Accident Analysis and Prevention 75 (2015) 272–284
Contents lists available at ScienceDirect
Accident Analysis and Prevention journal homepage: www.elsevier.com/locate/aap
Reasons underlying behaviour of motorcyclists disregarding traffic regulations in urban areas of Indonesia Yusak O. Susilo a, * , Tri Basuki Joewono b , Upali Vandebona c a b c
Department of Transport Science, Royal Institute of Technology (KTH), Teknikringen 10, 100 44 Stockholm, Sweden Graduate Program, Parahyangan Catholic University, Jl. Merdeka 30 Bandung 40117, Indonesia School of Civil and Environmental Engineering, University of New South Wales, Kensington, Campus, Sydney 2052, Australia
A R T I C L E I N F O
A B S T R A C T
Article history: Received 7 May 2014 Received in revised form 16 December 2014 Accepted 16 December 2014 Available online xxx
Over the last decade, motorcycle use has been rapidly increasing in Indonesia as have violations of traffic rules committed by motorcyclists. This study aims to explore the impacts of motorcyclists’ attitudes, habits, preferences, and travel patterns on their behaviour in disregarding traffic regulations in three cities in Indonesia. The theory of planned behaviour and structural equation modelling are employed to explore these relationships. Consistent with results from previous studies in developed countries, an individual’s beliefs and attitudes, social norms and perceived behaviour control significantly influence behaviour in disregarding traffic rules. Young adults and students are found to be more likely to frequently violate traffic regulations. However, unlike previous findings from developed countries, in Indonesia, males are less likely to disregard traffic rules than females. Overall, pushing the motorcycle through a (very) narrow gap, speeding, driving recklessly, and overtaking on the wrong side are the most frequent traffic violations that make up repetitive violation behaviour among urban motorcyclists in Indonesia. The results highlight the need to revisit Indonesian National Traffic Law traffic violation classification and penalties and separate violations that are likely to cause fatal results, thus requiring tougher law enforcement, from violations that are unlikely to have fatal consequences. ã 2014 Elsevier Ltd. All rights reserved.
Keywords: Motorcyclist Repetitive traffic violation Indonesia Psychological factors Social norms Perceived control Structural equation modelling
1. Introduction In the last two decades, economic conditions and motorisation in Southeast Asian countries have experienced a rapid growth rate. The motorcycle has become a popular transport mode in this region, including Indonesia, Vietnam and Thailand (Kaltheier, 2002). Currently, Indonesia is the third largest motorcycle market in the world after China and India. In Indonesia, there were 13.6 million motorcycles in 2000, increasing to 32.5 million in 2006 and 76.4 million in 2012. In 2012, the total number of motorised vehicles in Indonesia was 94.4 million (Statistics Indonesia, 2012). It has been estimated that by 2015, there will be one motorcycle for every two persons in this country (AISI, 2009) which has a population of about 240 million. The motorcycle is a popular transport mode among Southeast Asian travellers mainly because of the affordability, flexibility, and manoeuvrability on congested roads typical in the region.
* Corresponding author. Tel.: +46 87909635. E-mail addresses: [email protected] (Y.O. Susilo), [email protected] (T.B. Joewono), [email protected] (U. Vandebona). http://dx.doi.org/10.1016/j.aap.2014.12.016 0001-4575/ ã 2014 Elsevier Ltd. All rights reserved.
At the same time, the growth of motorcycle users has been seen as disruptive to users of other travel modes. Aggressive encroachment of pedestrian space and lack of compliance with road rules have been widely considered as a major cause of road accidents (Indriastuti and Sulistio, 2010) although statistically, motorcycle-related accidents are not over-represented in Indonesian traffic accident statistics. In 2012, motorcyclists were involved in 72% of all accidents in Indonesia, not very different to the percentage of motorcycles. However, a significant proportion of these accidents are fatal and, on average, one fatal accident every hour is attributed to motorcycles (Statistics Indonesia, 2012). Tagel’s (2013) studied a provincial capital in Indonesia, Denpasar, and showed that while motorcyclists represented only 82% of mode share they were responsible for 87% of traffic violations recorded in 2011 (Denpasar BPS, 2014). At the same time, private cars (14% of the mode share) were responsible for 6% of the traffic violations and the goods vehicles (4% of the total mode share) responsible for 3% of the traffic violations. It is important to consider motorcyclists as a distinct group from other road users. The number of motorcyclists is growing rapidly and they have been widely perceived as relatively more aggressive in their road use behaviour, as observed in other countries. The
Y.O. Susilo et al. / Accident Analysis and Prevention 75 (2015) 272–284
risk-taking behaviour of motorcyclists has been of serious concern also in Australia, India, and many other countries (e.g. Rowden et al., 2009; Dandona et al., 2006). Many studies (e.g. Jamson, 2004; Özkan et al., 2012) have shown that disregarding road rules, errors of judgement and aggression on the road are a result of a complex interaction of intentions, attitudes and the general desire to satisfy sensation-seeking behaviour. Whilst interactions of these behaviours in motorcycle accidents have been studied in developed countries (e.g. Elliott et al., 2007; Watson et al., 2007; Steg and van Brussel, 2009; Musselwhite et al., 2012, etc.), little is understood about the impacts of motorcyclist attitudes and social norms on their habit of violating traffic rules in developing countries. Given that social and cultural characteristics of developing countries are different to developed countries, it is important to understand the impacts and interactions of these attitudes and sensation-seeking behaviour in the local context. This is critical to the development of policies and intervention measures aimed at changing the behaviour of motorcyclists and to fostering a safe driving environment in developing countries (Steg and van Brussel, 2009; Joewono et al., 2014; Musselwhite et al., 2014). Motorcycle engines in Indonesia and other developing countries are generally less powerful than in developed countries. In Indonesia, the typical motorcycle is powered by an engine in the range of 125–150 cc, and often imported from China or Japan. Currently, Honda enjoys the majority market share. Some motor scooters can be in the 50 cc category and only a few motorcycles have engines above 500 cc (Statistics Indonesia, 2012). In the last decade, motorcycles with automatic transmission have rapidly becoming popular. In 2013, more than 50% of new motorcycles were sold with automatic transmission. These smaller and (semi-) automatic types of motorcycles make different cognitive and concentration demands on motorcycle riders in Indonesia to motorcyclists in US and European countries, who usually use motorcycles with a bigger engine size. Automatic transmission motorcycles give these scooter riders an opportunity to control the vehicle with one hand only. In this way, riders are able to engage in other activities, such as texting and eating, which riders of conventional geared motorcycles generally cannot do. Nonetheless, it is important to note here that multitasking activities whilst driving a vehicle, including listening to music, smoking and texting, are prohibited in Indonesia (National Traffic Law, 2009). However, despite the fact that these violations entitle the traffic police to issue a fine to the motorcyclist on the spot, these multitasking and many other traffic violation behaviours are still very common among Indonesian motorcyclists. To understand the underlying reasons for motorcyclists’ behaviours in repetitively disregarding traffic regulations, this study aims to investigate the effects of motorcyclists’ attitudes, habits, preferences, and travel pattern on these types of repetitive traffic violations using data from three major metropolitan cities in Indonesia (Bandung, Yogyakarta and Surabaya). A structural equation modelling technique with a structure design based on the theory of planned behaviour (Ajzen, 1991) is employed here to investigate interactions between internal and external characteristics that influence attitudes and self-justification in disregarding traffic regulations, including the impacts of social norms and lack of law enforcement. In the next section, a brief discussion of the unique attitudes and behaviour of risk taking and thrill seeking prevalent among motorists is provided. A description of the proposed model and data collection process follows. Then, the analysis and interpretation of the model results are presented. A final discussion is included in Section 6. 2. Attitudes and risk-taking behaviour among motorcyclists Motorcycle safety has been a focus of safety experts for decades. The main reason for this interest is the high probability of serious
273
injuries and fatalities to riders involved in such accidents. A study in Taiwan has shown that on average, motorcyclists have approximately three times the risk of fatality than non-motorcycle drivers after adjusting for mileage (Chang and Yeh, 2006). In the UK, a DETR (2000) study has also shown similar outcomes, where motorcyclists killed or sustaining serious injuries, per million vehicle kilometres, is approximately twice that of pedal cyclists and over 16 times that of car drivers and passengers. A similar proportion is also found among Swedish motorcycle and moped riders, compared to car passengers per distance travelled (Aare and von Holst, 2003). This accident severity condition is not solely due to the design properties of the two wheel vehicle and the vulnerability of riders due to lack restraint and protection compared to car drivers. Motorcyclist behaviour is also partly responsible. For example, Dandona et al. (2006) argued that the reason that motorcyclists in India have a high level of risk of road injuries is that many of them frequently ignore traffic rules, in addition to a significant proportion being unlicensed drivers, shunning use of helmets and driving vehicles in poor condition. Moreover, Chang and Yeh (2006) highlighted that poor driving skills and less experience were increasing the risk of accidents for all. They also found that young male motorcyclists were more likely to disobey traffic regulations and that young riders in particular have a greater tendency to ignore potential risks and motorcycle safety checks. Musselwhite et al. (2012) found that motorcyclists are aware that being on a motorbike exposes them to greater danger. However, they tend to view safety in terms of being able to handle the bike, knowing its limitations and capabilities, without having to lose the thrill of riding. These researchers highlight the need for an in-depth study going beyond impacts and attitudes towards behaviours to investigate motives and decision-making processes behind risky behaviours and attitudes of motorcyclists. Among the motorcyclist population, young adult motorcyclists have been considered a high risk population (Wong et al., 2010), mainly because they are eager to compromise their behaviours, more than other socio-demographic groups. Steg and van Brussel (2009) noted that violations are deliberate actions that result from social and motivational factors. Based on their study in Australia, Watson et al. (2007) indicated that risktaking intentions of motorcycle riders were primarily influenced by attitudes and sensation seeking, while safe behaviour intentions were influenced by perceived behavioural control. These attitudes and intentions vary according to socio-demographic groups, social norm and past experience. These findings are also consistent with Elliot et al.’s study (2007) in the United Kingdom. Jamson (2004) found that past behaviour (i.e. engaging in risky behaviour in the past), attitudes and behavioural beliefs (e.g. beliefs that speeding is enjoyable and that speeding allows me to beat the traffic) emerged as significant predictors of intentions to engage in various risky riding behaviours among older motorcyclists. In term of social influence, Steg and van Brussel (2009),Jamson (2004) and Watson et al. (2007) found that riders who had a positive attitude towards speeding were more likely to speed. It was also found that riders who had a stronger intention to ‘show off’ when they had a positive attitude towards speeding and thought of having others’ approval of their behaviours were more likely to perform stunts (Özkan et al., 2012). At the same time, younger and male drivers tend to express a lower level of normative motivation to comply with traffic laws than female and older drivers (Yagil, 1998). Tunnicliff et al. (2012) describes this propensity for sensation seeking and aggression as a factor in motorcycle riders’ decisions to engage in unsafe and, particularly, risky behaviour. High sensation seekers are proposed to either underestimate or accept risks as the price of the sensation or experience.
274
Y.O. Susilo et al. / Accident Analysis and Prevention 75 (2015) 272–284
Whilst the interactions of these behaviours in motorcycle and other motorised vehicles’ accidents have been well studied in developed countries, there is a lack of knowledge about how these behaviours and social norms influence traffic violation habits in developing countries. It is believed that both law enforcement and level of traffic obedience in developing countries are much weaker than in developed countries. To fully understand these interactions, this study intends to systematically analyse patterns and choices influenced by peer behaviour, values and cultures. Ajzen (1991) argued that individual action was a function of individual intention, influenced by the individual’s attitude towards the behaviour, subjective norms and perceived behavioural control to execute the choices. These aspects are framed together under the theory of planned behaviour (Ajzen, 1991). Attitude here refers to evaluation of a behaviour, which could be favourable or unfavourable. Perceived behavioural control refers to how easy or difficult it would be to carry out an act. Subjective norm describes the perceived pressure from others to commit the behaviour. Jakobsson et al. (2002) argue that the norms, trust and shared values between individuals and within social groups shape individual preferences and dissuade behaviour different to the acceptable norm. This concept is valid in the context of the behaviour of road users violating traffic regulations. Previously, Steg and van Brussel (2009) and also other studies (e.g. Özkan et al., 2012) have successfully used this concept to explain the impacts of attitudes, subjective norm and perceived behaviour control to explain the variance of speeding behaviours among Dutch moped riders. Nevertheless, it is important to be noted here that the use of Ajzen’s (1991) concept to explain the behaviour is very practical and it does not explain the cause of the behavioural change. Furthermore, it is important to understand that the relative importance of attitudes, subjective norms and perceived of behavioural control for the prediction of intentions vary from behaviour to behaviour and population to population (Ajzen and Fishbein, 2004). Thus, the estimated relationships based on a developing country data are likely to be different from the ones based on data from developed countries. There are several plausible hypotheses that can be considered based on previous findings. It could be expected that beliefs and attitudes of individuals, social norm and perceived behavioural control would significantly influence repetitive violation behaviour. It could also be expected that different types of violations would have different frequencies of occurrences, depending on influencing factors and conditions involved. Unlike studies in developed countries, this case study took place in developing Asian cities, where people hesitate to be seen as different to others. Thus, it was expected that social norm may play a stronger role in encouraging (or discouraging) motorcyclists to disregard traffic rules than in developed countries. 3. Data collection and sample profiles A total of nearly 3000 motorcyclists from three major cities in Indonesia, i.e. Surabaya, Bandung and Yogyakarta, were interviewed. Surabaya is the capital of East Java province and the second largest city in Indonesia with a population of 2.7 million (5.6 million within the greater metropolitan area). Bandung is the capital of West Java province and, with its conurbation, has a population of about 2.5 million. Yogyakarta is the provincial capital of the Special Region of Yogyakarta. It has a population of half a million. These three cities were considered as a representative sample of cities according to the spread of population, travel mode share and land use characteristics in Indonesia. Surabaya is a major business centre and port city. Surabaya is the capital of East Java province and had the
highest proportion of motorcycle ownership from 2007 to 2013 (BPS, 2011, 2012, 2013). Bandung and Yogyakarta are well known as university towns. However, the cultures and the city and geographical layouts of these two cities are very different. Bandung is a medium to large sized Sundanese capital city and is located around 2 h drive from the national capital, Jakarta. Culturally, the city is very modern and is one of the most popular destinations for Jakarta residents at weekends. Yogyakarta is also a medium sized city but is distant (550 km) from the national capital. It is the capital of Yogyakarta province. Bandung has a hilly and mountainous terrain whereas Yogyakarta has a generally flat terrain amenable to pedal cycle use. Yogyakarta is associated with a rich cultural Javanese heritage including the popularity of a traditional man powered tricycle known as a becak. On the other hand, Bandung has a traditional form of paratransit (minibus service) known as angkot (Susilo, 2014; Susilo et al., 2010). To meet the statistical requirements, 1000 respondents were selected from each city. Based on a suggestion from Israel (1992) in determining a minimum sample size, a sample of 400 was recommended. This was suitable for a population size of over 100,000 motorcycles with a 5% level of precision. Data showed that there were 448,651 motorcycles in Bandung in 2006, (BPS Bandung, 2007) and 256,224 in Yogyakarta in 2007 (BPS Yogyakarta, 2009), while in Surabaya in 2010 there were 3,122,901 units (Kompas, 2011). The number of respondents targeted was 1000 in order to anticipate non-responses or incomplete answers. Face to face interviews were conducted at a variety of public areas such as transport terminals, bus stops, malls, schools and public service offices. Only drivers of motorcycles were admitted to participate in the survey. To achieve that, accompanying passengers were rejected during the pre-selection process. At the completion of the survey, there were 983, 980, and 978 samples that could be used for further analysis for Bandung, Yogyakarta and Surabaya, respectively. The questionnaire attempted to collect information regarding (1) general socio-demographic characteristics such as age, gender, marital status, position in family, level of education, employment status and household size of the motorcyclist; (2) usage characteristics of the motorcycle such as average frequency of daily usage, average distance per trip, average distance per day, average travel time per day, motorcycle ownership, reason for using the motorcycle; (3) more detailed information on the typical daily motorcycle trips such as average travel time per trip, trip purpose, number of accompanying passengers and level of usage of motorcycle for intercity trips; (4) the reasons of why they violated the traffic rules. This includes the possible impacts of external factors, such as road and built environment conditions, vehicle, weather and traffic characteristics, influencing the frequency of violation. Beside those, the respondents were also asked to report (5) the frequency of violating traffic rules and reasons underlying the behaviour; and (6) their views on the seriousness of the violations, (7) their habit of violating traffic rules and (8) their believe about norms and level of acceptance in committing to such violations. Answers were recorded in a Likert-scale manner. Responses were scored in a range 1–4 from ‘very often’ to ‘never’, for questions that related to the frequency of violations. The range was 1–5, from ‘very often violate’ to ‘never violate’ the traffic rules, for questions that explored the perceived effects of internal and external factors that influence personal behaviour and decisions. More detailed information regarding the survey was provided by Joewono et al. (2012) and Joewono et al. (2014). Various instruments have been proposed to assess driving style, and relationships between driving style and accident/crash involvement in the past. These include the Driver Behaviour Questionnaire (DBQ, Reason et al., 1990), which has subsequently been the inspiration for various more recent measurements,
Y.O. Susilo et al. / Accident Analysis and Prevention 75 (2015) 272–284
including the Motorcycle Rider Behaviour Questionnaire (MRBQ, Elliott et al., 2007). They distinguished three types of behaviours that increase the risks of being involved in an accident: errors, lapses and violations. These three elements have different psychological origins and should be analysed in different ways. Steg and van Brussel (2009) explained that errors refer to occasions when a planned sequence of activities fails to achieve its intended outcome; they reflect driving mistakes involving failures of observation and misjudgement, such as breaking too quickly on a slippery road or misjudging the crossing interval when turning (Reason et al., 1990). Lapses refer to attention and memory failures that are relatively harmless, such as hitting something inadvertently when reversing or attempting to drive away from traffic lights in the wrong gear. Violation, which is the focus of this study, refers to deliberate deviations from normal safe driving practices (Reason et al., 1990), which may be deviation from formal prohibitions as well as from unofficial local norms of what is deemed to be safe or acceptable. Examples are speeding and overtaking on the inside. Errors and lapses are related to cognitive processes of the individual and are generally not deliberate actions, whereas violations result from motivational processes and deliberate actions (cf. Steg and van Brussel, 2009; Reason et al., 1990). The types of violations that are included in this study are the violations that are currently outlawed and/or barred by the present Indonesian National Traffic Regulations (National Traffic Law, 2009). The general breakdown of respondent profile can be seen in Table 1. The dotted lines separate different headings of demographic characteristics considered. As shown in Table 1, about 70–80% of respondents are in the young productive age group (17–39 years old), with the 17–29 years old group representing half of the respondents. There were only a few motorcyclists over the age of 50 years in the samples. There are also more male than female respondents, especially in Bandung. The majority of the respondents were students in Bandung and Yogyakarta, and private employees in Surabaya. In this study, the respondents were asked to state the type of driving licence(s) owned. Possible answers were: no licence, licence category A (driving licence designated only for common and utility car), category B (driving licence only for truck, bus or trailer), category C (driving licence designated only for motorcycle), or having more than one type of driving licence (of any combination). Since the target of respondents is motorcyclists, as expected, the highest proportion of driving licence type owned by respondents was category C. Locals refer to a driving licence as SIM (Surat Ijin Mengemudi)1 and this category is locally known as SIM-C. Worryingly, Table 1 shows that 23% of respondents in Surabaya rode motorcycles without having an appropriate licence (i.e. riding a motorcycle with a car licence or no licence). Also, the proportion of unlicensed motorcycle drivers in Yogyakarta and Bandung is 14% and 11%, respectively. 3.1. Trip pattern and trip purposes Respondents were also requested to provide information regarding their average motorcycle usage per trip and per day. These questions were intended to reveal the general pattern of motorcycle usage. Respondents reported their most frequent type of travel pattern which is not necessarily the same as during the survey day. Questions were provided in a closed structured form, which meant the respondents selected one of the available options
1 The general requirement for having a driving license in Indonesia is that you must be at least 17 years old, pass the theory test and pass the practical test. For some cases and for renewal of licenses, medical and eye examinations are required.
275
as applicable from the second column (titled Characteristics) of Table 2 for each respective city. In this study, one trip is defined as a one-way trip performed for a particular trip purpose (e.g. home–destination–home is counted as two trips). As shown in the table, different cities have significant differences in motorcycle use, across different travel characteristics, when compared to each other. The statistical test indicators can be seen in the left column of the table. Nevertheless, it is consistent that the main trip purposes of motorcyclists in all three cities are related to work and study. Most respondents performed two trips a day. However, a significant proportion of respondents used a motorcycle more than four times per day. In terms of travel time by motorcycle, most trips took 15–30 min and the second largest group took between 30 min and 1 h. Almost all of the respondents travelled further than 1 km per trip with their motorcycle. The results also show that many respondents were using a motorcycle for relatively long-distance travel (perhaps across the city) and motorcycle use is not limited to short trips near home. 3.2. Types of reported violations As mentioned earlier, the type of violations that are used in this study are those that are currently outlawed and/or barred by the current Indonesian National Traffic Regulations (National Traffic Law, 2009) and those that are widely enforced by various community groups and/or local neighbourhoods in Indonesia.2 The reported violation frequencies of such regulations in different cities are shown in Table 3. In this table, the violations are grouped according to the order of severity of the penalty that the rider can receive according to the National Law (National Traffic Law, 2009). Group-a violations are those that have both fine and jail consequences, whilst group-b has only fine consequences. Group-c includes violations that are strongly enforced by the community as a part of a widely accepted norm but are not written in the National Law and group-d refers to violations resulting in an accident which carry a lengthy jail term. It is important to note here that, unlike in developed countries, law enforcement in Indonesia towards motorcycle riders is rather lenient. Thus, despite different levels of penalties being attributed to different levels of violations, especially for group-b violations (as shown in Table 3), penalties are hardly implemented to their full extent. Table 3 shows that motorcyclists in Bandung are found to be involved in violations more frequently compared to the other two cities. The most common reported violation in Bandung city is chatting with other motorcyclist riders and/or passengers whilst driving. In the city of Yogyakarta, the most frequent violation is never or rarely turning on the turning-light when turning at an intersection. For motorcyclists in Surabaya, never or rarely obeying traffic signs is the most common type of violation. The least frequent type of violation involving motorcyclists in Bandung is crossing the road via the pedestrian crossing bridge. Driving recklessly and carrying loads that are too heavy or too many people are the least frequent types of violation in Yogyakarta and Surabaya, respectively. Fortunately, despite many repetitive violation behaviours, only a relatively small number of respondents reported trying to run from the scene when involved in a traffic accident. The Kruskal–Wallis test was carried out to test the
2 In Indonesia, similar to other Asian countries, there are some unwritten rules about ‘acceptable’ and ‘unacceptable’ behaviour, mostly related to respecting interests of the local community and social hierarchy. An example of this includes to drive slowly and patiently (no honking) near religious venues and military bases. A repeat violation of this kind can trigger social friction and the violator can receive various “social punishments”, from just being reprimanded by local elders to being jailed by the authorities for causing social disturbance.
276
Y.O. Susilo et al. / Accident Analysis and Prevention 75 (2015) 272–284
Table 1 Respondents’ socio-demographic characteristics. Proportion (%)
Characteristics
Bandung (n = 983)
Yogyakarta (n = 980)
Surabaya (n = 978)
Gender
The traveller is male
82.1
62.8
57.2
Age
Aged Aged Aged Aged Aged
3.8 64.4 20.2 8.7 2.8
4.2 49.3 25.6 12.1 8.7
3.1 49.1 33.9 10.1 3.8
Education
Reached high school and special skills diploma level Reached university or higher level education
63.5 34.6
65.4 32.1
65.8 33.1
Employment status
The The The The
42.3 52.0 3.4 2.3
52.8 39.9 4.7 2.7
66.0 27.8 4.3 1.9
Motorcycle ownership
Own the motorcycle
53.6
60.1
65.2
Driving licence ownership
Not own any driving licence Have a driving licence just for common and utility car (Category A) Have a driving licence just for truck and bus (Category B) Have a driving licence just for motorcycle (Category C) Have more than one driving licence (any combination)
7.5 2.8 0.3 47.4 41.9
12.9 1.2 0.1 68.1 17.8
14.9 7.1 0.6 56.4 21.0
17 or less between 17 and 29 between 30 and 39 between 40 and 49 50 or more
traveller traveller traveller traveller
is is is is
a full-time employed a student/pupil a housewife an unemployed or retired person
significant differences between the types of violations found within the three cities. It was found that, apart from two attributes that were not significantly different among cities (see Table 3) the rest of the violation distributions are statistically different from each other. 3.3. Reported reasons underlying the behaviours Table 4 shows the self-reported reasons of the respondents for violating and/or disregarding traffic regulations. The Kruskal–Wallis test was carried out to confirm that the reported
reasons for violation behaviours are statistically different among the studied cities. It is shown that Bandung provided smaller mean scores, which means a higher frequency, for self-reported repetitive violations than the other two cities (the range value of the variables was 1–5, from ‘very often violate’ to ‘never violate’). Presumably this was because Bandung respondents were predominantly young male students, who were more likely to violate traffic laws than others. This trend is consistent with Jonah (1986) and Wong et al. (2010) who highlighted an increased level of accident risk among young motorists and potential factors that may account for it. The other
Table 2 The motorcycle usage of the respondents. Characteristics
x2; df; p-value
Proportion (%) Bandung (n = 983)
Yogyakarta (n = 980)
Surabaya (n = 978)
Trip purpose
Leisure Working Studying Visiting friends or relatives Religious activities Other
19.4 31.6 37.4 2.5 0.8 8.1
16.6 42.3 28.6 3.9 0.7 7.9
15.8 54.5 22.0 1.9 1.0 4.7
121.919; 10;
Contents lists available at ScienceDirect
Accident Analysis and Prevention journal homepage: www.elsevier.com/locate/aap
Reasons underlying behaviour of motorcyclists disregarding traffic regulations in urban areas of Indonesia Yusak O. Susilo a, * , Tri Basuki Joewono b , Upali Vandebona c a b c
Department of Transport Science, Royal Institute of Technology (KTH), Teknikringen 10, 100 44 Stockholm, Sweden Graduate Program, Parahyangan Catholic University, Jl. Merdeka 30 Bandung 40117, Indonesia School of Civil and Environmental Engineering, University of New South Wales, Kensington, Campus, Sydney 2052, Australia
A R T I C L E I N F O
A B S T R A C T
Article history: Received 7 May 2014 Received in revised form 16 December 2014 Accepted 16 December 2014 Available online xxx
Over the last decade, motorcycle use has been rapidly increasing in Indonesia as have violations of traffic rules committed by motorcyclists. This study aims to explore the impacts of motorcyclists’ attitudes, habits, preferences, and travel patterns on their behaviour in disregarding traffic regulations in three cities in Indonesia. The theory of planned behaviour and structural equation modelling are employed to explore these relationships. Consistent with results from previous studies in developed countries, an individual’s beliefs and attitudes, social norms and perceived behaviour control significantly influence behaviour in disregarding traffic rules. Young adults and students are found to be more likely to frequently violate traffic regulations. However, unlike previous findings from developed countries, in Indonesia, males are less likely to disregard traffic rules than females. Overall, pushing the motorcycle through a (very) narrow gap, speeding, driving recklessly, and overtaking on the wrong side are the most frequent traffic violations that make up repetitive violation behaviour among urban motorcyclists in Indonesia. The results highlight the need to revisit Indonesian National Traffic Law traffic violation classification and penalties and separate violations that are likely to cause fatal results, thus requiring tougher law enforcement, from violations that are unlikely to have fatal consequences. ã 2014 Elsevier Ltd. All rights reserved.
Keywords: Motorcyclist Repetitive traffic violation Indonesia Psychological factors Social norms Perceived control Structural equation modelling
1. Introduction In the last two decades, economic conditions and motorisation in Southeast Asian countries have experienced a rapid growth rate. The motorcycle has become a popular transport mode in this region, including Indonesia, Vietnam and Thailand (Kaltheier, 2002). Currently, Indonesia is the third largest motorcycle market in the world after China and India. In Indonesia, there were 13.6 million motorcycles in 2000, increasing to 32.5 million in 2006 and 76.4 million in 2012. In 2012, the total number of motorised vehicles in Indonesia was 94.4 million (Statistics Indonesia, 2012). It has been estimated that by 2015, there will be one motorcycle for every two persons in this country (AISI, 2009) which has a population of about 240 million. The motorcycle is a popular transport mode among Southeast Asian travellers mainly because of the affordability, flexibility, and manoeuvrability on congested roads typical in the region.
* Corresponding author. Tel.: +46 87909635. E-mail addresses: [email protected] (Y.O. Susilo), [email protected] (T.B. Joewono), [email protected] (U. Vandebona). http://dx.doi.org/10.1016/j.aap.2014.12.016 0001-4575/ ã 2014 Elsevier Ltd. All rights reserved.
At the same time, the growth of motorcycle users has been seen as disruptive to users of other travel modes. Aggressive encroachment of pedestrian space and lack of compliance with road rules have been widely considered as a major cause of road accidents (Indriastuti and Sulistio, 2010) although statistically, motorcycle-related accidents are not over-represented in Indonesian traffic accident statistics. In 2012, motorcyclists were involved in 72% of all accidents in Indonesia, not very different to the percentage of motorcycles. However, a significant proportion of these accidents are fatal and, on average, one fatal accident every hour is attributed to motorcycles (Statistics Indonesia, 2012). Tagel’s (2013) studied a provincial capital in Indonesia, Denpasar, and showed that while motorcyclists represented only 82% of mode share they were responsible for 87% of traffic violations recorded in 2011 (Denpasar BPS, 2014). At the same time, private cars (14% of the mode share) were responsible for 6% of the traffic violations and the goods vehicles (4% of the total mode share) responsible for 3% of the traffic violations. It is important to consider motorcyclists as a distinct group from other road users. The number of motorcyclists is growing rapidly and they have been widely perceived as relatively more aggressive in their road use behaviour, as observed in other countries. The
Y.O. Susilo et al. / Accident Analysis and Prevention 75 (2015) 272–284
risk-taking behaviour of motorcyclists has been of serious concern also in Australia, India, and many other countries (e.g. Rowden et al., 2009; Dandona et al., 2006). Many studies (e.g. Jamson, 2004; Özkan et al., 2012) have shown that disregarding road rules, errors of judgement and aggression on the road are a result of a complex interaction of intentions, attitudes and the general desire to satisfy sensation-seeking behaviour. Whilst interactions of these behaviours in motorcycle accidents have been studied in developed countries (e.g. Elliott et al., 2007; Watson et al., 2007; Steg and van Brussel, 2009; Musselwhite et al., 2012, etc.), little is understood about the impacts of motorcyclist attitudes and social norms on their habit of violating traffic rules in developing countries. Given that social and cultural characteristics of developing countries are different to developed countries, it is important to understand the impacts and interactions of these attitudes and sensation-seeking behaviour in the local context. This is critical to the development of policies and intervention measures aimed at changing the behaviour of motorcyclists and to fostering a safe driving environment in developing countries (Steg and van Brussel, 2009; Joewono et al., 2014; Musselwhite et al., 2014). Motorcycle engines in Indonesia and other developing countries are generally less powerful than in developed countries. In Indonesia, the typical motorcycle is powered by an engine in the range of 125–150 cc, and often imported from China or Japan. Currently, Honda enjoys the majority market share. Some motor scooters can be in the 50 cc category and only a few motorcycles have engines above 500 cc (Statistics Indonesia, 2012). In the last decade, motorcycles with automatic transmission have rapidly becoming popular. In 2013, more than 50% of new motorcycles were sold with automatic transmission. These smaller and (semi-) automatic types of motorcycles make different cognitive and concentration demands on motorcycle riders in Indonesia to motorcyclists in US and European countries, who usually use motorcycles with a bigger engine size. Automatic transmission motorcycles give these scooter riders an opportunity to control the vehicle with one hand only. In this way, riders are able to engage in other activities, such as texting and eating, which riders of conventional geared motorcycles generally cannot do. Nonetheless, it is important to note here that multitasking activities whilst driving a vehicle, including listening to music, smoking and texting, are prohibited in Indonesia (National Traffic Law, 2009). However, despite the fact that these violations entitle the traffic police to issue a fine to the motorcyclist on the spot, these multitasking and many other traffic violation behaviours are still very common among Indonesian motorcyclists. To understand the underlying reasons for motorcyclists’ behaviours in repetitively disregarding traffic regulations, this study aims to investigate the effects of motorcyclists’ attitudes, habits, preferences, and travel pattern on these types of repetitive traffic violations using data from three major metropolitan cities in Indonesia (Bandung, Yogyakarta and Surabaya). A structural equation modelling technique with a structure design based on the theory of planned behaviour (Ajzen, 1991) is employed here to investigate interactions between internal and external characteristics that influence attitudes and self-justification in disregarding traffic regulations, including the impacts of social norms and lack of law enforcement. In the next section, a brief discussion of the unique attitudes and behaviour of risk taking and thrill seeking prevalent among motorists is provided. A description of the proposed model and data collection process follows. Then, the analysis and interpretation of the model results are presented. A final discussion is included in Section 6. 2. Attitudes and risk-taking behaviour among motorcyclists Motorcycle safety has been a focus of safety experts for decades. The main reason for this interest is the high probability of serious
273
injuries and fatalities to riders involved in such accidents. A study in Taiwan has shown that on average, motorcyclists have approximately three times the risk of fatality than non-motorcycle drivers after adjusting for mileage (Chang and Yeh, 2006). In the UK, a DETR (2000) study has also shown similar outcomes, where motorcyclists killed or sustaining serious injuries, per million vehicle kilometres, is approximately twice that of pedal cyclists and over 16 times that of car drivers and passengers. A similar proportion is also found among Swedish motorcycle and moped riders, compared to car passengers per distance travelled (Aare and von Holst, 2003). This accident severity condition is not solely due to the design properties of the two wheel vehicle and the vulnerability of riders due to lack restraint and protection compared to car drivers. Motorcyclist behaviour is also partly responsible. For example, Dandona et al. (2006) argued that the reason that motorcyclists in India have a high level of risk of road injuries is that many of them frequently ignore traffic rules, in addition to a significant proportion being unlicensed drivers, shunning use of helmets and driving vehicles in poor condition. Moreover, Chang and Yeh (2006) highlighted that poor driving skills and less experience were increasing the risk of accidents for all. They also found that young male motorcyclists were more likely to disobey traffic regulations and that young riders in particular have a greater tendency to ignore potential risks and motorcycle safety checks. Musselwhite et al. (2012) found that motorcyclists are aware that being on a motorbike exposes them to greater danger. However, they tend to view safety in terms of being able to handle the bike, knowing its limitations and capabilities, without having to lose the thrill of riding. These researchers highlight the need for an in-depth study going beyond impacts and attitudes towards behaviours to investigate motives and decision-making processes behind risky behaviours and attitudes of motorcyclists. Among the motorcyclist population, young adult motorcyclists have been considered a high risk population (Wong et al., 2010), mainly because they are eager to compromise their behaviours, more than other socio-demographic groups. Steg and van Brussel (2009) noted that violations are deliberate actions that result from social and motivational factors. Based on their study in Australia, Watson et al. (2007) indicated that risktaking intentions of motorcycle riders were primarily influenced by attitudes and sensation seeking, while safe behaviour intentions were influenced by perceived behavioural control. These attitudes and intentions vary according to socio-demographic groups, social norm and past experience. These findings are also consistent with Elliot et al.’s study (2007) in the United Kingdom. Jamson (2004) found that past behaviour (i.e. engaging in risky behaviour in the past), attitudes and behavioural beliefs (e.g. beliefs that speeding is enjoyable and that speeding allows me to beat the traffic) emerged as significant predictors of intentions to engage in various risky riding behaviours among older motorcyclists. In term of social influence, Steg and van Brussel (2009),Jamson (2004) and Watson et al. (2007) found that riders who had a positive attitude towards speeding were more likely to speed. It was also found that riders who had a stronger intention to ‘show off’ when they had a positive attitude towards speeding and thought of having others’ approval of their behaviours were more likely to perform stunts (Özkan et al., 2012). At the same time, younger and male drivers tend to express a lower level of normative motivation to comply with traffic laws than female and older drivers (Yagil, 1998). Tunnicliff et al. (2012) describes this propensity for sensation seeking and aggression as a factor in motorcycle riders’ decisions to engage in unsafe and, particularly, risky behaviour. High sensation seekers are proposed to either underestimate or accept risks as the price of the sensation or experience.
274
Y.O. Susilo et al. / Accident Analysis and Prevention 75 (2015) 272–284
Whilst the interactions of these behaviours in motorcycle and other motorised vehicles’ accidents have been well studied in developed countries, there is a lack of knowledge about how these behaviours and social norms influence traffic violation habits in developing countries. It is believed that both law enforcement and level of traffic obedience in developing countries are much weaker than in developed countries. To fully understand these interactions, this study intends to systematically analyse patterns and choices influenced by peer behaviour, values and cultures. Ajzen (1991) argued that individual action was a function of individual intention, influenced by the individual’s attitude towards the behaviour, subjective norms and perceived behavioural control to execute the choices. These aspects are framed together under the theory of planned behaviour (Ajzen, 1991). Attitude here refers to evaluation of a behaviour, which could be favourable or unfavourable. Perceived behavioural control refers to how easy or difficult it would be to carry out an act. Subjective norm describes the perceived pressure from others to commit the behaviour. Jakobsson et al. (2002) argue that the norms, trust and shared values between individuals and within social groups shape individual preferences and dissuade behaviour different to the acceptable norm. This concept is valid in the context of the behaviour of road users violating traffic regulations. Previously, Steg and van Brussel (2009) and also other studies (e.g. Özkan et al., 2012) have successfully used this concept to explain the impacts of attitudes, subjective norm and perceived behaviour control to explain the variance of speeding behaviours among Dutch moped riders. Nevertheless, it is important to be noted here that the use of Ajzen’s (1991) concept to explain the behaviour is very practical and it does not explain the cause of the behavioural change. Furthermore, it is important to understand that the relative importance of attitudes, subjective norms and perceived of behavioural control for the prediction of intentions vary from behaviour to behaviour and population to population (Ajzen and Fishbein, 2004). Thus, the estimated relationships based on a developing country data are likely to be different from the ones based on data from developed countries. There are several plausible hypotheses that can be considered based on previous findings. It could be expected that beliefs and attitudes of individuals, social norm and perceived behavioural control would significantly influence repetitive violation behaviour. It could also be expected that different types of violations would have different frequencies of occurrences, depending on influencing factors and conditions involved. Unlike studies in developed countries, this case study took place in developing Asian cities, where people hesitate to be seen as different to others. Thus, it was expected that social norm may play a stronger role in encouraging (or discouraging) motorcyclists to disregard traffic rules than in developed countries. 3. Data collection and sample profiles A total of nearly 3000 motorcyclists from three major cities in Indonesia, i.e. Surabaya, Bandung and Yogyakarta, were interviewed. Surabaya is the capital of East Java province and the second largest city in Indonesia with a population of 2.7 million (5.6 million within the greater metropolitan area). Bandung is the capital of West Java province and, with its conurbation, has a population of about 2.5 million. Yogyakarta is the provincial capital of the Special Region of Yogyakarta. It has a population of half a million. These three cities were considered as a representative sample of cities according to the spread of population, travel mode share and land use characteristics in Indonesia. Surabaya is a major business centre and port city. Surabaya is the capital of East Java province and had the
highest proportion of motorcycle ownership from 2007 to 2013 (BPS, 2011, 2012, 2013). Bandung and Yogyakarta are well known as university towns. However, the cultures and the city and geographical layouts of these two cities are very different. Bandung is a medium to large sized Sundanese capital city and is located around 2 h drive from the national capital, Jakarta. Culturally, the city is very modern and is one of the most popular destinations for Jakarta residents at weekends. Yogyakarta is also a medium sized city but is distant (550 km) from the national capital. It is the capital of Yogyakarta province. Bandung has a hilly and mountainous terrain whereas Yogyakarta has a generally flat terrain amenable to pedal cycle use. Yogyakarta is associated with a rich cultural Javanese heritage including the popularity of a traditional man powered tricycle known as a becak. On the other hand, Bandung has a traditional form of paratransit (minibus service) known as angkot (Susilo, 2014; Susilo et al., 2010). To meet the statistical requirements, 1000 respondents were selected from each city. Based on a suggestion from Israel (1992) in determining a minimum sample size, a sample of 400 was recommended. This was suitable for a population size of over 100,000 motorcycles with a 5% level of precision. Data showed that there were 448,651 motorcycles in Bandung in 2006, (BPS Bandung, 2007) and 256,224 in Yogyakarta in 2007 (BPS Yogyakarta, 2009), while in Surabaya in 2010 there were 3,122,901 units (Kompas, 2011). The number of respondents targeted was 1000 in order to anticipate non-responses or incomplete answers. Face to face interviews were conducted at a variety of public areas such as transport terminals, bus stops, malls, schools and public service offices. Only drivers of motorcycles were admitted to participate in the survey. To achieve that, accompanying passengers were rejected during the pre-selection process. At the completion of the survey, there were 983, 980, and 978 samples that could be used for further analysis for Bandung, Yogyakarta and Surabaya, respectively. The questionnaire attempted to collect information regarding (1) general socio-demographic characteristics such as age, gender, marital status, position in family, level of education, employment status and household size of the motorcyclist; (2) usage characteristics of the motorcycle such as average frequency of daily usage, average distance per trip, average distance per day, average travel time per day, motorcycle ownership, reason for using the motorcycle; (3) more detailed information on the typical daily motorcycle trips such as average travel time per trip, trip purpose, number of accompanying passengers and level of usage of motorcycle for intercity trips; (4) the reasons of why they violated the traffic rules. This includes the possible impacts of external factors, such as road and built environment conditions, vehicle, weather and traffic characteristics, influencing the frequency of violation. Beside those, the respondents were also asked to report (5) the frequency of violating traffic rules and reasons underlying the behaviour; and (6) their views on the seriousness of the violations, (7) their habit of violating traffic rules and (8) their believe about norms and level of acceptance in committing to such violations. Answers were recorded in a Likert-scale manner. Responses were scored in a range 1–4 from ‘very often’ to ‘never’, for questions that related to the frequency of violations. The range was 1–5, from ‘very often violate’ to ‘never violate’ the traffic rules, for questions that explored the perceived effects of internal and external factors that influence personal behaviour and decisions. More detailed information regarding the survey was provided by Joewono et al. (2012) and Joewono et al. (2014). Various instruments have been proposed to assess driving style, and relationships between driving style and accident/crash involvement in the past. These include the Driver Behaviour Questionnaire (DBQ, Reason et al., 1990), which has subsequently been the inspiration for various more recent measurements,
Y.O. Susilo et al. / Accident Analysis and Prevention 75 (2015) 272–284
including the Motorcycle Rider Behaviour Questionnaire (MRBQ, Elliott et al., 2007). They distinguished three types of behaviours that increase the risks of being involved in an accident: errors, lapses and violations. These three elements have different psychological origins and should be analysed in different ways. Steg and van Brussel (2009) explained that errors refer to occasions when a planned sequence of activities fails to achieve its intended outcome; they reflect driving mistakes involving failures of observation and misjudgement, such as breaking too quickly on a slippery road or misjudging the crossing interval when turning (Reason et al., 1990). Lapses refer to attention and memory failures that are relatively harmless, such as hitting something inadvertently when reversing or attempting to drive away from traffic lights in the wrong gear. Violation, which is the focus of this study, refers to deliberate deviations from normal safe driving practices (Reason et al., 1990), which may be deviation from formal prohibitions as well as from unofficial local norms of what is deemed to be safe or acceptable. Examples are speeding and overtaking on the inside. Errors and lapses are related to cognitive processes of the individual and are generally not deliberate actions, whereas violations result from motivational processes and deliberate actions (cf. Steg and van Brussel, 2009; Reason et al., 1990). The types of violations that are included in this study are the violations that are currently outlawed and/or barred by the present Indonesian National Traffic Regulations (National Traffic Law, 2009). The general breakdown of respondent profile can be seen in Table 1. The dotted lines separate different headings of demographic characteristics considered. As shown in Table 1, about 70–80% of respondents are in the young productive age group (17–39 years old), with the 17–29 years old group representing half of the respondents. There were only a few motorcyclists over the age of 50 years in the samples. There are also more male than female respondents, especially in Bandung. The majority of the respondents were students in Bandung and Yogyakarta, and private employees in Surabaya. In this study, the respondents were asked to state the type of driving licence(s) owned. Possible answers were: no licence, licence category A (driving licence designated only for common and utility car), category B (driving licence only for truck, bus or trailer), category C (driving licence designated only for motorcycle), or having more than one type of driving licence (of any combination). Since the target of respondents is motorcyclists, as expected, the highest proportion of driving licence type owned by respondents was category C. Locals refer to a driving licence as SIM (Surat Ijin Mengemudi)1 and this category is locally known as SIM-C. Worryingly, Table 1 shows that 23% of respondents in Surabaya rode motorcycles without having an appropriate licence (i.e. riding a motorcycle with a car licence or no licence). Also, the proportion of unlicensed motorcycle drivers in Yogyakarta and Bandung is 14% and 11%, respectively. 3.1. Trip pattern and trip purposes Respondents were also requested to provide information regarding their average motorcycle usage per trip and per day. These questions were intended to reveal the general pattern of motorcycle usage. Respondents reported their most frequent type of travel pattern which is not necessarily the same as during the survey day. Questions were provided in a closed structured form, which meant the respondents selected one of the available options
1 The general requirement for having a driving license in Indonesia is that you must be at least 17 years old, pass the theory test and pass the practical test. For some cases and for renewal of licenses, medical and eye examinations are required.
275
as applicable from the second column (titled Characteristics) of Table 2 for each respective city. In this study, one trip is defined as a one-way trip performed for a particular trip purpose (e.g. home–destination–home is counted as two trips). As shown in the table, different cities have significant differences in motorcycle use, across different travel characteristics, when compared to each other. The statistical test indicators can be seen in the left column of the table. Nevertheless, it is consistent that the main trip purposes of motorcyclists in all three cities are related to work and study. Most respondents performed two trips a day. However, a significant proportion of respondents used a motorcycle more than four times per day. In terms of travel time by motorcycle, most trips took 15–30 min and the second largest group took between 30 min and 1 h. Almost all of the respondents travelled further than 1 km per trip with their motorcycle. The results also show that many respondents were using a motorcycle for relatively long-distance travel (perhaps across the city) and motorcycle use is not limited to short trips near home. 3.2. Types of reported violations As mentioned earlier, the type of violations that are used in this study are those that are currently outlawed and/or barred by the current Indonesian National Traffic Regulations (National Traffic Law, 2009) and those that are widely enforced by various community groups and/or local neighbourhoods in Indonesia.2 The reported violation frequencies of such regulations in different cities are shown in Table 3. In this table, the violations are grouped according to the order of severity of the penalty that the rider can receive according to the National Law (National Traffic Law, 2009). Group-a violations are those that have both fine and jail consequences, whilst group-b has only fine consequences. Group-c includes violations that are strongly enforced by the community as a part of a widely accepted norm but are not written in the National Law and group-d refers to violations resulting in an accident which carry a lengthy jail term. It is important to note here that, unlike in developed countries, law enforcement in Indonesia towards motorcycle riders is rather lenient. Thus, despite different levels of penalties being attributed to different levels of violations, especially for group-b violations (as shown in Table 3), penalties are hardly implemented to their full extent. Table 3 shows that motorcyclists in Bandung are found to be involved in violations more frequently compared to the other two cities. The most common reported violation in Bandung city is chatting with other motorcyclist riders and/or passengers whilst driving. In the city of Yogyakarta, the most frequent violation is never or rarely turning on the turning-light when turning at an intersection. For motorcyclists in Surabaya, never or rarely obeying traffic signs is the most common type of violation. The least frequent type of violation involving motorcyclists in Bandung is crossing the road via the pedestrian crossing bridge. Driving recklessly and carrying loads that are too heavy or too many people are the least frequent types of violation in Yogyakarta and Surabaya, respectively. Fortunately, despite many repetitive violation behaviours, only a relatively small number of respondents reported trying to run from the scene when involved in a traffic accident. The Kruskal–Wallis test was carried out to test the
2 In Indonesia, similar to other Asian countries, there are some unwritten rules about ‘acceptable’ and ‘unacceptable’ behaviour, mostly related to respecting interests of the local community and social hierarchy. An example of this includes to drive slowly and patiently (no honking) near religious venues and military bases. A repeat violation of this kind can trigger social friction and the violator can receive various “social punishments”, from just being reprimanded by local elders to being jailed by the authorities for causing social disturbance.
276
Y.O. Susilo et al. / Accident Analysis and Prevention 75 (2015) 272–284
Table 1 Respondents’ socio-demographic characteristics. Proportion (%)
Characteristics
Bandung (n = 983)
Yogyakarta (n = 980)
Surabaya (n = 978)
Gender
The traveller is male
82.1
62.8
57.2
Age
Aged Aged Aged Aged Aged
3.8 64.4 20.2 8.7 2.8
4.2 49.3 25.6 12.1 8.7
3.1 49.1 33.9 10.1 3.8
Education
Reached high school and special skills diploma level Reached university or higher level education
63.5 34.6
65.4 32.1
65.8 33.1
Employment status
The The The The
42.3 52.0 3.4 2.3
52.8 39.9 4.7 2.7
66.0 27.8 4.3 1.9
Motorcycle ownership
Own the motorcycle
53.6
60.1
65.2
Driving licence ownership
Not own any driving licence Have a driving licence just for common and utility car (Category A) Have a driving licence just for truck and bus (Category B) Have a driving licence just for motorcycle (Category C) Have more than one driving licence (any combination)
7.5 2.8 0.3 47.4 41.9
12.9 1.2 0.1 68.1 17.8
14.9 7.1 0.6 56.4 21.0
17 or less between 17 and 29 between 30 and 39 between 40 and 49 50 or more
traveller traveller traveller traveller
is is is is
a full-time employed a student/pupil a housewife an unemployed or retired person
significant differences between the types of violations found within the three cities. It was found that, apart from two attributes that were not significantly different among cities (see Table 3) the rest of the violation distributions are statistically different from each other. 3.3. Reported reasons underlying the behaviours Table 4 shows the self-reported reasons of the respondents for violating and/or disregarding traffic regulations. The Kruskal–Wallis test was carried out to confirm that the reported
reasons for violation behaviours are statistically different among the studied cities. It is shown that Bandung provided smaller mean scores, which means a higher frequency, for self-reported repetitive violations than the other two cities (the range value of the variables was 1–5, from ‘very often violate’ to ‘never violate’). Presumably this was because Bandung respondents were predominantly young male students, who were more likely to violate traffic laws than others. This trend is consistent with Jonah (1986) and Wong et al. (2010) who highlighted an increased level of accident risk among young motorists and potential factors that may account for it. The other
Table 2 The motorcycle usage of the respondents. Characteristics
x2; df; p-value
Proportion (%) Bandung (n = 983)
Yogyakarta (n = 980)
Surabaya (n = 978)
Trip purpose
Leisure Working Studying Visiting friends or relatives Religious activities Other
19.4 31.6 37.4 2.5 0.8 8.1
16.6 42.3 28.6 3.9 0.7 7.9
15.8 54.5 22.0 1.9 1.0 4.7
121.919; 10;
