- Email: [email protected]
Red-light running rates at five intersections by road user in Changsha, China: An observational study
G Model AAP 3844 No. of Pages 6
Accident Analysis and Prevention xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Accident Analysis and Prevention journal homepage: www.elsevier.com/locate/aap
Red-light running rates at five intersections by road user in Changsha, China: An observational study Fangfang Yana , Beixi Lia , Wei Zhangb , Guoqing Hua,* a b
Department of Epidemiology and Health Statistics, School of Public Health, Central South University, Changsha, China Longhua District Centers for Disease Control and Prevention, Shenzhen Municipality, Shenzhen, China
A R T I C L E I N F O
A B S T R A C T
Article history: Received 19 December 2014 Received in revised form 9 April 2015 Accepted 14 June 2015 Available online xxx
The red-light running rate by type of road users has not been reported in China so far. We conducted an observation study to report the violation rate in Changsha, China. Portable digital devices were used to record red-light running violations at five selected intersections. The observation was performed for three days (weekday, weekend and holiday), four time periods per day and an hour per time period (peak and off-peak hours in the morning and in the afternoon). Violation rate was calculated as number of violations divided by total number of vehicles/pedestrians 100%. We used adjusted violation rate ratio (VRR) to quantify the effects of type of day and time period based on Poisson model. Totally, 162,124 vehicles (including motor vehicles, motorcycles and bicycles) and 31,649 pedestrians were recorded. The red-light running rate was 0.14% for motor vehicle drivers, far lowering than those for motorcyclists (18.64%), bicyclists (18.74%) and pedestrians (18.54%). The rate on holiday was 1.89 times that on weekday for drivers. The rate for motorcyclists was high in off-peak hours (adjusted VRR: 1.11), but low on weekend and on holiday (adjusted VRRs: 0.80 and 0.65). The rate for bicyclists was 32% lower on weekend than on weekday. For pedestrians, the rates were high on weekend and holiday and in off-peak hours (adjusted VRR: 1.09, 1.67 and 1.30). The red-light running rate of motor vehicle drivers is far lower than those for motorcyclists, bicyclists and pedestrians. The effects of type of day and time period on violation rate vary with road users, indicating the type of day and time period should be considered when developing and implementing interventions to reduce red-light running of different road users. ã 2015 Elsevier Ltd. All rights reserved.
Keywords: Red-light running violation rate Motor vehicle drivers Motorcyclists Bicyclists Pedestrians China
1. Introduction Road traffic injuries are the eighth leading cause of deaths globally (Lozano et al., 2012). About 1.30 million people were killed on roads worldwide in 2010 (Lozano et al., 2012). According to the Global status report on road safety 2013: supporting a decade of action (World Health Organization, 2013), 275,983 persons in China were estimated to die of road traffic crashes in 2010, accounting for 22% of global road traffic deaths. Traffic violations are important risk factors of road traffic deaths. Based on the statistics released by the Traffic Management Bureau of the Ministry of Public Security of China, 96% of road traffic deaths were caused by traffic violations from motor vehicles (91%), non-motor vehicles (3%), and pedestrians and passengers (2%) in 2013 (Note: the traffic violations are defined according to the Traffic Safety Law of China) (The Traffic
* Corresponding author. E-mail addresses: [email protected] (F. Yan), [email protected] (B. Li) , [email protected] (W. Zhang), [email protected] (G. Hu).
Management Bureau of the Ministry of Public Security of China, 2014). Red-light running is a common traffic violation. Many published studies related to red light running focused on drivers. About 260,000 red light running crashes are estimated to occur annually in the United States, of which approximately 750 lead to deaths (Retting et al., 1999). A national telephone survey revealed that approximately 20% of drivers reported having one or more red light running when entering the last ten signalized intersections in the United States (Porter and Berry, 2001). A study in Greater Manchester, United Kingdom showed that 11.3% of drivers run through red lights at urban shuttle-lane road works (Yousif et al., 2014). 153 of 1190 drivers (12.9%) were observed running red lights at 15 rural and suburban signalized intersections in Jordan (Al-Omari and Al-Masaeid, 2003). In addition, researchers reported various rates of running red lights for other road users: 4.8% for electric bicyclists in Suzhou, China (Du et al., 2013), 6.9% for urban commuter cyclists in Melbourne (Johnson et al., 2011), 37.3% for cyclists in Australia (Johnson et al., 2013), and 56% for two-wheelers riders at urban intersections in Beijing, China (Wu et al., 2012).
http://dx.doi.org/10.1016/j.aap.2015.06.006 0001-4575/ ã 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: F. Yan, et al., Red-light running rates at five intersections by road user in Changsha, China: An observational study, Accid. Anal. Prev. (2015), http://dx.doi.org/10.1016/j.aap.2015.06.006
G Model AAP 3844 No. of Pages 6
2
F. Yan et al. / Accident Analysis and Prevention xxx (2015) xxx–xxx
To comprehensively describe the red-light running rate of different road users is critical for developing appropriate interventions to reduce red light running. However, only a study by Kim et al. (2008) compared the odds of running red lights between drivers and pedestrians in Hawaii and concludes that drivers tend to commit proportionately more red-light running violations than pedestrians in Hawaii, United States. Due to distinctions in culture and road traffic management, the violations of running red lights may differ between China and the United States. We conducted an observational study to estimate the redlight running rates of different road users and to examine differences in rates from type of day and time period by road user in Changsha, China.
2.3. Outcome measures According to the Traffic Safety Law of the People’s Republic of China (Road traffic safety law of China, 2011), the motor vehicles and non-motor vehicles and pedestrians should stop driving or walking in front of red traffic lights. Thus, we defined red light violation as crossing the intersection against the red light. The cameras were placed at the site where the information of crosswalk traffic lights, vehicle traffic lights and pedestrians, bicyclists, motorcyclists, and motor vehicles for the crosswalk being observed can be recorded simultaneously (Fig. 1). The violation rate of red-light running was calculated as the numbers of vehicles or pedestrians being observed running red light divided by total number of vehicles or pedestrians 100%.
2. Material and methods 2.4. Statistical analysis 2.1. Design An observational study was designed to record the red-light running violations of different road users at intersections. Changsha is located in the middle of China and is the capital city of Hunan Province. In 2012, Changsha has a residential population of 7.04 million. The amount of motor vehicles has reached 8.73 million in 2013. According to the statistics of Changsha Traffic Management Bureau (Ren et al., 2012), we selected five intersections that have the highest traffic crash rates, including Yao Ling, Shi Zi Ling, Wen Yi Lu Kou, Yu Hua Ting, and Song Gui Yuan. Of five intersections, three have an underpass in one direction, the other two have no underpass. For the three intersections with an underpass, we chose the crosswalks in the other direction as observation sites. For the other two intersections, we randomly chose one from four crosswalks in two directions as observational sites.
Based on preliminary analysis (not shown here), we divided road user into four categories (motor vehicle driver, motorcyclist, bicyclist, and pedestrian) and report violation rates for all four categories. We reported violation rates by type of day and time period. Poisson regression was used to examine the significance of type of day and time period. Adjusted violation rate ratio (VRR) and 95% confidence interval (CI) were used to quantify the effects of type of day and time period. ‘p < 0.05’ was considered statistically significant. We used Stata/IC 12.1 to analyze data. 2.5. Ethnic considerations The research plan was approved by the Ethnic Committee of School of Public Health, Central South University. The videos that were recorded at five intersections were strictly managed and were used for only counting the number of violations and vehicles/ pedestrians.
2.2. Data collection 3. Results Considering that the road traffic crash rates were reported to vary with time (peak hours vs. off-peak hours) and the type of day (weekdays, weekends, and holidays) (Zhao et al., 2009), we conducted the observations at each intersection for all three types of days. The selections of weekday, weekend and holiday were determined at random. For each selected day, we conducted the observations in four time periods, including two peak hours (7:30–8:30 am and 5:30–6:30 pm) and two off-peak hours (9:30–10:30 am and 3:30–4:30 pm). In total, the traffic flows of 60 h were recorded at the five intersections. Portable digital cameras and smart phones with high-definition camera were used to record the traffic flow at given dates and time periods. The basic information of five intersections were also collected, including length of red light and green light for pedestrian to cross the road, width of roads and crosswalks, whether the intersection has a safety strip, and type of warning light (flashing, beeping or both). The traffic signal phases at intersections kept constant when being observed because Changsha, as well as many other cities in China, does not adopt automatic traffic signal system that can adjust signal phase based on traffic volume and time of day. The field observations were performed by researchers (Yang F., Li B., Zhang W.) and 15 volunteers. The volunteers were recruited from undergraduates of the School of Public Health, Central South University. All volunteers received the training on the requirements of field observations and the collection of road characteristics, in addition to issues to prevent being injured in the collection of data and privacy protection of individuals being recorded. The data were collected between September, 2012 and April, 2013.
3.1. Characteristics of five selected intersections The width of five roads being observed ranged from 29 to 40 m, and the width of five crosswalks ranged from 4.9 to 5.4 m (Table 1). Three crosswalks had safety strips. The time lengths of red lights for crosswalks were 1.6–4.8 times those of green lights (102–140 s
Fig. 1. The selection of observation site at intersections. Note: The red-light running of different road users within the red line can be clearly recorded.
Please cite this article in press as: F. Yan, et al., Red-light running rates at five intersections by road user in Changsha, China: An observational study, Accid. Anal. Prev. (2015), http://dx.doi.org/10.1016/j.aap.2015.06.006
G Model AAP 3844 No. of Pages 6
F. Yan et al. / Accident Analysis and Prevention xxx (2015) xxx–xxx
3
Table 1 Basic characteristics of five selected intersections in Changsha, China. Characteristics
Yao Ling
Shi Zi Ling
Wen Yi Lu Kou
Yu Hua Ting
Song Gui Yuan
Length of red light (s) Length of green light (s) Width of road (m) Width of crosswalk (m) Having safety strips or not Type of warning lighta
120 39 30 5.4 No Flashing
125 77 40 5.1 No Beeping
102 56 30 4.9 Yes Beeping
120 25 24 5.1 Yes Beeping
140 58 29 5.1 Yes Flashing and beeping
a Warning light means the light to remind pedestrians the change of traffic lights from green to red. We only reported the type of warning light of crosswalks being observed.
vs. 25–77 s). The type of warning signal for the four crosswalks was ‘beeping’, and the other one was ‘flashing and beeping’. In total, 162,124 vehicles and 31,649 pedestrians were recorded, including 117,557 cars, 11,946 coaches, 333 trucks, 27,974 motorcycles and 4314 bicycles. In general, cars, pedestrians, and motorcycles were most observed accounting for 60.7%, 16.3 and 14.4%, respectively. The composition of vehicle/pedestrian changed a little between five intersections (Fig. 2). 3.2. Red-light running rate The overall violation rates were much higher for motorcyclists, bicyclists and pedestrians than for motor vehicle drivers (18.54–18.74 vs. 0.14 per 100 vehicles/pedestrians) (Table 2). The effects of type of day and time period on violation rates (violation rate ratio) did not change significantly between univariate and multivariate Poisson models (Tables 2 and 3). The violation rate for motor vehicle drivers on holiday were 1.89 times that on weekday (95% CI: 1.33–2.70) (Table 3). The violation rate of red-light running for motorcyclists was higher in off-peak hours than in peak hours (adjusted VRR: 1.11; 95% CI: 1.06–1.18), but lower on weekend and on holiday than on weekday (adjusted VRRs: 0.80, 95% CI: 0.75–0.85; 0.65, 95% CI: 0.61–0.69) (Table 3). The violation rate was 32% lower on weekend than on weekday (adjusted VRR: 0.68; 95% CI: 0.57–0.81) for bicyclists (Table 3). For pedestrians, the violation rates were higher on weekend and holiday and in off-peak hours than those on weekday and in
peak hours, having adjusted VRR of 1.09, 1.67 and 1.30, respectively (Table 3). 4. Discussion Using the observational method, we reported the violation rates of red-light running by road user in Changsha, China for the first time. Motorcyclists, bicyclists, and pedestrians were observed having much higher violation rates than motor vehicle drivers. The effects of type of date and time period on violation rates were inconsistent: (1) the rates on weekend and holiday were higher than on weekday for motor vehicle drivers and pedestrians, but lower than on weekday for motorcyclists and bicyclists; and (2) motorcyclists and pedestrians had higher violation rates in offpeak hours than in peak hours. Compared with few relevant publications related to China (The Traffic Management Bureau of the Ministry of Public Security of China, 2014; Du et al., 2013; Wu et al., 2012; Bai et al., 2014), our study provides the violation rate of red-light running by road user. The annual road traffic crash statistics that are released by the Transportation Management Bureau under the Ministry of Public Security of China only cover total number of law violations (The Traffic Management Bureau of the Ministry of Public Security of China, 2014). The other three studies focused on the violations of electronic two-wheelers (Du et al., 2013; Wu et al., 2012; Bai et al., 2014). To some extent, our findings reflect the red-light running rates of many cities of China in terms of high similarities in road
Fig. 2. Distribution of road users at five intersections.
Please cite this article in press as: F. Yan, et al., Red-light running rates at five intersections by road user in Changsha, China: An observational study, Accid. Anal. Prev. (2015), http://dx.doi.org/10.1016/j.aap.2015.06.006
G Model AAP 3844 No. of Pages 6
4
F. Yan et al. / Accident Analysis and Prevention xxx (2015) xxx–xxx
Table 2 Red-light running violation rate/100 vehicles or pedestrians by vehicle/pedestrian, type of date and time period. Unadjusted VRRa
Road user
Date/time period
N
Number of violations
Violation rate (95% CI) (%)
Motor vehicle driver
Total Type of date Weekday Weekends Holiday Time Peak hours Off-peak hours
129,836
184
0.14 (0.12, 0.16)
49,204 43,511 37,121
52 58 74
0.11 (0.08, 0.14) 0.13 (0.10, 0.17) 0.20 (0.16, 0.25)
1.00 1.26 (0.87, 1.83) 1.89 (1.32, 2.69)**
62,748 67,088
91 93
0.14 (0.11, 0.18) 0.14 (0.11, 0.17)
1.00 0.96 (0.72, 1.28)
Total Type of day Weekday Weekends Holiday Time Peak hours Off-peak hours
27,974
5215
18.64 (18.21, 19.10)
10,493 8367 9114
2370 1510 1335
22.59 (21.82, 23.41) 18.05 (17.24, 18.90) 14.65 (13.80, 16.32)
1.00 0.80 (0.75, 0.85)** 0.65 (0.61, 0.69)**
13,820 14,154
2436 2779
17.63 (17.02, 18.31) 19.63 (19.01, 20.22)
1.00 1.11 (1.05, 1.18)**
Total Type of day Weekday Weekends Holiday Time Peak hours Off-peak hours
4343
814
18.74 (17.60, 19.91)
1646 1493 1204
343 212 259
20.84 (18.92, 22.95) 14.20 (12.51, 16.12) 21.51 (19.22, 23.90)
1.00 0.68 (0.57, 0.81)** 1.03 (0.88, 1.21)
2283 2060
427 387
18.70 (17.10, 20.41) 18.79 (17.12, 20.51)
1.00 1.00 (0.88, 1.15)
Total Type of day Weekday Weekends Holiday Time Peak hours Off-peak hours
31,649
5868
18.54 (18.10, 19.02)
12,047 10,062 9540
1806 1641 2421
14.99 (14.41, 15.60) 16.31 (15.61, 17.02) 25.38 (24.53, 26.34)
1.00 1.09 (1.02, 1.16)* 1.69 (1.59, 1.80)**
14,903 16,746
2358 3510
15.82 (15.22, 16.43) 20.96 (20.02, 21.63)
1.00 1.32 (1.26, 1.40)**
Motorcyclist
Bicyclist
Pedestrian
a
Unadjusted violation rate ratio. 95% CI: 95% confidence interval. p < 0.05. ** p < 0.01. *
Table 3 Adjusted red-light running violation rate ratio of type of date and time period by vehicle/pedestrian based on Poisson regression. Vehicle/pedestrian
Date/time period
Motor vehicle driver
Type of date (reference = weekday) Weekend Holiday Time (reference = peak hours) Off-peak hours
Motorcyclist
Bicyclist
Pedestrian
Type of date (reference = weekday) Weekend Holiday Time (reference = peak hours) Off-peak hours Type of date (reference = weekday) Weekend Holiday Time (reference = peak hours) Off-peak hours Type of date (reference = weekday) Weekend Holiday Time (reference = peak hours) Off-peak hours
Adjusted VRR (95% CI)a
z
p
1.27 (0.87, 1.84) 1.89 (1.33, 2.70)
1.23 3.52
0.218 <0.001**
0.94 (0.71, 1.26)
0.40
0.691
0.80 (0.75, 0.85) 0.65 (0.61, 0.69)
6.81 12.66
<0.001** <0.001**
1.11 (1.06, 1.18)
3.92
<0.001**
0.68 (0.57, 0.81) 1.03 (0.88, 1.21)
4.40 0.39
<0.001** 0.693
1.02 (0.89, 1.18)
0.35
0.727
1.09 (1.02, 1.17) 1.67 (1.57, 1.78)
2.54 16.53
0.011* <0.001**
1.30 (1.23, 1.37)
9.73
<0.001**
a
Adjusted violation rate ratio. 95% CI: 95% confidence interval. p < 0.05. ** p < 0.01. *
Please cite this article in press as: F. Yan, et al., Red-light running rates at five intersections by road user in Changsha, China: An observational study, Accid. Anal. Prev. (2015), http://dx.doi.org/10.1016/j.aap.2015.06.006
G Model AAP 3844 No. of Pages 6
F. Yan et al. / Accident Analysis and Prevention xxx (2015) xxx–xxx
traffic management between cities, although our findings are based on a 60-h observation of five intersections in Changsha. The over 130-fold gap in violation rates between motor vehicle drivers and all other road users is contrary to the findings by Kim et al. (2008) in Hawaii, United States. The inconsistent results between the two studies may be due to differences in motorization, safety culture and road traffic management between China and the United States. Further, the over 130-fold gap may mainly reflect differences in perceived opportunities and risks of violations at specific road environments for different road users, the penalties to different violations that the road traffic safety law of China requires and differences in the implementation of penalties. China has adopted a new road traffic safety law since May 1, 2011, in which unprecedentedly severe penalties are given to law violations of motor vehicle drivers (excluding motorcyclists), e.g., ‘deprivation of driving license forever’ and ‘being sentenced to jail’ for drunken driving (Road traffic safety law of China, 2011). At the same time, the penalties to other traffic violations for motor vehicle drivers such as red light running and speeding are also raised to some extent although the changes (increased amount of fine and penalty scores) are incomparable to those for drunken driving. These enhanced penalties may also reduce red-light running of drivers. In contrast, equally severe punishments are not required by the national law to be enacted for the violations of riders of electric motorbike, bicyclists, and pedestrians, probably because the penalties are easier to be implemented for drivers than for riders of electric motorbike, bicyclists and pedestrians in practice. In China, motor vehicles drivers are required to register and update their driving license regularly while the mandatory requirements do not apply to riders of electric motorbike that do not use gasoline bicyclists and pedestrians. The rate of red-light running was extremely low for motor vehicle drivers in Changsha (0.14%) compared with the rates in the United States (approximate 20%) (Porter and Berry, 2001), Greater Manchester, United Kingdom (11.3%) (Yousif et al., 2014), and Jordan (12.9%) (Al-Omari and Al-Masaeid, 2003). On the one hand, the variation may reflect differences in data collection (e.g., selfreporting (Porter and Berry, 2001) vs. observation (Yousif et al., 2014; Al-Omari and Al-Masaeid, 2003), choices of observation sites, day, and time period). On the other hand, it is possibly the effects of implementation of unprecedentedly severe road traffic safety law in China since 2011. Sze et al. (2011) reported that the frequency of red-light violations distinctly dropped for drivers after the implementation of the new penalty system in Hong Kong (Note: the new penalty system here is not the same to the enhanced penalties that the traffic law of the mainland of China requires since 2011). High violation rates on holiday for motor vehicles are probably the effects of much high traffic volume of holiday. People prefer to go outside on holiday for traveling, shopping and entertainment, leading to extremely high traffic volume (Lin and Ye, 2013). Porter et al. (2013) reported that the rates of red-light running were not associated with driver characteristics but were significantly associated with using camera or not and traffic volume (and their interaction). Countdown timers (Long et al., 2011) and yellow timing changes (Retting et al., 2008) that were reported to substantially reduce red-light violations of drivers can be applied to reducing violations in China. The violation rate of bicyclists in Changsha (18.74%) differs from the relevant rates in other places (4.8%-56%) (Johnson et al., 2011, 2013; Wu et al., 2012; Kim et al., 2008). This may be due to the differences in data collection (definition, selection of site, date and time, data recording), traffic volume, motorization, road traffic management and safety culture.
5
The relationships between violation rates and type of day, time period vary with the type of road user in this study. The inconsistent results highlight the value of further identifying protective and risk factors of red-light running at context and individual levels in the future. In general, Ajzen’s theory of planned behavior (TPB) is helpful for exploring modifiable factors of law violations, which interprets the formation of behaviors from three elements: attitude, subjective norm, and perceived behavioral control (Ajzen, 1991). The TPB theory suggests that people are more likely to violate traffic rules on roads when they judge the behavior is to be safe at certain situations based on prior experience (Evans and Norman, 2003). These findings suggest that specific interventions should be developed for the violations of certain road users, such as safety education and mandatory requirements for riders of motor bikes, bicyclists and pedestrians as soon as possible. In the future, the TPB could be used to guide the exploration of protective and risk factors of red-light running and the development of specific interventions. Our study has two primary limitations. First, we did not collect the individual information of motor vehicle drivers, motorcyclists, bicyclists, and pedestrians, such as sex, age, level education that may be associated with violation variations across road users. The lack of individual information greatly limits us to interpret differences in violation rates of specific road users from type of day and time period. Second, we did not observe red-light running violations in rural areas of China. A recent study (Huang et al., 2013) reported that road traffic mortality in rural areas were much higher than in urban areas between 2005 and 2010 in China. Considering the lack of necessary intersection camera and weak enforcement of law (Hu et al., 2010), the violation rates in rural areas may be higher than in urban areas. In conclusion, the rates of red-light running are much higher for motorcyclists, bicyclists, and pedestrians than for motor vehicle drivers in Changsha, China. This may be mainly due to the differences in the requirements of punishments to violations by the national road traffic safety law and the implementation of law between different road users. The effects of type of day and time period on red-light running rates vary with the type of road user. Further studies are needed to identify the modifiable factors of high violation rates of red-light running in motorcyclist, bicyclists, and pedestrians. Acknowledgements This study was supported in part by Training Programs of Innovation and Entrepreneurship for Undergraduates in Central South University. The authors owe thanks Changsha Urban Traffic Management Department for helping access to official statistics. The authors are also grateful to Qianluan Yue, Mingjie Yi and all other volunteers who participated in data collection. References Al-Omari, B.H., Al-Masaeid, H.R., 2003. Red light violations at rural and suburban signalized intersections in Jordan. Traffic Inj. Prev. 4 (2), 169–172. Ajzen, I., 1991. The theory of planned behavior. Organ. Behav. Hum. Decis. Process. 50 (2), 179–211. Bai, L., Liu, P., Guo, Y., Yu, H., 2014. Comparative analysis of risky behaviors of electric bicycles at signalized intersections. Traffic Inj. Prev. 18, 0 (Epub ahead of print). Du, W., Yang, J., Powis, B., Zheng, X., Ozanne-Smith, J., Bilston, L., Wu, M., 2013. Understanding on-road practices of electric bike riders: an observational study in a developed city of China. Accid. Anal. Prev. 59, 319–326. doi:http://dx.doi. org/10.1016/j.aap.2013.06.011 (Epub 2013 June 25). Evans, D., Norman, P., 2003. Predicting adolescent pedestrians’ road-crossing intentions: an application and extension of the Theory of Planned Behaviour. Health Educ. Res. 18 (3), 267–277. Hu, G., Baker, T.D., Baker, S.P., 2010. Urban–rural disparities in injury mortality in China, 2006. J. Rural Health 26, 73–77.
Please cite this article in press as: F. Yan, et al., Red-light running rates at five intersections by road user in Changsha, China: An observational study, Accid. Anal. Prev. (2015), http://dx.doi.org/10.1016/j.aap.2015.06.006
G Model AAP 3844 No. of Pages 6
6
F. Yan et al. / Accident Analysis and Prevention xxx (2015) xxx–xxx
Huang, Y., Tian, D., Gao, L., Li, L., Deng, X., Mamady, K., Hu, G., 2013. Neglected increases in rural road traffic mortality in China: findings based on health data from 2005 to 2010. BMC Public Health 13 (1), 1111. Johnson, M., Newstead, S., Charlton, J., Oxley, J., 2011. Riding through red lights: the rate, characteristics and risk factors of non-compliant urban commuter cyclists. Accid. Anal. Prev. 43 (1), 323–328. Johnson, M., Charlton, J., Oxley, J., Newstead, S., 2013. Why do cyclists infringe at red lights? An investigation of Australian cyclists’ reasons for red light infringement. Accid. Anal. Prev. 50, 840–847. doi:http://dx.doi.org/10.1016/j. aap.2012.07.008 (Epub 2012 July 25). Kim, K., Brunner, I.M., Yamashita, E., 2008. Modeling violation of Hawaii’s crosswalk law. Accid. Anal. Prev 40 (3), 894–904. Lin, Q., Ye, S., 2013. How to solve the traffic jam on holiday? Wenhui newspaper. Available at: http://news.hexun.com/2013-10-03/158484361.html (accessed 20.11.14.). Long, K., Han, L.D., Yang, Q., 2011. Effects of countdown timers on driver behavior after the yellow onset at Chinese intersections. Traffic Inj. Prev. 12 (5), 538–544. Lozano, R., Naghavi, M., Foreman, K., et al., 2012. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 380 (9859), 2095–2128 Erratum in: Lancet.2013 Feb 23; 381 (9867): 628. AlMazroa, Mohammad A. [added]; Memish, Ziad A. [added]. Porter, B.E., Berry, T.D., 2001. A nationwide survey of self-reported red light running: measuring prevalence, predictors, and perceived consequences. Accid. Anal. Prev. 33 (6), 735–741. Porter, B.E., Johnson, K.L., Bland, J.F., 2013. Turning off the cameras: red light running characteristics and rates after photo enforcement legislation expired. Accid. Anal. Prev. 50, 1104–1111.
Ren, W., Li, M., Xiao, Q., 2012. Five intersections were most jammed in Changsha City in February. Xiaoxiang Morning News. Available at: http://hn.rednet.cn/c/2012/ 03/08/2541581.htm (accessed 5.06.12.). Retting, R.A., Ulmer, R.G., Williams, A.F., 1999. Prevalence and characteristics of red light running crashes in the United States. Accid. Anal. Prev. 31 (6), 687–694. Retting, R.A., Ferguson, S.A., Farmer, C.M., 2008. Reducing red light running through longer yellow signal timing and red light camera enforcement: results of a field investigation. Accid. Anal. Prev. 40 (1), 327–333. Sze, N.N., Wong, S.C., Pei, X., Choi, P.W., Lo, Y.K., 2011. Is a combined enforcement and penalty strategy effective in combating red light violations? An aggregate model of violation behavior in Hong Kong. Accid. Anal. Prev. 43 (1), 265–271. The Standing Committee of the National People’s Congress. Road traffic safety law of China. Released on April 22, 2011. Available at: http://www.law-lib.com/law/ law_view.asp?id=346438 (accessed 29.11.14.). The Traffic Management Bureau of the Ministry of Public Security of China, 2014. Statistical Yearbook of Road Traffic Crashes of China, 2013. Traffic Management Research Institute of the Ministry of Public Security, Nanjing, Jiangsu. World Health Organization, 2013. Global Status Report on Road Safety 2013: Supporting a Decade of Action. World Health Organization, Geneva. Wu, C., Yao, L., Zhang, K., 2012. The red-light running behavior of electric bike riders and cyclists at urban intersections in China: an observational study. Accid. Anal. Prev. 49, 186–192. Yousif, S., Alterawi, M., Henson, R.R., 2014. Red light running and close following behaviour at urban shuttle-lane roadworks. Accid. Anal. Prev. 66, 147–157. doi: http://dx.doi.org/10.1016/j.aap.2014.01.021 (Epub 2014 February 1). Zhao, X.G., He, X.D., Wu, J.S., Zhao, G.F., Ma, Y.F., Zhang, M., Gan, J.X., Xu, S.W., Jiang, G. Y., 2009. Risk factors for urban road traffic injuries in Hangzhou, China. Arch. Orthop. Trauma Surg. 129 (4), 507–513.
Please cite this article in press as: F. Yan, et al., Red-light running rates at five intersections by road user in Changsha, China: An observational study, Accid. Anal. Prev. (2015), http://dx.doi.org/10.1016/j.aap.2015.06.006
Accident Analysis and Prevention xxx (2015) xxx–xxx
Contents lists available at ScienceDirect
Accident Analysis and Prevention journal homepage: www.elsevier.com/locate/aap
Red-light running rates at five intersections by road user in Changsha, China: An observational study Fangfang Yana , Beixi Lia , Wei Zhangb , Guoqing Hua,* a b
Department of Epidemiology and Health Statistics, School of Public Health, Central South University, Changsha, China Longhua District Centers for Disease Control and Prevention, Shenzhen Municipality, Shenzhen, China
A R T I C L E I N F O
A B S T R A C T
Article history: Received 19 December 2014 Received in revised form 9 April 2015 Accepted 14 June 2015 Available online xxx
The red-light running rate by type of road users has not been reported in China so far. We conducted an observation study to report the violation rate in Changsha, China. Portable digital devices were used to record red-light running violations at five selected intersections. The observation was performed for three days (weekday, weekend and holiday), four time periods per day and an hour per time period (peak and off-peak hours in the morning and in the afternoon). Violation rate was calculated as number of violations divided by total number of vehicles/pedestrians 100%. We used adjusted violation rate ratio (VRR) to quantify the effects of type of day and time period based on Poisson model. Totally, 162,124 vehicles (including motor vehicles, motorcycles and bicycles) and 31,649 pedestrians were recorded. The red-light running rate was 0.14% for motor vehicle drivers, far lowering than those for motorcyclists (18.64%), bicyclists (18.74%) and pedestrians (18.54%). The rate on holiday was 1.89 times that on weekday for drivers. The rate for motorcyclists was high in off-peak hours (adjusted VRR: 1.11), but low on weekend and on holiday (adjusted VRRs: 0.80 and 0.65). The rate for bicyclists was 32% lower on weekend than on weekday. For pedestrians, the rates were high on weekend and holiday and in off-peak hours (adjusted VRR: 1.09, 1.67 and 1.30). The red-light running rate of motor vehicle drivers is far lower than those for motorcyclists, bicyclists and pedestrians. The effects of type of day and time period on violation rate vary with road users, indicating the type of day and time period should be considered when developing and implementing interventions to reduce red-light running of different road users. ã 2015 Elsevier Ltd. All rights reserved.
Keywords: Red-light running violation rate Motor vehicle drivers Motorcyclists Bicyclists Pedestrians China
1. Introduction Road traffic injuries are the eighth leading cause of deaths globally (Lozano et al., 2012). About 1.30 million people were killed on roads worldwide in 2010 (Lozano et al., 2012). According to the Global status report on road safety 2013: supporting a decade of action (World Health Organization, 2013), 275,983 persons in China were estimated to die of road traffic crashes in 2010, accounting for 22% of global road traffic deaths. Traffic violations are important risk factors of road traffic deaths. Based on the statistics released by the Traffic Management Bureau of the Ministry of Public Security of China, 96% of road traffic deaths were caused by traffic violations from motor vehicles (91%), non-motor vehicles (3%), and pedestrians and passengers (2%) in 2013 (Note: the traffic violations are defined according to the Traffic Safety Law of China) (The Traffic
* Corresponding author. E-mail addresses: [email protected] (F. Yan), [email protected] (B. Li) , [email protected] (W. Zhang), [email protected] (G. Hu).
Management Bureau of the Ministry of Public Security of China, 2014). Red-light running is a common traffic violation. Many published studies related to red light running focused on drivers. About 260,000 red light running crashes are estimated to occur annually in the United States, of which approximately 750 lead to deaths (Retting et al., 1999). A national telephone survey revealed that approximately 20% of drivers reported having one or more red light running when entering the last ten signalized intersections in the United States (Porter and Berry, 2001). A study in Greater Manchester, United Kingdom showed that 11.3% of drivers run through red lights at urban shuttle-lane road works (Yousif et al., 2014). 153 of 1190 drivers (12.9%) were observed running red lights at 15 rural and suburban signalized intersections in Jordan (Al-Omari and Al-Masaeid, 2003). In addition, researchers reported various rates of running red lights for other road users: 4.8% for electric bicyclists in Suzhou, China (Du et al., 2013), 6.9% for urban commuter cyclists in Melbourne (Johnson et al., 2011), 37.3% for cyclists in Australia (Johnson et al., 2013), and 56% for two-wheelers riders at urban intersections in Beijing, China (Wu et al., 2012).
http://dx.doi.org/10.1016/j.aap.2015.06.006 0001-4575/ ã 2015 Elsevier Ltd. All rights reserved.
Please cite this article in press as: F. Yan, et al., Red-light running rates at five intersections by road user in Changsha, China: An observational study, Accid. Anal. Prev. (2015), http://dx.doi.org/10.1016/j.aap.2015.06.006
G Model AAP 3844 No. of Pages 6
2
F. Yan et al. / Accident Analysis and Prevention xxx (2015) xxx–xxx
To comprehensively describe the red-light running rate of different road users is critical for developing appropriate interventions to reduce red light running. However, only a study by Kim et al. (2008) compared the odds of running red lights between drivers and pedestrians in Hawaii and concludes that drivers tend to commit proportionately more red-light running violations than pedestrians in Hawaii, United States. Due to distinctions in culture and road traffic management, the violations of running red lights may differ between China and the United States. We conducted an observational study to estimate the redlight running rates of different road users and to examine differences in rates from type of day and time period by road user in Changsha, China.
2.3. Outcome measures According to the Traffic Safety Law of the People’s Republic of China (Road traffic safety law of China, 2011), the motor vehicles and non-motor vehicles and pedestrians should stop driving or walking in front of red traffic lights. Thus, we defined red light violation as crossing the intersection against the red light. The cameras were placed at the site where the information of crosswalk traffic lights, vehicle traffic lights and pedestrians, bicyclists, motorcyclists, and motor vehicles for the crosswalk being observed can be recorded simultaneously (Fig. 1). The violation rate of red-light running was calculated as the numbers of vehicles or pedestrians being observed running red light divided by total number of vehicles or pedestrians 100%.
2. Material and methods 2.4. Statistical analysis 2.1. Design An observational study was designed to record the red-light running violations of different road users at intersections. Changsha is located in the middle of China and is the capital city of Hunan Province. In 2012, Changsha has a residential population of 7.04 million. The amount of motor vehicles has reached 8.73 million in 2013. According to the statistics of Changsha Traffic Management Bureau (Ren et al., 2012), we selected five intersections that have the highest traffic crash rates, including Yao Ling, Shi Zi Ling, Wen Yi Lu Kou, Yu Hua Ting, and Song Gui Yuan. Of five intersections, three have an underpass in one direction, the other two have no underpass. For the three intersections with an underpass, we chose the crosswalks in the other direction as observation sites. For the other two intersections, we randomly chose one from four crosswalks in two directions as observational sites.
Based on preliminary analysis (not shown here), we divided road user into four categories (motor vehicle driver, motorcyclist, bicyclist, and pedestrian) and report violation rates for all four categories. We reported violation rates by type of day and time period. Poisson regression was used to examine the significance of type of day and time period. Adjusted violation rate ratio (VRR) and 95% confidence interval (CI) were used to quantify the effects of type of day and time period. ‘p < 0.05’ was considered statistically significant. We used Stata/IC 12.1 to analyze data. 2.5. Ethnic considerations The research plan was approved by the Ethnic Committee of School of Public Health, Central South University. The videos that were recorded at five intersections were strictly managed and were used for only counting the number of violations and vehicles/ pedestrians.
2.2. Data collection 3. Results Considering that the road traffic crash rates were reported to vary with time (peak hours vs. off-peak hours) and the type of day (weekdays, weekends, and holidays) (Zhao et al., 2009), we conducted the observations at each intersection for all three types of days. The selections of weekday, weekend and holiday were determined at random. For each selected day, we conducted the observations in four time periods, including two peak hours (7:30–8:30 am and 5:30–6:30 pm) and two off-peak hours (9:30–10:30 am and 3:30–4:30 pm). In total, the traffic flows of 60 h were recorded at the five intersections. Portable digital cameras and smart phones with high-definition camera were used to record the traffic flow at given dates and time periods. The basic information of five intersections were also collected, including length of red light and green light for pedestrian to cross the road, width of roads and crosswalks, whether the intersection has a safety strip, and type of warning light (flashing, beeping or both). The traffic signal phases at intersections kept constant when being observed because Changsha, as well as many other cities in China, does not adopt automatic traffic signal system that can adjust signal phase based on traffic volume and time of day. The field observations were performed by researchers (Yang F., Li B., Zhang W.) and 15 volunteers. The volunteers were recruited from undergraduates of the School of Public Health, Central South University. All volunteers received the training on the requirements of field observations and the collection of road characteristics, in addition to issues to prevent being injured in the collection of data and privacy protection of individuals being recorded. The data were collected between September, 2012 and April, 2013.
3.1. Characteristics of five selected intersections The width of five roads being observed ranged from 29 to 40 m, and the width of five crosswalks ranged from 4.9 to 5.4 m (Table 1). Three crosswalks had safety strips. The time lengths of red lights for crosswalks were 1.6–4.8 times those of green lights (102–140 s
Fig. 1. The selection of observation site at intersections. Note: The red-light running of different road users within the red line can be clearly recorded.
Please cite this article in press as: F. Yan, et al., Red-light running rates at five intersections by road user in Changsha, China: An observational study, Accid. Anal. Prev. (2015), http://dx.doi.org/10.1016/j.aap.2015.06.006
G Model AAP 3844 No. of Pages 6
F. Yan et al. / Accident Analysis and Prevention xxx (2015) xxx–xxx
3
Table 1 Basic characteristics of five selected intersections in Changsha, China. Characteristics
Yao Ling
Shi Zi Ling
Wen Yi Lu Kou
Yu Hua Ting
Song Gui Yuan
Length of red light (s) Length of green light (s) Width of road (m) Width of crosswalk (m) Having safety strips or not Type of warning lighta
120 39 30 5.4 No Flashing
125 77 40 5.1 No Beeping
102 56 30 4.9 Yes Beeping
120 25 24 5.1 Yes Beeping
140 58 29 5.1 Yes Flashing and beeping
a Warning light means the light to remind pedestrians the change of traffic lights from green to red. We only reported the type of warning light of crosswalks being observed.
vs. 25–77 s). The type of warning signal for the four crosswalks was ‘beeping’, and the other one was ‘flashing and beeping’. In total, 162,124 vehicles and 31,649 pedestrians were recorded, including 117,557 cars, 11,946 coaches, 333 trucks, 27,974 motorcycles and 4314 bicycles. In general, cars, pedestrians, and motorcycles were most observed accounting for 60.7%, 16.3 and 14.4%, respectively. The composition of vehicle/pedestrian changed a little between five intersections (Fig. 2). 3.2. Red-light running rate The overall violation rates were much higher for motorcyclists, bicyclists and pedestrians than for motor vehicle drivers (18.54–18.74 vs. 0.14 per 100 vehicles/pedestrians) (Table 2). The effects of type of day and time period on violation rates (violation rate ratio) did not change significantly between univariate and multivariate Poisson models (Tables 2 and 3). The violation rate for motor vehicle drivers on holiday were 1.89 times that on weekday (95% CI: 1.33–2.70) (Table 3). The violation rate of red-light running for motorcyclists was higher in off-peak hours than in peak hours (adjusted VRR: 1.11; 95% CI: 1.06–1.18), but lower on weekend and on holiday than on weekday (adjusted VRRs: 0.80, 95% CI: 0.75–0.85; 0.65, 95% CI: 0.61–0.69) (Table 3). The violation rate was 32% lower on weekend than on weekday (adjusted VRR: 0.68; 95% CI: 0.57–0.81) for bicyclists (Table 3). For pedestrians, the violation rates were higher on weekend and holiday and in off-peak hours than those on weekday and in
peak hours, having adjusted VRR of 1.09, 1.67 and 1.30, respectively (Table 3). 4. Discussion Using the observational method, we reported the violation rates of red-light running by road user in Changsha, China for the first time. Motorcyclists, bicyclists, and pedestrians were observed having much higher violation rates than motor vehicle drivers. The effects of type of date and time period on violation rates were inconsistent: (1) the rates on weekend and holiday were higher than on weekday for motor vehicle drivers and pedestrians, but lower than on weekday for motorcyclists and bicyclists; and (2) motorcyclists and pedestrians had higher violation rates in offpeak hours than in peak hours. Compared with few relevant publications related to China (The Traffic Management Bureau of the Ministry of Public Security of China, 2014; Du et al., 2013; Wu et al., 2012; Bai et al., 2014), our study provides the violation rate of red-light running by road user. The annual road traffic crash statistics that are released by the Transportation Management Bureau under the Ministry of Public Security of China only cover total number of law violations (The Traffic Management Bureau of the Ministry of Public Security of China, 2014). The other three studies focused on the violations of electronic two-wheelers (Du et al., 2013; Wu et al., 2012; Bai et al., 2014). To some extent, our findings reflect the red-light running rates of many cities of China in terms of high similarities in road
Fig. 2. Distribution of road users at five intersections.
Please cite this article in press as: F. Yan, et al., Red-light running rates at five intersections by road user in Changsha, China: An observational study, Accid. Anal. Prev. (2015), http://dx.doi.org/10.1016/j.aap.2015.06.006
G Model AAP 3844 No. of Pages 6
4
F. Yan et al. / Accident Analysis and Prevention xxx (2015) xxx–xxx
Table 2 Red-light running violation rate/100 vehicles or pedestrians by vehicle/pedestrian, type of date and time period. Unadjusted VRRa
Road user
Date/time period
N
Number of violations
Violation rate (95% CI) (%)
Motor vehicle driver
Total Type of date Weekday Weekends Holiday Time Peak hours Off-peak hours
129,836
184
0.14 (0.12, 0.16)
49,204 43,511 37,121
52 58 74
0.11 (0.08, 0.14) 0.13 (0.10, 0.17) 0.20 (0.16, 0.25)
1.00 1.26 (0.87, 1.83) 1.89 (1.32, 2.69)**
62,748 67,088
91 93
0.14 (0.11, 0.18) 0.14 (0.11, 0.17)
1.00 0.96 (0.72, 1.28)
Total Type of day Weekday Weekends Holiday Time Peak hours Off-peak hours
27,974
5215
18.64 (18.21, 19.10)
10,493 8367 9114
2370 1510 1335
22.59 (21.82, 23.41) 18.05 (17.24, 18.90) 14.65 (13.80, 16.32)
1.00 0.80 (0.75, 0.85)** 0.65 (0.61, 0.69)**
13,820 14,154
2436 2779
17.63 (17.02, 18.31) 19.63 (19.01, 20.22)
1.00 1.11 (1.05, 1.18)**
Total Type of day Weekday Weekends Holiday Time Peak hours Off-peak hours
4343
814
18.74 (17.60, 19.91)
1646 1493 1204
343 212 259
20.84 (18.92, 22.95) 14.20 (12.51, 16.12) 21.51 (19.22, 23.90)
1.00 0.68 (0.57, 0.81)** 1.03 (0.88, 1.21)
2283 2060
427 387
18.70 (17.10, 20.41) 18.79 (17.12, 20.51)
1.00 1.00 (0.88, 1.15)
Total Type of day Weekday Weekends Holiday Time Peak hours Off-peak hours
31,649
5868
18.54 (18.10, 19.02)
12,047 10,062 9540
1806 1641 2421
14.99 (14.41, 15.60) 16.31 (15.61, 17.02) 25.38 (24.53, 26.34)
1.00 1.09 (1.02, 1.16)* 1.69 (1.59, 1.80)**
14,903 16,746
2358 3510
15.82 (15.22, 16.43) 20.96 (20.02, 21.63)
1.00 1.32 (1.26, 1.40)**
Motorcyclist
Bicyclist
Pedestrian
a
Unadjusted violation rate ratio. 95% CI: 95% confidence interval. p < 0.05. ** p < 0.01. *
Table 3 Adjusted red-light running violation rate ratio of type of date and time period by vehicle/pedestrian based on Poisson regression. Vehicle/pedestrian
Date/time period
Motor vehicle driver
Type of date (reference = weekday) Weekend Holiday Time (reference = peak hours) Off-peak hours
Motorcyclist
Bicyclist
Pedestrian
Type of date (reference = weekday) Weekend Holiday Time (reference = peak hours) Off-peak hours Type of date (reference = weekday) Weekend Holiday Time (reference = peak hours) Off-peak hours Type of date (reference = weekday) Weekend Holiday Time (reference = peak hours) Off-peak hours
Adjusted VRR (95% CI)a
z
p
1.27 (0.87, 1.84) 1.89 (1.33, 2.70)
1.23 3.52
0.218 <0.001**
0.94 (0.71, 1.26)
0.40
0.691
0.80 (0.75, 0.85) 0.65 (0.61, 0.69)
6.81 12.66
<0.001** <0.001**
1.11 (1.06, 1.18)
3.92
<0.001**
0.68 (0.57, 0.81) 1.03 (0.88, 1.21)
4.40 0.39
<0.001** 0.693
1.02 (0.89, 1.18)
0.35
0.727
1.09 (1.02, 1.17) 1.67 (1.57, 1.78)
2.54 16.53
0.011* <0.001**
1.30 (1.23, 1.37)
9.73
<0.001**
a
Adjusted violation rate ratio. 95% CI: 95% confidence interval. p < 0.05. ** p < 0.01. *
Please cite this article in press as: F. Yan, et al., Red-light running rates at five intersections by road user in Changsha, China: An observational study, Accid. Anal. Prev. (2015), http://dx.doi.org/10.1016/j.aap.2015.06.006
G Model AAP 3844 No. of Pages 6
F. Yan et al. / Accident Analysis and Prevention xxx (2015) xxx–xxx
traffic management between cities, although our findings are based on a 60-h observation of five intersections in Changsha. The over 130-fold gap in violation rates between motor vehicle drivers and all other road users is contrary to the findings by Kim et al. (2008) in Hawaii, United States. The inconsistent results between the two studies may be due to differences in motorization, safety culture and road traffic management between China and the United States. Further, the over 130-fold gap may mainly reflect differences in perceived opportunities and risks of violations at specific road environments for different road users, the penalties to different violations that the road traffic safety law of China requires and differences in the implementation of penalties. China has adopted a new road traffic safety law since May 1, 2011, in which unprecedentedly severe penalties are given to law violations of motor vehicle drivers (excluding motorcyclists), e.g., ‘deprivation of driving license forever’ and ‘being sentenced to jail’ for drunken driving (Road traffic safety law of China, 2011). At the same time, the penalties to other traffic violations for motor vehicle drivers such as red light running and speeding are also raised to some extent although the changes (increased amount of fine and penalty scores) are incomparable to those for drunken driving. These enhanced penalties may also reduce red-light running of drivers. In contrast, equally severe punishments are not required by the national law to be enacted for the violations of riders of electric motorbike, bicyclists, and pedestrians, probably because the penalties are easier to be implemented for drivers than for riders of electric motorbike, bicyclists and pedestrians in practice. In China, motor vehicles drivers are required to register and update their driving license regularly while the mandatory requirements do not apply to riders of electric motorbike that do not use gasoline bicyclists and pedestrians. The rate of red-light running was extremely low for motor vehicle drivers in Changsha (0.14%) compared with the rates in the United States (approximate 20%) (Porter and Berry, 2001), Greater Manchester, United Kingdom (11.3%) (Yousif et al., 2014), and Jordan (12.9%) (Al-Omari and Al-Masaeid, 2003). On the one hand, the variation may reflect differences in data collection (e.g., selfreporting (Porter and Berry, 2001) vs. observation (Yousif et al., 2014; Al-Omari and Al-Masaeid, 2003), choices of observation sites, day, and time period). On the other hand, it is possibly the effects of implementation of unprecedentedly severe road traffic safety law in China since 2011. Sze et al. (2011) reported that the frequency of red-light violations distinctly dropped for drivers after the implementation of the new penalty system in Hong Kong (Note: the new penalty system here is not the same to the enhanced penalties that the traffic law of the mainland of China requires since 2011). High violation rates on holiday for motor vehicles are probably the effects of much high traffic volume of holiday. People prefer to go outside on holiday for traveling, shopping and entertainment, leading to extremely high traffic volume (Lin and Ye, 2013). Porter et al. (2013) reported that the rates of red-light running were not associated with driver characteristics but were significantly associated with using camera or not and traffic volume (and their interaction). Countdown timers (Long et al., 2011) and yellow timing changes (Retting et al., 2008) that were reported to substantially reduce red-light violations of drivers can be applied to reducing violations in China. The violation rate of bicyclists in Changsha (18.74%) differs from the relevant rates in other places (4.8%-56%) (Johnson et al., 2011, 2013; Wu et al., 2012; Kim et al., 2008). This may be due to the differences in data collection (definition, selection of site, date and time, data recording), traffic volume, motorization, road traffic management and safety culture.
5
The relationships between violation rates and type of day, time period vary with the type of road user in this study. The inconsistent results highlight the value of further identifying protective and risk factors of red-light running at context and individual levels in the future. In general, Ajzen’s theory of planned behavior (TPB) is helpful for exploring modifiable factors of law violations, which interprets the formation of behaviors from three elements: attitude, subjective norm, and perceived behavioral control (Ajzen, 1991). The TPB theory suggests that people are more likely to violate traffic rules on roads when they judge the behavior is to be safe at certain situations based on prior experience (Evans and Norman, 2003). These findings suggest that specific interventions should be developed for the violations of certain road users, such as safety education and mandatory requirements for riders of motor bikes, bicyclists and pedestrians as soon as possible. In the future, the TPB could be used to guide the exploration of protective and risk factors of red-light running and the development of specific interventions. Our study has two primary limitations. First, we did not collect the individual information of motor vehicle drivers, motorcyclists, bicyclists, and pedestrians, such as sex, age, level education that may be associated with violation variations across road users. The lack of individual information greatly limits us to interpret differences in violation rates of specific road users from type of day and time period. Second, we did not observe red-light running violations in rural areas of China. A recent study (Huang et al., 2013) reported that road traffic mortality in rural areas were much higher than in urban areas between 2005 and 2010 in China. Considering the lack of necessary intersection camera and weak enforcement of law (Hu et al., 2010), the violation rates in rural areas may be higher than in urban areas. In conclusion, the rates of red-light running are much higher for motorcyclists, bicyclists, and pedestrians than for motor vehicle drivers in Changsha, China. This may be mainly due to the differences in the requirements of punishments to violations by the national road traffic safety law and the implementation of law between different road users. The effects of type of day and time period on red-light running rates vary with the type of road user. Further studies are needed to identify the modifiable factors of high violation rates of red-light running in motorcyclist, bicyclists, and pedestrians. Acknowledgements This study was supported in part by Training Programs of Innovation and Entrepreneurship for Undergraduates in Central South University. The authors owe thanks Changsha Urban Traffic Management Department for helping access to official statistics. The authors are also grateful to Qianluan Yue, Mingjie Yi and all other volunteers who participated in data collection. References Al-Omari, B.H., Al-Masaeid, H.R., 2003. Red light violations at rural and suburban signalized intersections in Jordan. Traffic Inj. Prev. 4 (2), 169–172. Ajzen, I., 1991. The theory of planned behavior. Organ. Behav. Hum. Decis. Process. 50 (2), 179–211. Bai, L., Liu, P., Guo, Y., Yu, H., 2014. Comparative analysis of risky behaviors of electric bicycles at signalized intersections. Traffic Inj. Prev. 18, 0 (Epub ahead of print). Du, W., Yang, J., Powis, B., Zheng, X., Ozanne-Smith, J., Bilston, L., Wu, M., 2013. Understanding on-road practices of electric bike riders: an observational study in a developed city of China. Accid. Anal. Prev. 59, 319–326. doi:http://dx.doi. org/10.1016/j.aap.2013.06.011 (Epub 2013 June 25). Evans, D., Norman, P., 2003. Predicting adolescent pedestrians’ road-crossing intentions: an application and extension of the Theory of Planned Behaviour. Health Educ. Res. 18 (3), 267–277. Hu, G., Baker, T.D., Baker, S.P., 2010. Urban–rural disparities in injury mortality in China, 2006. J. Rural Health 26, 73–77.
Please cite this article in press as: F. Yan, et al., Red-light running rates at five intersections by road user in Changsha, China: An observational study, Accid. Anal. Prev. (2015), http://dx.doi.org/10.1016/j.aap.2015.06.006
G Model AAP 3844 No. of Pages 6
6
F. Yan et al. / Accident Analysis and Prevention xxx (2015) xxx–xxx
Huang, Y., Tian, D., Gao, L., Li, L., Deng, X., Mamady, K., Hu, G., 2013. Neglected increases in rural road traffic mortality in China: findings based on health data from 2005 to 2010. BMC Public Health 13 (1), 1111. Johnson, M., Newstead, S., Charlton, J., Oxley, J., 2011. Riding through red lights: the rate, characteristics and risk factors of non-compliant urban commuter cyclists. Accid. Anal. Prev. 43 (1), 323–328. Johnson, M., Charlton, J., Oxley, J., Newstead, S., 2013. Why do cyclists infringe at red lights? An investigation of Australian cyclists’ reasons for red light infringement. Accid. Anal. Prev. 50, 840–847. doi:http://dx.doi.org/10.1016/j. aap.2012.07.008 (Epub 2012 July 25). Kim, K., Brunner, I.M., Yamashita, E., 2008. Modeling violation of Hawaii’s crosswalk law. Accid. Anal. Prev 40 (3), 894–904. Lin, Q., Ye, S., 2013. How to solve the traffic jam on holiday? Wenhui newspaper. Available at: http://news.hexun.com/2013-10-03/158484361.html (accessed 20.11.14.). Long, K., Han, L.D., Yang, Q., 2011. Effects of countdown timers on driver behavior after the yellow onset at Chinese intersections. Traffic Inj. Prev. 12 (5), 538–544. Lozano, R., Naghavi, M., Foreman, K., et al., 2012. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. Lancet 380 (9859), 2095–2128 Erratum in: Lancet.2013 Feb 23; 381 (9867): 628. AlMazroa, Mohammad A. [added]; Memish, Ziad A. [added]. Porter, B.E., Berry, T.D., 2001. A nationwide survey of self-reported red light running: measuring prevalence, predictors, and perceived consequences. Accid. Anal. Prev. 33 (6), 735–741. Porter, B.E., Johnson, K.L., Bland, J.F., 2013. Turning off the cameras: red light running characteristics and rates after photo enforcement legislation expired. Accid. Anal. Prev. 50, 1104–1111.
Ren, W., Li, M., Xiao, Q., 2012. Five intersections were most jammed in Changsha City in February. Xiaoxiang Morning News. Available at: http://hn.rednet.cn/c/2012/ 03/08/2541581.htm (accessed 5.06.12.). Retting, R.A., Ulmer, R.G., Williams, A.F., 1999. Prevalence and characteristics of red light running crashes in the United States. Accid. Anal. Prev. 31 (6), 687–694. Retting, R.A., Ferguson, S.A., Farmer, C.M., 2008. Reducing red light running through longer yellow signal timing and red light camera enforcement: results of a field investigation. Accid. Anal. Prev. 40 (1), 327–333. Sze, N.N., Wong, S.C., Pei, X., Choi, P.W., Lo, Y.K., 2011. Is a combined enforcement and penalty strategy effective in combating red light violations? An aggregate model of violation behavior in Hong Kong. Accid. Anal. Prev. 43 (1), 265–271. The Standing Committee of the National People’s Congress. Road traffic safety law of China. Released on April 22, 2011. Available at: http://www.law-lib.com/law/ law_view.asp?id=346438 (accessed 29.11.14.). The Traffic Management Bureau of the Ministry of Public Security of China, 2014. Statistical Yearbook of Road Traffic Crashes of China, 2013. Traffic Management Research Institute of the Ministry of Public Security, Nanjing, Jiangsu. World Health Organization, 2013. Global Status Report on Road Safety 2013: Supporting a Decade of Action. World Health Organization, Geneva. Wu, C., Yao, L., Zhang, K., 2012. The red-light running behavior of electric bike riders and cyclists at urban intersections in China: an observational study. Accid. Anal. Prev. 49, 186–192. Yousif, S., Alterawi, M., Henson, R.R., 2014. Red light running and close following behaviour at urban shuttle-lane roadworks. Accid. Anal. Prev. 66, 147–157. doi: http://dx.doi.org/10.1016/j.aap.2014.01.021 (Epub 2014 February 1). Zhao, X.G., He, X.D., Wu, J.S., Zhao, G.F., Ma, Y.F., Zhang, M., Gan, J.X., Xu, S.W., Jiang, G. Y., 2009. Risk factors for urban road traffic injuries in Hangzhou, China. Arch. Orthop. Trauma Surg. 129 (4), 507–513.
Please cite this article in press as: F. Yan, et al., Red-light running rates at five intersections by road user in Changsha, China: An observational study, Accid. Anal. Prev. (2015), http://dx.doi.org/10.1016/j.aap.2015.06.006