Speed Characteristics in Relation to Speed Limit Increase and Its Influence on Driver’s Speed Selection Behavior
Round 1
Reviewer 1 Report
The manuscript titled “Speed characteristics in relation to speed limit increase and its influence on driver’s speed selection behaviour” deals with an interesting topic with implications for traffic safety. Furthermore, the manuscript has several positive issues, such as the review of the literature in the introduction section or the use of objective data. However, I don’t see the novelty of this research and how it contributes to advancing scientific knowledge. In addition to this major concern, I would like to highlight some other minor issues. I will organize my comments following the manuscript structure:
Introduction
The authors have reviewed previous research about the effects of speed limits on speed behaviour well. From this review, it can be concluded that there is wide empirical evidence showing the impact of speed limits on drivers’ speed (see references 4 to 13, both included). However, the authors have not provided a rationale for the relevance of their research. The goal of their research is to “examine the effect of speed limit increase on actual speeds of vehicles” (see page 3, lines 93-94). Even if this could be a relevant topic, it seems as if we already have enough empirical evidence about this relationship. The authors should make a point about what is new about their research, why it is relevant to investigate this topic and how they will contribute to advancing scientific knowledge. Currently, this study does not appear to offer any novel or relevant contribution to advancing previous scientific knowledge. The goal of the study (see page 3, lines 93-95) includes a contextualization “in both rural and urban highways” that is confusing as the distinction between rural and urban highways is not taken into account further in the analyses.Method
I would remove Figure 1. It doesn’t provide substantial information. Table 1 stipulates that there are only 15 ATR. This is surprising because previously the authors had stated that “27 sites were selected as control sites and 39 sites as treated sites” (page 3, lines 110-111). Perhaps I have missed something. The authors should preferably explain in this section (not in the results), the reasons for using the average speed and the 85th percentile speed for your analyses. Too much detail is given about t and K-S tests. The t-test is a very well-known statistical test. The authors should just: (1) report in the results section those data showing that assumptions for these kinds of analyses were fulfilled and, (2) to explain the specific method used taking into account the different sample sizes and the equal or unequal variance. Although the K-S test is less known, I still found the information provided about the analysis too long, too detailed and not very clear (e.g., page 7, lines 199 to 207). Page 6 line 179: what does “also” refer to?Results
The findings in table 3 appear to be cut-off. Page 8, lines 242-244. As the authors have not included all ATRs’ distribution curves, I would say something similar to “As an example...” or “As a way of illustrating...”. The authors should include a table similar to Table 3 showing the t-test results for “average speed”, as this was one of the variables they studied.Discussion
The authors should discuss their findings and not only summarise the findings that were already explained in the results section. A good discussion should relate their findings with previous research, reflect about theoretical contributions and practical implications, discuss the limitations of their research and point out some avenues for future research. I missed some reflection about the implications of this study and their findings for sustainability research. In fact, sustainability was not mentioned at all along the manuscript.I hope the authors find these comments helpful and constructive for their research.
Author Response
1-Currently, this study does not appear to offer any novel or relevant contribution to advancing previous scientific knowledge:
Answer:Thank you for pointing out on that. Since this project was related to speed limit increase, it was important for Kansas Department of Transportation to see how speed limit increase caused changes in drivers’ behaviors. So, according to several similar studies in other states, new results were obtained regarding how this increase caused drivers' behavior change accordingly.
2-The goal of the study (see page 3, lines 93-95) includes a contextualization “in both rural and urban highways” that is confusing as the distinction between rural and urban highways is not taken into account further in the analyses.
Answer:Line 93-95 is revised and the Table#1 is now showing the distinction between rural and urban exact locations, which are also shown on the map too.
3- I would remove Figure 1. It doesn’t provide substantial information. Table 1 stipulates that there are only 15 ATR. This is surprising because previously the authors had stated that “27 sites were selected as control sites and 39 sites as treated sites” (page 3, lines 110-111).
Answer: Thank you for your suggestion. Figure 1 is now representing more information about the rural and urban locations. Also, there are 15 ATRs total available that are located on the sections affected by speed limit increase and on the sections without speed limit increase. We have had total 39 sections affected by speed limit increase in Kansas, while the sections that no speed limit increase has occurred are only 27.
4-The authors should preferably explain in this section (not in the results), the reasons for using the average speed and the 85th percentile speed for your analyses.
Answer:The information is now moved to this section (line 116-118).
5-Too much detail is given about t and K-S tests. The t-test is a very well-known statistical test. The authors should just: (1) report in the results section those data showing that assumptions for these kinds of analyses were fulfilled and, (2) to explain the specific method used taking into account the different sample sizes and the equal or unequal variance.
Answer: The excessive information is now removed and the results of the two methods are only shown.
6- The findings in table 3 appear to be cut-off. Page 8, lines 242-244. As the authors have not included all ATRs’ distribution curves, I would say something similar to “As an example...” or “As a way of illustrating...”. The authors should include a table similar to Table 3 showing the t-test results for “average speed”, as this was one of the variables they studied.
Answer:As an example is added and an additional table is also included with more explanations.
7-The authors should discuss their findings and not only summarize the findings that were already explained in the results section. A good discussion should relate their findings with previous research, reflect about theoretical contributions and practical implications, discuss the limitations of their research and point out some avenues for future research. I missed some reflection about the implications of this study and their findings for sustainability research. In fact, sustainability was not mentioned at all along the manuscript.
Answer:The conclusion section is rewritten and limitation, future work, and sustainable transportation word is included in this section and other parts too.
Reviewer 2 Report
The paper "Speed characteristics in relation to speed limit increase and its influence on driver's speed selection behavior" delivers two new insights in speed characteristics resulting from statistical tests. Whereas the contribution is not extensive, the paper works with clear and understandable language, and, a well-structrured form suitable for the short lenght of the paper. Therefore, the number (and selection) of references is suitable.
One more confusing element is the mentioning of the STATA software with "The t-test procedure of STATA software was also utilized in this research to compute the t-values". I'm not sure if there are any specifics in the procedure. For the sake of providing software-independent methodologies, I would not mention the software in this paper.
Author Response
1-the contribution is not extensive:
Answer:Thank you for pointing out on that. Since this project was related to speed limit increase, it was important for Kansas Department of Transportation to see how speed limit increase caused changes in drivers’ behaviors compared to surrounding states after the speed limit change. So, according to several similar studies in other states, new results were obtained regarding 5mph increase in the speed limit, which caused drivers' behavior change too.
2-For the sake of providing software-independent methodologies, I would not mention the software in this paper
Answer: Since the t-test is conducted using STATA software package specifically and as there are many more software packages too, which are capable for conducting the t-test, I thought mentioning the software name could result in less confusion for readers on how the t-test is applied and using which package the results have released.
Reviewer 3 Report
Authors of the paper present results of studies on influence of speed limit increase on drivers’ speed selection behavior. The subject of the paper is interesting and worth investigating. The results are rather predictable and obvious. The methodology of the research is understandable, but it raises some doubts about the quality of the results and some issues require further clarification.
The speeding behavior of drivers depends on a number of factors such as weather conditions or traffic volume (free flow or saturated or over saturated traffic), but also on road characteristics. In this context:
a) Isn't it reasonable to classify ATRs' and differentiate them according to the characteristics of the road section and its surroundings? This would allow conclusions to be drawn about the speed choice of drivers taking into account the characteristics of the road section and its surroundings.
b) Was the selection of data made taking into account weather conditions and traffic volume (free, saturated or over saturated conditions)? An increase in traffic or weather conditions influence the choice of speed and should be the basis for selecting a set of data for statistical comparison. Are data on road accidents available? It would be interesting to compare the change in the number of accidents on the sections where ATRs are located in order to determine the impact of raising speed limits on road safety levels. The results indicating the choice of higher speeds by drivers as a result of a change in speed limits seem to be obvious under comparable traffic conditions and with comparable specific road characteristics.
Author Response
1-Isn't it reasonable to classify ATRs' and differentiate them according to the characteristics of the road section and its surroundings? This would allow conclusions to be drawn about the speed choice of drivers taking into account the characteristics of the road section and its surroundings.
Answer: Thank you for your suggestion. Figure 1 and Table 1 are now representing more information about the rural and urban locations of the road sections and surroundings. Also, there are 15 ATRs total available that are located on the sections affected by speed limit increase and on the sections without speed limit increase in this study.
2- Was the selection of data made taking into account weather conditions and traffic volume (free, saturated or over saturated conditions)? An increase in traffic or weather conditions influence the choice of speed and should be the basis for selecting a set of data for statistical comparison. Are data on road accidents available? It would be interesting to compare the change in the number of accidents on the sections where ATRs are located in order to determine the impact of raising speed limits on road safety levels. The results indicating the choice of higher speeds by drivers as a result of a change in speed limits seem to be obvious under comparable traffic conditions and with comparable specific road characteristics.
Answer: There has been no data regarding weather condition or traffic volume but there has been general data for total number of crashes and fatal and injury crashes related to sections before speed limit increase and after speed increase. So, Figure 1 has been added to the revised manuscript, which depicts total number of crashes and fatal and injury crashes separately for your information.
Round 2
Reviewer 3 Report
Thank you for your comments. Further research work on the presented issues is necessary, taking into account the impact of speed limits on road safety.
Author Response
Thank you for your constructive comment. Yes, you are totally correct and there is another research, which is under going for further variables considerations that caused safety issue for interstates based on speed limit increase and other parameters. The new results will be submitted to your journal for review in near future.
