Driving anger in China: Psychometric properties of the Driving Anger Scale (DAS) and its relationship with aggressive driving

https://doi.org/10.1016/j.paid.2014.04.018Get rights and content

Highlights

  • We assessed the psychometric properties of the DAS on a Chinese sample.

  • CFA results supported the original six-factor structure of the DAS.

  • Differences on age groups and countries supported the discriminant validity.

  • Anger from police presence contributed to most of variances in subjective aggression.

  • Anger from slow driving contributed to most of variances in aggressive actions.

Abstract

In this study, we examined the psychometric properties of the Driving Anger Scale (DAS; Deffenbacher, Oetting, & Lynch, 1994) and its relationship with aggressive driving in Chinese context. A total of 411 drivers from five cities in China completed the survey. Confirmatory factor analysis indicated that the fit of the original six-dimension solution was good. Chinese drivers reported lower level of anger than their American counterparts on all six subscales, as well as New Zealand and Spanish drivers on discourtesy and illegal driving. No gender effects were revealed on driving anger. Those drivers reporting a higher level of anger on some subscales tended to be younger, from a more congested city, reporting a lower weekly mileage, and more experienced. The overall driving anger was significantly related to aggressive driving. Further relative importance analysis showed that anger from slow driving, police presence and hostile gestures contributed to most of variances in aggressive driving.

Introduction

The number of vehicles in China has increased dramatically with the strong economic growth (Xu, Li, & Jiang, 2014). As a result, road safety has become a serious problem. In 2011, there were 210,812 traffic accidents involving mortalities in China, leading to 62,387 fatalities, putting the number of road fatalities per 100,000 motor vehicles at 27.8, whereas the numbers in Japan and USA are only 6.1 and 14.9 fatalities per 100,000 vehicles respectively in 2009 (ChinaAudioWeb., 2012).

Driving anger, a situation-bound form of trait anger (Deffenbacher, Oetting, & Lynch, 1994), is one of the most prevalent factors in road accidents (Dahlen et al., 2012, González-Iglesias et al., 2012). High trait-anger drivers might be angered by a wider range of situations, and they experience more intense anger while driving than those low trait-anger drivers (Sullman, 2006). Today, the topic of driving anger has gained popularity among studies (Sullman, Gras, Cunill, Planes, & Font-Mayolas, 2007).

In China, media reports appear to indicate that driving anger is becoming more common. An online survey of 9620 respondents found that 60.7% of them had experienced anger during driving, especially when involved in traffic violations, slow driving, traffic jams and congestion (Sohu, 2008). However, no published academic studies have investigated driving anger in China, although the nature of the phenomenon has been explored in numerous Western countries (e.g., Sullman, 2006, Sullman et al., 2007).

An important issue that has emerged from previous research is the need to examine what sorts of situations provoke anger. The Driving Anger Scale (DAS) developed by Deffenbacher et al. (1994) in the United States is one of the most frequently used instruments for assessing the trait driving anger (Sârbescu, 2012). The original DAS contains 33 common, potentially provocative situations that may arouse anger during driving. A cluster analysis of 1526 incoming freshmen’s responses resulted in six clusters of provocative situations: hostile gestures, illegal driving, police presence, slow driving, discourtesy, and traffic obstructions.

The DAS has been validated in several Western countries, and different factor structures have been revealed. In the UK, factor analysis yielded three dimensions after dropping 6 items that evoked slight anger (Lajunen, Parker, & Stradling, 1998). Later, Sullman (2006) found that four dimensions described his data well in New Zealand when 33 items were subjected to principal component analysis and parallel analysis. More recently, confirmatory factor analysis has indicated that the original six-factor model was satisfactory in the Spanish context (Sullman et al., 2007). Different analytic approaches may be the primary cause for these differences; and cultural differences between these countries may also have some contribution (Sullman et al., 2007). As Yasak and Esiyok (2009) noted that although anger is universal, the situations casing anger vary across culture. For instance, the presence of police did not evoke any appreciable amounts of anger amongst their sample of British drivers, while causing anger amongst Spanish, New Zealand, and American drivers. Then, it is necessary to explore whether all 33 situations would provoke anger in a Chinese context (Lajunen et al., 1998, Sullman, 2006). Also, we wanted to know whether the multidimensionality of driving anger could be verified in China since culture might threaten the factor invariance given the extreme or acquiescence response style (Cheung & Rensvold, 2000). Thus, the primary purposes of this study were to explore whether situations in DAS would cause Chinese drivers to experience anger and what the factor structure of the DAS is in China.

Another concern of this study was the relationship between driving anger and aggressive driving, a construct defined as any form of behavior exhibited by a driver on road with the intent to physically and/or psychological harm other road users (Dula & Geller, 2003). To evaluate the validity of the DAS, Deffenbacher and his colleagues (2001) proposed that high driving anger individuals would engage in more aggressive driving behaviors compared to those with low driving anger. This proposed positive relationship between driving anger and aggressive behavior has been confirmed in several cross-sectional survey studies (Dahlen et al., 2012, Deffenbacher et al., 2001b), in experimental studies (Neighbors, Vietor, & Knee, 2002) and in a meta-analytic review (Nesbit, Conger, & Conger, 2007). However, as a multidimensional construct, driving anger was only measured or operationalized as a single, unidimensional construct in these studies. For instance, Dahlen et al. (2012) measured driving anger with the 14-item short form of the DAS and used the total score in their model. Such a practice might confound the different contributions of each dimension of the DAS to aggression. For example, only anger at slow driving and illegal driving has been found to significantly explain traffic violations (González-Iglesias et al., 2012). Thus, another purpose of the current study was to explore whether the different dimensions of the DAS differentially contribute to aggressive driving and to ascertain the relative importance of each dimension if possible. In so doing, we could be able to identify which dimension of anger contributes the most to aggression on the road and then provide a corresponding training or treatment program to improve driving safety.

Thus, the main purpose of current study was to explore the reliability as well as the construct and discriminant validity of the DAS in the Chinese context. As an extension, the relationship between driving anger and aggressive driving was examined, particularly the relative importance of different types of anger.

Section snippets

Participants and procedure

The current study was one part of a large survey conducted by a professional survey agency in five cities in China (Beijing, Shanghai, Guangzhou, Wuhan, and Chengdu), which are assumed to representative examples of major Chinese cities. The target population of this large survey was office workers who were interviewed by trained personnel in their homes. Before each interview, the interviewer found the target household according to the block sampling plan. Then, the interviewer introduced the

Initial analysis of the items

The means and standard deviations of the 33 DAS items are firstly explored. According to Lajunen et al. (1998), items with means less than 1.5 on the 5-point scale are insufficiently anger-provoking and should not be included in further analysis. This cut-off value has been adopted in later studies (e.g., Sullman, 2006, Sullman et al., 2007). In the current study, all of the mean values of the 33 items were above 1.5 and thus were kept for the following analyses.

Confirmatory factor analysis for the DAS

Using the maximum likelihood

Discussion

The main purpose of the current research was to assess the psychometric properties the DAS in terms of internal consistency reliability, factorial validity, and discriminant validity in the Chinese context. The confirmatory factor analysis showed that the original six-dimension solution found in an American sample (Deffenbacher et al., 1994) provided a good fit to the current data. Each factor loading was significant, with values greater than .60; and all six subscales had good reliability,

Acknowledgments

This work was partly supported by the Scientific Foundation of Institute of Psychology, Chinese Academy of Sciences (No. Y1CX233005) and the Beijing Natural Science Foundation (No. 9112017).

References (26)

Cited by (91)

  • The expression of driving anger in a sample of Italian drivers

    2023, Transportation Research Part F: Traffic Psychology and Behaviour
  • Anger between bus drivers and passengers or among passengers: Development of a bus passenger anger scale (BPAS) and a bus driver anger scale (BDAS)

    2023, Transportation Research Part A: Policy and Practice
    Citation Excerpt :

    Since 1994, various countries have used the DAS and validated it in samples of drivers from their own countries. After more than 20 years of development, many countries have adapted the original DAS to their national contexts and applied these adapted scales (Brandenburg and Oehl, 2021; Li et al., 2014; McLinton and Dollard, 2010; Negre and Delhomme, 2017; Sullman, 2006; Sullman et al., 2014; Yasak and Esiyok, 2009). In addition, Chinese scholars have adapted the scale to professional drivers and bus drivers (Feng et al., 2016; Lin and Deng, 2019).

View all citing articles on Scopus
View full text