.
Q: In the Limitations of the study section, it would be useful to pick up a theme related to the use of methods based on self-reporting to study selected variables, as well as the sample size and the possibilities to generalise the conclusions of the research to the whole population of cyclists.
A: Thank you for your valuable feedback. I have added a paragraph in the Limitations of the Study section discussing the limitations of using self-reported measurements in this research (see lines 578-586). Additionally, I addressed the representativeness of the sample size, noting that the number of cyclists included in our study reflects the relatively small population size in Latvia compared to other EU countries. While we acknowledge that a larger sample would enhance the robustness of our findings, we believe that these initial results provide a meaningful foundation for future research. Expanding the sample size is indeed part of our plans moving forward, as we aim to deepen our understanding of cycling behavior in Latvia.
For this study, I have also included a G*Power calculation indicating that, for MANCOVA analysis, we needed 134 cyclists per group. Since we exceeded this number, the sample size was deemed acceptable for this analysis (see lines 388-391). Regarding the regression analysis, we found the sample size to be too small, and therefore, we decided to exclude it from the article, as the results were not significant enough. Instead, we replaced it with pathway analysis, which we believe is much more suitable for this research context. Thank you again for your suggestions, which have significantly improved the clarity and comprehensiveness of the manuscript. The analysis itself, including the path analysis, which replaced the regression analysis, can be seen in lines 440-471. Upon reflection, we concluded that regression analysis was not suitable for this study, and we are confident that the current approach offers more meaningful insights and better supports the research objectives. Thank you for the suggestions!
Q: In the Practical implications of the study section, it is crucial to extend the theme only mentioned by the Authors and to describe not only that (i.e., "… it can be expected that promoting cyclists’ risk perception (i.e., increasing their awareness of the potential risks and consequences of different cycling behaviors) could enhance both safety and positive behavior among cyclists", lines 604-606), but also how the perception of risk in cyclists should be increased, through which exact actions, also with reference to existing activities and efforts undertaken towards this group of road users in other countries.
Also, the second paragraph in this section (lines 608-612) definitely needs to be elaborated and detailed, i.e., in what exact ways and through what activities "address positive riding habits".
A: Thank you for your insightful feedback regarding the practical implications of the study section. I have addressed your comment by adding a paragraph that emphasizes the importance of promoting risk perception among Latvian cyclists (see lines 717-721). In this addition, I discuss the implementation of targeted educational campaigns inspired by successful initiatives in other countries. These campaigns could include interactive training sessions, workshops, and partnerships with local cycling organizations to foster a culture of safety among cyclists (see lines 726-733).
Q: There are some minor linguistic deficiencies in the paper that require correction (e.g., "self-repotted", line 14; "… am age-based perspective", line 526).
I hope that the above-mentioned suggestions will be helpful in improving the paper, which, in my opinion, is both interesting and carefully prepared.
A: Thank you for your careful review and for highlighting the minor linguistic deficiencies in the paper. I have made the necessary corrections to ensure clarity and accuracy in the language used throughout the manuscript (please see coloured in yellow).
I genuinely appreciate all the valuable suggestions and constructive feedback you have provided throughout this process. Your insights have significantly enhanced the quality of my work, enabling me to refine my arguments and clarify important points.
All the mentioned additions have been highlighted in yellow in the text. Thank you so much for your valuable contribution to enhancing the quality of the publication!
Reviewer 3 Report
Comments and Suggestions for AuthorsDear Authors,
you have prepared an interesting study, to which I have the following comments:
Comments:
Introduction:
In the line 32 you write: “For many residents of Latvia, cycling to work is becoming increasingly popular each year, offering a range of benefits.”
l And what about, for example, students, children commuting to school?
It would also be appropriate to provide statistics on the number of accidents in Latvia as well as cyclist fatalities.
I would also emphasize the lack of research, not only regarding information about Latvia, but also about new knowledge that will expand the field of cyclist safety
Methodology:
Since the online survey captured a sample of residents aged between 18 and 76, the discussion should mention that it would be appropriate to analyze and obtain data on younger persons as well, for example, children and teenagers.
We also need to find out if urban and rural residents behave differently. Was there any information in the survey?
The sample of 299 cyclists is not very large, from the point of view of the sample size methodology and taking into account the size of Latvia, it would be more appropriate to get more people. I find it rather short that it was open for 9 months. In the Data collection procedure chapter, state which promotional channels you used? For example only a website, Facebook group, etc.
It is also important whether the respondents were regular cyclists or only occasionally. Whether they cycled to work, or it was a ride associated with relaxation or sports.
In the study, I would expect the participants to be divided into several groups, which could be better analyzed in relation to their behavior during cycling. The statistical analysis is quite simple.
It would also be appropriate to add questions, or the parameters you asked about in the survey in subchapter 2.3.1.
It would also be interesting to state what are the official data on the modal split for Latvia, or the other Baltic countries with regard to cyclists.
Perhaps it would be interesting to bring your opinion into the discussion, what differentiates the behavior of the Baltic countries, for example, from the results of the studies you mentioned.
In conclusion, I would recommend supplementing the information I mentioned above and linking it with outputs that have already been implemented in this area.
Author Response
Dear Reviewer,
Thank you very much for your thoughtful and detailed feedback on the manuscript. I greatly appreciate the time and effort you have invested in providing your comments and suggestions, all of which have helped to improve the quality and clarity of the work. In this response, I have addressed each of your questions, comments, and suggestions in detail and in the order provided. I hope my revisions and clarifications meet your expectations.
Q: In the line 32 you write: “For many residents of Latvia, cycling to work is becoming increasingly popular each year, offering a range of benefits.”
And what about, for example, students, children commuting to school?
It would also be appropriate to provide statistics on the number of accidents in Latvia as well as cyclist fatalities.
I would also emphasize the lack of research, not only regarding information about Latvia, but also about new knowledge that will expand the field of cyclist safety.
A: Thank you for your insightful suggestions. I have revised the text to include not only residents commuting to work but also students, as cycling is becoming increasingly popular among all age groups (see lines 34-35). Additionally, I have provided relevant statistics on cycling-related accidents and fatalities in Latvia to give a clearer context of the current safety situation (see lines 49-55). I would like to point out that within the scope of this research, the sample consisted of cyclists from the age of 18.
Furthermore, I have addressed the issue of the lack of research and the need for new knowledge that will expand the field of cyclist safety in the Practical Implications of the Study section. Thank you for the suggestions.
Q: Since the online survey captured a sample of residents aged between 18 and 76, the discussion should mention that it would be appropriate to analyze and obtain data on younger persons as well, for example, children and teenagers.
A: Thank you for your suggestion. In this research, we specifically aimed to analyze the behavior of cyclists aged 18 and older, as our primary focus was on adult cyclists. However, we acknowledge that including children and teenagers in future studies would provide a valuable perspective. Given that this is the first study of its kind in Latvia, we see great potential for future work to explore the cycling habits of younger age groups. Your insight will certainly help guide our research directions moving forward.
Q: We also need to find out if urban and rural residents behave differently. Was there any information in the survey?
A: Thank you for your question regarding the potential differences in cycling behavior between urban and rural residents. Yes, we collected data from cyclists across Latvia, encompassing both urban and rural areas. The statistics regarding the sample can be found in lines 154-157.
Q: As mentioned in the article, we had participants from various locations, including the capital city and smaller towns. However, in our analysis, we did not find significant differences in the behavior of residents from the capital compared to those from other cities. Consequently, this aspect was not explored in-depth in the paper, but it remains a valuable avenue for future research. Thank you for highlighting this important consideration.
A: Thank you for your thoughtful comments regarding the sample size and data collection procedure. We acknowledge that a larger sample would provide a more robust analysis, and we certainly aim to increase the sample size in future studies. However, considering the relatively small population of Latvia, we believe that the sample of 299 cyclists is representative for initial research in this field. For this study, I have included a G*Power calculation indicating that, for MANCOVA analysis, we needed 134 cyclists per group. Since we exceeded this number, the sample size was deemed acceptable for this analysis (see lines 399-402). Regarding the regression analysis, we found the sample size to be too small and decided to exclude it from the article, as the results were not significant enough. Instead, we replaced it with pathway analysis and SEM model (see line 441-471), which we believe is more suitable for this research context. Based on it we have added more paragraph in introduction part, discussion, conclusions and limitation (highlighted in yellow).
In terms of data collection, we utilized various promotional channels, including Facebook cycling groups and emails sent to relevant institutions, to reach a broad range of cyclists. We appreciate your input and will certainly aim for a broader reach in future research.
Q: It is also important whether the respondents were regular cyclists or only occasionally. Whether they cycled to work, or it was a ride associated with relaxation or sports.
In the study, I would expect the participants to be divided into several groups, which could be better analyzed in relation to their behavior during cycling. The statistical analysis is quite simple.It would also be appropriate to add questions, or the parameters you asked about in the survey in subchapter 2.3.1.
A: Thank you for your excellent suggestion regarding the inclusion of questions or parameters in subchapter 2.3.1. In our survey, we did collect data on the frequency and purpose of cycling, such as commuting versus recreational purposes. However, we did not categorize participants into distinct groups for this particular analysis.
Your recommendation to group cyclists based on their frequency and purpose of cycling is a valuable one, and we will certainly consider this for further analysis. Additionally, the parameters related to cyclist behaviors are described in the Results section under the subheading Traffic violations, Riding errors, Positive behaviors, and Traffic safety literacy. This section details the specific behaviors that were directly queried in the questionnaires.
We recognize that expanding our analysis by dividing participants into more specific groups based on their cycling habits would provide richer insights. We plan to explore these distinctions in future research, as we believe it will enhance our understanding of cyclist behavior.
Q: It would also be interesting to state what are the official data on the modal split for Latvia, or the other Baltic countries with regard to cyclists.
A: Thank you for suggestion regarding the official data on the modal split for Latvia and the other Baltic countries concerning cyclists. Unfortunately, there is currently limited research available on this topic, particularly regarding specific behavioral aspects of cycling. The existing statistics for the Baltic states are generally limited to the numbers of accidents and fatalities among cyclists on the road. Within the framework of the Baltic region, this kind of methodology has not been employed in Lithuania or Estonia, which highlights the uniqueness of our study. I noted after the reviews that some statements regarding the Baltic states could have been misleading, and I have corrected these throughout the article to provide greater clarity.
To address the safety context in Latvia, I have added statistics on accidents, deaths, and injuries related to cycling in lines 49-55. This study represents the first attempt to investigate such a complex aspect of cyclist behavior and safety in Latvia, and we agree that incorporating more data on modal split would be valuable for future research.
Q: Perhaps it would be interesting to bring your opinion into the discussion, what differentiates the behavior of the Baltic countries, for example, from the results of the studies you mentioned.
In conclusion, I would recommend supplementing the information I mentioned above and linking it with outputs that have already been implemented in this area.
A: We agree that incorporating such a discussion would provide valuable context and enhance the depth of our analysis. In future discussions, we will aim to include more in-depth comparisons with the results of studies conducted in other countries, particularly focusing on the cultural, infrastructural, and policy-related factors that may influence these differences in cyclist behavior. Additionally, we will strive to supplement the information you mentioned and link it to existing outputs in this area to create a more cohesive narrative.
Thank you once again for your insightful feedback, which will certainly guide our future research endeavors.
All the mentioned additions have been highlighted in yellow in the text. Thank you so much for your valuable contribution to enhancing the quality of the publication!
Reviewer 4 Report
Comments and Suggestions for AuthorsThis is another paper in what appears to be a series of 'across the globe' surveys of bike riders using the RPRS and also the CBQ. It appears to be a reasonable study that further supports the hypothesis, developed previously by the Useche group, that traffic violations and riding errors are good predictors of the likelihood of crashes when riding. The part that does not appear to make sense is the conclusion that age is positively correlated with the likelihood of crashes. First, the actual data show only a very weak correlation (0.094; ns) between rider age and self-reported crashes. In addition, the authors report that the logit analysis revealed a significant effect of rider age on OR, but the change in OR is only 1.03 per year. Is this a case of a statistically reliable effect that is nonetheless very small in magnitude? Further, the idea that old drivers in Latvia (present study) are more likely to be involved in crash seems at odds with previous studies by the Useche group (see Safety Science, 2018; Data in Brief, 2019), but the authors do not fully address this apparent discrepancy.
The manuscript has many instances where the quality of the English language could be improved (see below for the details).
In sum, the paper is an interesting read that furthers the hypothesis that the RPRS questionnaire can be used to identify relationships between rider bad behaviors (traffic violations and riding errors) and crash risk, but not good rider behaviors (speed and headway control, checking intersections) and crash risk. Further, as mentioned above, the authors do not provide an explanation for the apparent discrepancy between their Latvian riders (age is associated with more crashes) and the riders from other countries (age is associated with fewer crashes).
Comments on the Quality of English LanguageComments on the Quality of the English Language, and Other Relatively Minor Comments
Line 14: self-reported cyclists’
Line 21: the term anti-social behavior is imprecise; I think you are referring to traffic violations and riding errors
Line 24: delete the phrase all in sum
Line 53: do you mean reduces their severity?
Line 65: might discourage people from
Line 68: fatality when involved in a crash
Lines 78-80. Cyclists’ behavior. Recent studies show
Lines 110-111: aware of potential risks may reduce
Lines 122-125: This study aimed to assess, in Latvia, the relationships
Line 143-144: occupation, the majority of the research respondents reported
Lines 149-151: 48% of subjects being involved in a crash seems like a lot; is this consistent with other riders in other studies?
Line 159: they were willing to respond
Lines 169-214: Perhaps you should just insert the RPRS and QBC scales?
Line 215: Statistical
Lines 249-251, and Table 1: Would you consider inserting a column in the table that provided Mean Scores from other papers as a reference? That would make your statement in lines 149-151 more concrete.
Line 284: holes or obstacles
Lines 293-295: Again, what is the reference value for ‘insufficient’ frequency of helmet use? One could argue that anything < 100% is insufficient.
Lines 355-358: Do the authors feel that all of the behaviors identified as traffic violations should be equally weighted? Is it possible that some violations are more dangerous than others?
Line 377: crash occurrence was significantly influenced by the combined
Line 440 (redundancy): behavioral patterns of behavior among
Line 481: Also, except for variations in
Line 510: notable findings in the present study
Lines 526-549: The authors offer no explanation for the discrepancy between their results (age is associated with more crashes) and previous work.
Line 551: covered several different
Lines 586-587: crash occurrences over the past 5 years among Latvian cyclists
Lines 589-591: What is the evidence for this statement? Can you provide the reference data, AND is the current data from Latvian subjects significantly different?
Author Response
Dear Reviewer,
Thank you very much for your thoughtful and detailed feedback on the manuscript. I greatly appreciate the time and effort you have invested in providing your comments and suggestions, all of which have helped to improve the quality and clarity of the work. In this response, I have addressed each of your questions, comments, and suggestions in detail and in the order provided. I hope my revisions and clarifications meet your expectations.
Q: The part that does not appear to make sense is the conclusion that age is positively correlated with the likelihood of crashes. First, the actual data show only a very weak correlation (0.094; ns) between rider age and self-reported crashes. In addition, the authors report that the logit analysis revealed a significant effect of rider age on OR, but the change in OR is only 1.03 per year. Is this a case of a statistically reliable effect that is nonetheless very small in magnitude? Further, the idea that old drivers in Latvia (present study) are more likely to be involved in crash seems at odds with previous studies by the Useche group (see Safety Science, 2018; Data in Brief, 2019), but the authors do not fully address this apparent discrepancy.
A: Thank you very much for your valuable feedback. We greatly appreciate your detailed insights and suggestions.
We understand your concern regarding the weak correlation between rider age and self-reported crashes, as well as the small effect size observed in our logistic regression analysis. As noted in the manuscript, while older cyclists often display more responsible behaviors on the road such as better adherence to traffic rules, they still appear to experience more crashes compared to younger cyclists. This finding aligns with previous research (e.g., Engber et al., 2018; Ayuso et al., 2020), which attributes the higher crash likelihood to factors such as cognitive decline, increased physical fragility, and riding errors, rather than to risky behavior.
We have conducted and added a G*Power calculation for the MANCOVA analysis, which indicated that a minimum of 134 cyclists per group would be required. As we exceeded this sample size, we deemed the data sufficient for this analysis (see lines 399-402). However, during the regression analysis, we found the sample size to be insufficient, and due to the lack of significant results, we chose to exclude it from the manuscript. Instead, we opted for a pathway analysis (SEM model (see lines 251-264; 440-471)), which we believe provides a more comprehensive and appropriate approach for our research context. The analysis itself, including the path analysis, which replaced the regression analysis, can be seen in lines 440-471. Upon reflection, we concluded that regression analysis was not suitable for this study, and we are confident that the current approach offers more meaningful insights and better supports the research objectives. Thank you for the suggestions!
After reviewers' suggestions and careful review, we concluded that regression analysis was not an appropriate method in the context of this study. I think you have created a more successful solution now.
In addition, we have addressed the age-related aspects of risky behavior and crash history in more detail in section Risky behavior and crash history: an age-based perspective. We have made adjustments to this section for added clarity (see lines 578-579 and 601-606), incorporating a more nuanced discussion of the relationship between age and crash likelihood. We also acknowledge that other researchers have found similar trends regarding age, which is a compelling area of investigation. These findings have led us to include further discussion on the association between age and behavior, as demonstrated by the Pearson correlations showing age-related associations across various behavior and risk perception scales.
Once again, we sincerely thank you for your thoughtful comments, which have significantly enhanced the clarity and depth of our manuscript.
Q: The manuscript has many instances where the quality of the English language could be improved (see below for the details).
In sum, the paper is an interesting read that furthers the hypothesis that the RPRS questionnaire can be used to identify relationships between rider bad behaviors (traffic violations and riding errors) and crash risk, but not good rider behaviors (speed and headway control, checking intersections) and crash risk. Further, as mentioned above, the authors do not provide an explanation for the apparent discrepancy between their Latvian riders (age is associated with more crashes) and the riders from other countries (age is associated with fewer crashes).
A: We greatly appreciate your thoughtful comments and suggestions. We have thoroughly revised the manuscript to improve the language quality, addressing the specific aspects you pointed out. These changes have been implemented throughout the text, and we have highlighted all the improvements in yellow. We believe these revisions have enhanced the readability and clarity of our work.
We also acknowledge your comments regarding the apparent discrepancy between our findings (where age is associated with more crashes among Latvian cyclists) and studies from other countries, which suggest that age is associated with fewer crashes. As discussed in our revised manuscript, we believe this difference may be attributed to specific factors such as increased physical fragility and cognitive decline in older cyclists, as well as potential differences in infrastructure, cycling culture, and rider behavior between regions. We appreciate your positive feedback regarding the use of the RPRS questionnaire. We are glad that you found the study interesting in its exploration of the relationship between cyclists’ behaviors and crash risk.
After carefully reviewing your comments, we realized that the original conclusion did not fully align with the aim of our research or the analyses conducted. Based on this understanding, we have significantly revised the conclusion to better reflect the core findings of our study. Specifically, we have removed the conclusion regarding the Baltic states, as we acknowledged that this was beyond the scope of our research. Unfortunately, there is currently limited research available on this topic, especially in terms of specific behavioral aspects of cycling. The existing data for the Baltic states generally focus only on accident and fatality numbers among cyclists, without addressing detailed behavioral factors. Furthermore, within the framework of the Baltic region, this kind of methodology has not yet been applied in Lithuania or Estonia, which highlights the uniqueness of our study. After reflecting on the reviews, I also recognized that some statements regarding the Baltic states could have been potentially misleading, and I have corrected these throughout the article to ensure greater clarity. We have also enhanced the conclusion section to provide a more focused discussion (see lines 651-659), aligning the conclusion with our research goals. In addition, we improved the sections regarding the practical implications of our findings (see lines 688-710 and 717-721; 726-733), clarifying how the results of our study could be applied in real-world contexts.
Once again, thank you for your valuable feedback, which has greatly contributed to the improvement of our manuscript. We hope that these revisions meet your expectations and address all of your concerns.
All the mentioned additions have been highlighted in yellow in the text. Thank you so much for your valuable contribution to enhancing the quality of the publication!
Reviewer 5 Report
Comments and Suggestions for AuthorsThis paper investigates the associations among cyclists’ self-reported behavior, traffic safety knowledge, and crash involvement in Latvia. In 2023 and 2024, 300 cyclists were surveyed online using a cycling behavior questionnaire and a cyclist risk perception and regulation scale. In the current form, the data and analysis do not respond well to the objective. Overall, several major issues should be carefully considered in the revision.
Any new findings, including Baltic contexts, should be clarified. This should be supported by reviewing similar studies in the Baltic in the introduction and discussions later in Section 4.
The analysis is basic, with limited results. There were only a few significant bivariate relationships, and the logistic model had only two significant variables, suggesting the method was not suitable. It is strongly recommended to look at individual behaviors as the combined constructs do not seem to provide sufficient evidence or new findings.
Figure 1 is inappropriate. There is no such model in this paper. It appears to include results of regression and correlation test, which is confusing and misleading. A SEM model, for example, could be applied if this is the aim of the research.
In the conclusion, comparisons with other Baltic countries are stated without any details or discussion in previous sections.
The writing generally needs to be improved with formal language, supported by references in many places. For example, statistics about cycling and cyclist crashes in Latvia were missing.
Author Response
Dear Reviewer,
Thank you very much for your thoughtful and detailed feedback on the manuscript. I greatly appreciate the time and effort you have invested in providing your comments and suggestions, all of which have helped to improve the quality and clarity of the work. In this response, I have addressed each of your questions, comments, and suggestions in detail and in the order provided. I hope my revisions and clarifications meet your expectations.
Q: This paper investigates the associations among cyclists’ self-reported behavior, traffic safety knowledge, and crash involvement in Latvia. In 2023 and 2024, 300 cyclists were surveyed online using a cycling behavior questionnaire and a cyclist risk perception and regulation scale. In the current form, the data and analysis do not respond well to the objective. Overall, several major issues should be carefully considered in the revision.
A: Thank you very much for your thorough analysis and thoughtful feedback on our manuscript. After carefully considering your comments, we have recognized certain inaccuracies, particularly regarding the Baltic context, and we have made several revisions to the paper to address these concerns.
Q: Any new findings, including Baltic contexts, should be clarified. This should be supported by reviewing similar studies in the Baltic in the introduction and discussions later in Section 4.
A: I understand that the aim of our study may have appeared misleading. Our original objective was not to specifically analyze cyclists' behavior within the Baltic context. Unfortunately, there is limited research available on the specific behavioral aspects of cycling in the Baltic states. Most existing statistics for the region primarily focus on accidents and fatalities among cyclists, without delving deeply into behavior. Furthermore, the methodology we employed has not yet been applied in either Lithuania or Estonia, highlighting the uniqueness of our study in Latvia. After reflecting on your review, we recognized that certain statements in the manuscript could have been misinterpreted, and we have made the necessary corrections throughout the article to provide greater clarity (highlighted in yellow). While direct comparisons with other Baltic states are not possible due to the lack of similar research, this gap presents a valuable opportunity for future studies to explore these topics more broadly across the region. Additionally, we have clarified these points throughout the text and also improved the conclusion section to ensure it aligns more closely with the study's objectives and the conducted analyses (see lines 688-697). To improve this, we have also added a hypothesis study (see line 151-145).
Q: The analysis is basic, with limited results. There were only a few significant bivariate relationships, and the logistic model had only two significant variables, suggesting the method was not suitable. It is strongly recommended to look at individual behaviors as the combined constructs do not seem to provide sufficient evidence or new findings.
A: We agree with your observation that the initial analysis, particularly the logistic model, was limited in terms of significant results and did not fully address the research objectives. After carefully reviewing this, we decided to revise our analytical approach to better align with the study's goals. We have now included a G*Power calculation for the MANCOVA analysis, which indicated that a minimum of 134 cyclists per group was required. As we exceeded this sample size, we deemed the data adequate for this analysis (see lines 399-402). However, recognizing the insufficient sample size for the logistic regression analysis and the lack of significant findings, we made the decision to exclude this method from the manuscript. Instead, we have employed a pathway analysis, which we believe provides a more appropriate and nuanced exploration of the data in this context. This method allows for a more detailed understanding of the relationships between individual behaviors and crash involvement, addressing the limitations of the initial approach.
Furthermore, to resolve the basic statistical analysis, we have also added the methods described under the Mathematical Statistics section that answer the hypotheses we defined, and have provided a description of the suggested SEM model (see lines 251-264). The analysis itself, including the path analysis, which replaced the regression analysis, can be seen in lines 440-471. Upon reflection, we concluded that regression analysis was not suitable for this study, and we are confident that the current approach offers more meaningful insights and better supports the research objectives. Thank you for the suggestions!
Q: Figure 1 is inappropriate. There is no such model in this paper. It appears to include results of regression and correlation test, which is confusing and misleading. A SEM model, for example, could be applied if this is the aim of the research.
A: Thank you for your constructive feedback regarding Figure 1. We understand your concerns about the appropriateness of the figure and its potential to confuse readers. In response, we have made significant improvements to Figure 1 to ensure it aligns better with the research aim (see lines 431-439). We recognize that the original title may have been misleading, and we have clarified its purpose as a visual representation of the gathered data results. The figure has been re-drawn and renamed appropriately to eliminate any confusion.
Furthermore, as mentioned previously, we have replaced the regression analysis with a pathway analysis based on your suggestions. This adjustment allows the results to better match the study’s objectives and provides clearer findings regarding the relationships among the variables of interest. Additionally, as mentioned before, we have incorporated a Structural Equation Modeling (SEM) approach in the article, including path analysis (see lines 440-471). Based on this analysis, we have added result descriptions at several key points within the framework of the research (see lines 543-550; 560-564; 608-631; 698-701) to provide a more thorough and insightful interpretation of the findings. We greatly appreciate your valuable input and for pointing this out.
Q: In the conclusion, comparisons with other Baltic countries are stated without any details or discussion in previous sections.7
A: We recognize that including these comparisons without sufficient context or discussion in the preceding sections was unnecessary and outside the framework of our research. We have removed these comparisons from the conclusion, as we agree that it is inappropriate to draw parallels with other Baltic states, given the different methodologies employed in their research that do not correspond with our own study. This change has been made to enhance the clarity and focus of our conclusions (see lines 602-628).
Additionally, for the SEM model in the pathway analysis, we have also added a section to the discussion (see lines 543-551; 560-564; 608-631) to further elaborate on the results and their implications within the research framework.
Q: The writing generally needs to be improved with formal language, supported by references in many places. For example, statistics about cycling and cyclist crashes in Latvia were missing.
A: Thank you for your valuable feedback regarding the overall writing quality and the need for formal language and supporting references. We appreciate you pointing out the missing statistics about cycling and cyclist crashes in Latvia. We have now included these statistics to provide a clearer context for our study (see lines 49-55). Additionally, I have made efforts to address your concerns about references and the use of formal language throughout the text. Many of these improvements have been highlighted in yellow for your convenience.
Thank you once again for your insightful suggestions, which have significantly enhanced the clarity and rigor of our manuscript.
All the mentioned additions have been highlighted in yellow in the text. Thank you so much for your valuable contribution to enhancing the quality of the publication!
Round 2
Reviewer 1 Report
Comments and Suggestions for AuthorsThis study assesses the relationships among self-repotted cyclists’ behavior, traffic safety literacy, and their cycling crash involvement rates. A total of 299 cyclists aged M=32.8 from across Latvia (42% females) participated in an online survey, which comprised questions regarding respondents’ demographics, frequency of riding, and the number of crashes in the past five years. The Cycling Behavior Questionnaire (CBQ) and the Cyclist Risk Perception and Regulation Scale (RPRS) were applied to assess cyclists’ behavior patterns and traffic safety literacy. Results: It is noticed that cyclists frequently engage in riding errors and traffic violations while cycling. Those who exhibit more antisocial behavior patterns are also more likely to be involved in road crashes. Conversely, cyclists with greater positive behavior rates more often also tend to possess better knowledge of traffic rules and exhibit a heightened risk perception, indicating a greater awareness of road traffic safety.
The authors addressed all my comments.
Reviewer 2 Report
Comments and Suggestions for AuthorsI thank the Authors for considering the suggestions and have no further comments. I accept the manuscript for publication in its current form.
Best regards,
Reviewer
Reviewer 3 Report
Comments and Suggestions for AuthorsDear authors,
thank you for updating your manuscript.
Reviewer 4 Report
Comments and Suggestions for AuthorsThe authors have addressed most of the concerns that were raised initially, and effective deflected some of the other concerns. They also have beefed up the paper to make some additional points that were not in the original manuscript. At the end of the day, however, I think the paper is now suitable for publication in Safety.
Comments on the Quality of English LanguageNoted above
Reviewer 5 Report
Comments and Suggestions for AuthorsThe authors have addressed the comments adequately
