driving anger in spain

www.elsevier.com/locate/paid

Personality and Individual Differences 42 (2007) 701–713

Driving anger in Spain

Mark J.M. Sullman a,*, M. Eugenia Gras b, Monica Cunill b,Monserrat Planes b, Silvia Font-Mayolas b

a Centre for Ergonomics and Occupational Health and Safety, Massey University, Palmerston North, New Zealandb Quality of Life Research Institute, Department of Psychology, University of Girona, Spain

Received 7 February 2006; received in revised form 3 August 2006Available online 19 October 2006

Abstract

The present study examined the types of situations that cause Spanish drivers to become angry. The 33-item version of the Driver Anger Scale (Deffenbacher, Oetting, & Lynch, 1994) was used to investigatedriver anger amongst a sample of 371 drivers. Confirmatory factor analysis showed the fit of the originalsix-factor model was satisfactory. In most of the subscales Spanish drivers reported similar levels of angerto that reported by drivers from other nations. Female drivers reported more anger overall and on three ofthe six types of driving anger (discourtesy, traffic obstructions and illegal driving). Individually, police pres-ence, slow driving and overall anger were significant predictors of crash involvement (p < 0.05), even afterthe contributions of the background variables had been partialled out.� 2006 Elsevier Ltd. All rights reserved.

Keywords: Driving behaviour; Anger; Driving

1. Introduction

Driving anger has been defined as a narrow, situationally specific form of the personality char-acteristic trait anger (Deffenbacher et al., 1994). The issue of driving anger appears to have re-cently gained popularity amongst researchers, with a relatively large number of studies being

0191-8869/$ - see front matter � 2006 Elsevier Ltd. All rights reserved.doi:10.1016/j.paid.2006.08.014

* Corresponding author. Tel.: +64 6 350 5799; fax: +64 6 350 5796.E-mail address: [email protected] (M.J.M. Sullman).

mailto:[email protected]

702 M.J.M. Sullman et al. / Personality and Individual Differences 42 (2007) 701–713

published in the last decade (e.g. Dahlen, Martin, Ragan, & Kuhlman, 2005; Deffenbacher, Def-fenbacher, Lynch, & Richards, 2003, 2001; Parker, Lajunen, & Summala, 2002; Sullman, 2006a;Underwood, Chapman, Wright, & Crundall, 1999). One of the reasons cited by almost allresearchers on this topic is the reported increase in levels of anger and resultant road rage inci-dents. However, the evidence for an increase in driving anger is inconclusive with the only peerreviewed research on this topic reporting road rage had not increased in one region of Canada(Smart, Mann, & Zhao, 2005). Whether or not driving anger is increasing, there are a numberof other reasons for studying this phenomenon. Firstly, to become angry whilst driving appearsto be a relatively common event (Deffenbacher, Lynch, & Oetting, 2002). Furthermore, a numberof studies have found that angry drivers engage more often in aggressive and dangerous behav-iours (Dahlen et al., 2005; Deffenbacher et al., 1994, Deffenbacher, Lynch, Oetting, & Yingling,2001). Research has also found significant relationships to exist between driving anger and nearmisses (Underwood et al., 1999) and crash related conditions, such as losing control of their vehi-cle, losing concentration and crashing in simulator studies (Deffenbacher et al., 2001, 2003).

There are a number of ways in which driving anger can be measured, with one of the most com-monly used scales being the Driving Anger Scale (DAS). The 33-item version of the DAS has beenused to measure driving anger amongst drivers from the USA (Deffenbacher et al., 1994), UK (Laj-unen, Parker, & Stradling, 1998), and New Zealand (Sullman, 2006a). Findings using the 33-itemversion of the scale have produced widely different results in terms of the underlying structure ofthe scale. The original research found six different types of driving anger (slow driving, trafficobstructions, police presence, illegal driving, hostile gestures and discourtesy), while research inthe UK found that their data was best described by three factors (Lajunen et al., 1998). To furtheradd to the confusion, Sullman (2006a) found that four factors best described the data he collectedfrom a sample of New Zealand drivers. Interestingly, one factor, Hostile Gestures, was found by allthree studies using the 33-item version of the DAS (Lajunen et al., 1998 labelled it ‘‘direct hostility’’).

There are a number of possible reasons for the differences in the factor structures found in thethree previous studies. Firstly, there was the fact that Deffenbacher et al. (1994) analysed theirdata using cluster analysis, while the other two studies used exploratory PCA. Furthermore, Laj-unen et al. (1998) dropped six items from their factor analysis as the situations without othermotorists involved (e.g. you encounter a detour) evoked very little anger amongst the British driv-ers. This resulted in the production of a 21-item UK version of the DAS, as a further six items didnot load on any of their three factors.

It also seems possible that cultural differences between USA, UK and NZ may have contributed tothe different factor structures. Unfortunately, the only peer reviewed study to use a version of theDAS to collect data from more than one country (Parker et al., 2002) used the 21-item British versionof the scale, meaning that the study cannot be used as evidence of the 33-item scale’s factor structure.However, although there is no directly relevant research on the cultural differences in driving angerbetween these countries, there is evidence that suggests general anger scales translate across cultureswith more divergent origins than those mentioned above (e.g. Krohne, Schmukle, Spaderna, &Spielberger, 2002; Maxwell, Sukhodolsky, Chow, & Wong, 2005). Therefore it appears that the dif-ferences in factor structure may have been caused primarily by dissimilar analytical approaches.

Although there has been disagreement on the scale’s factor structure there has been a relativelyhigh degree of agreement in the relationship between self reported driving anger and a number ofbackground variables. For example, research has found that females generally report higher levels

M.J.M. Sullman et al. / Personality and Individual Differences 42 (2007) 701–713 703

of driving anger and that the level of driving anger declines with age and experience (e.g. Parkeret al., 2002; Sullman, 2006a). Research has also found that drivers who reported preferring todrive faster also reported higher levels of driving anger (Sullman, 2006a).

This study investigated driving anger in a sample of Spanish drivers. In particular the researchinvestigated; the factor structure of the DAS using confirmatory factor analysis, which situationsangered Spanish drivers, along with the relationships driving anger had with a number of descrip-tive variables.

2. Method

2.1. Materials

Driver anger was measured using the 33-item version of the Driver Anger Scale (DAS) (Def-fenbacher et al., 1994). Translation of the scale was carried out by two native Spanish speakersand these translations were then translated back into English by a native English speaker. No ma-jor issues were identified during the translation process. Participants were instructed to imagineeach of the 33 potentially anger provoking situations happening to them and to rate the amountof anger elicited by each on a six point Likert scale (0 = Not at all to 5 = Very much).

2.2. Participants

One third of the employees (600) working full or part time at the University of Girona were se-lected. Systematic random sampling within strata was used for selection. Strata criteria were; typeof work (academic and non-academic) and faculty (Arts, Law, Education, Business, Engineering,Science, Nursing and Tourism). Random numbers between one and three were used to decide whereto begin the selection process, and then every third person was chosen in each stratified sample frame.

2.3. Procedure

Participants were firstly contacted by an email which explained the purpose of the study andasked them to participate. The questionnaires were then sent to participants via university mail.The majority of the questionnaires were collected in person by the researchers, but a small numberof the participants returned the questionnaires by university mail. Anonymity and confidentialityof the data were guaranteed.

3. Results

3.1. Demographic and descriptive variables

Of the 600 employees selected, 19 had no driving license, 21 had not driven in the last sixmonths (the two exclusion criteria) and 189 people did not respond. In total 371 (62%) employeesreturned completed questionnaires.

Table 1Descriptive variables

Variable Value

Males 47.2%Females 52.8%Annual mileage (range) 17226 km (10–200000)Mean age (range) 37.5 (22–75)Mean driving experience (range) 17.8 (1–50)Crashes in previous 5 years 53.0%Speed on divided highway (km/h) 128.2 (±11.9)Speed on motorway (km/h) 98.1 (±11.2)Speed on city roads (km/h) 47.0 (±10.8)Speed on residential roads (km/h) 43.8 (±14.3)Speed on winding mountain road (km/h) 56.3 (±19.5)


Table 1 shows the participants had a mean age of 37.5 (SD = 9.4), with 47% being male and53% female. The mean annual mileage was 17226 km/year (range 10–200000) and the partici-pants had an average of 17.8 years driving experience (range 1–50). In total 53% of the drivershad been involved in at least one crash in the previous five years, while 46% reported no crashesand 1% did not answer the question.

Participants were also asked to indicate their preferred driving speed on five different types ofroads (Table 1). As would be expected, the average speed for the divided highway was the highest,with a mean of 128.2 km/h (speed limit 120 km/h). This was followed by the non-divided motor-way where the average was 98.1 km/h (speed limit 80–100 km/h), while the lowest average(43.8 km/h) was for the residential area (speed limit normally 50 km/h). For the remaining anal-ysis the five speed items were combined to form a mean preferred driving speed (see Sullman,Meadows, & Pajo, 2002).

3.2. Driving anger

The means and standard deviations of the 33 Driving Anger Scale items are presented in Table 2.Unlike in the UK data (Lajunen et al., 1998), there were no items with means less than 1.5.

The items with the highest means were: ‘‘Someone cuts in right in front of you on the motor-way’’ (3.77 ± 0.99) and ‘‘Someone runs a red light or stop sign’’ (3.77 ± 1.07). This was followedclosely by ‘‘Someone cuts in and takes the parking spot you have been waiting for’’ (3.64 ± 1.08).

Using the six factors identified in the original DAS research (Deffenbacher et al., 1994), the sub-scale means found here were compared with the USA, UK and NZ findings (Table 3). T-tests wereperformed to compare New Zealand and Spanish drivers. This found that Spanish drivers re-ported higher levels of anger on the Police presence and Illegal driving factors and reported sig-nificantly lower levels on the Discourtesy and Slow driving categories of driving (all p < 0.001).

Table 3 also shows that the highest overall scale mean for the Spanish sample was the illegaldriving scale (3.46 ± 0.88). This level was significantly higher than that reported by New Zealanddrivers, and New Zealand drivers reported a significantly higher level than UK drivers. The meanfor the US data (2.7) was also considerably lower than that found here. Therefore, it seems that

Table 2DAS means and standard deviations

Item No. Item Mean SD

Discourtesy

7 Someone cuts in front of you on motorway 3.77 0.998 Someone cuts in and takes the parking spot you were waiting for 3.64 1.0814 Someone coming towards you does not dim headlights at night 3.47 0.9815 At night someone is driving right behind you with bright lights on 3.46 1.025 Someone driving very close to your rear bumper 3.43 1.0712 Someone backs out right in front of you 3.25 1.0120 Someone pulls out right in front of you when there is no-one behind you 3.15 1.0217 Someone speeds up when you try to pass them 3.13 1.0228 Cyclist in the middle of the lane 2.79 1.03

Average 3.36 0.76

Traffic obstructions

22 Hit unmarked deep pothole 3.32 1.0631 Truck kicking up gravel 3.17 1.1030 Behind a badly smoking vehicle 2.97 1.1019 Stuck in a traffic jam 2.67 1.0832 Behind a large truck you cannot see around 2.48 1.1026 Behind a truck with material flapping around 2.46 1.1233 Road works or detours 2.31 0.96

Average 2.77 0.74

Hostile gestures

21 Someone makes an obscene gesture towards you 2.81 1.2427 Someone shouts at you 2.68 1.1624 Someone beeps at you 2.55 1.08

Average 2.68 1.07

Slow driving

4 Someone driving too slowly in the outside lane 3.02 0.969 Someone driving more slowly than is reasonable 2.68 0.9110 Slow vehicle will not pull over 2.39 0.951 Someone does not move straight away when lights green 2.16 0.8218 Someone slow in parking 2.03 0.853 Pedestrian walks slowly across the street 1.64 0.73

Average 2.32 0.61

Police presence

23 Police car driving close by 2.41 1.1829 Police officer pulls you over 1.85 1.0011 Hidden police car watching traffic 1.82 1.0216 Pass a speed camera 1.80 0.97

Average 2.00 0.99

Illegal driving

13 Someone runs a red light/stop sign 3.77 1.0725 Someone driving well above speed limit 3.50 1.24

(continued on next page)


Table 2 (continued)

Item No. Item Mean SD

6 Someone weaving in and out of traffic 3.46 1.112 Someone driving too fast for the road conditions 3.12 1.12

Average 3.46 0.88

Table 3Means for USA (Deffenbacher et al., 1994), UK (Lajunen et al., 1998), NZ (Sullman, 2006a) and Spain

No. of Items USA UK NZ (SD) Spain (SD)

Discourtesy 9 3.9 2.7 3.51 (±0.77) 3.36 (±0.76)***

Traffic obstructions 7 3.3 2.0 2.69 (±0.82) 2.77 (±0.74)Hostile gestures 3 3.2 2.3 2.73 (±1.07) 2.68 (±1.07)Slow driving 6 3.2 2.0 2.77 (±0.72) 2.32 (±0.61)***

Police presence 4 3.0 1.4 1.86 (±0.86) 2.00 (±0.99)***

Illegal driving 4 2.7 2.3 3.28 (±0.97) 3.46 (±0.88)***

*** p < 0.001.


Spanish drivers report an unusually high level of anger evoked by illegal driving, when comparedto overseas drivers. The discourtesy scale produced the second highest mean (3.36 ± 0.76), whilethis scale recorded the highest means in the NZ, USA and UK samples (Deffenbacher et al., 1994;Lajunen et al., 1998; Sullman, 2006a). In the New Zealand and UK samples illegal driving had thesecond highest mean. As with the New Zealand and UK research, anger caused by police presenceproduced the lowest level of anger amongst the Spanish sample.

3.2.1. Factor analysisConfirmatory factor analysis (CFA) was used to investigate whether the factors identified by

Deffenbacher et al. (1994) were a good fit for the Spanish DAS data. LISREL 8.71 was usedand the fit was analysed using maximum likelihood estimation procedures. As the programmeused listwise deletion of missing data, the sample size was reduced from 371 to 327 participants.The processes began with the same six factor model produced by Deffenbacher et al. (1994). Table4 presents a summary of the goodness of fit indices. The goodness of fit indices used were; theSatorra-Bentler Scaled Chi-Squared statistic and the Non-Normed Fit Index (NNFI), due tothe non-normality of the data. The Standardised Root Mean Squared (SMRM) and RootMean-Squared Error of Approximation Measures Residuals (RMSEA) were also used.

Although in the first model the NNFI was acceptable, the SMRM and RMSEA were bothhigher than desired. As the goodness of fit indices for the first model were not as satisfactoryas desired (see Table 4) the modification indices were examined. This showed a number of prob-lem pairs of items (see Table 5). Redundant content appears to be the most likely explanation forthese issues, as the pairs of items involved were similar. For example, item 9 ‘‘Someone is drivingmore slowly than is reasonable for the traffic flow’’ and item 10 ‘‘A slow vehicle on a winding roadwill not pull over and let people pass’’ are very similar. Allowing these items to be correlated pro-

Table 5Error Correlations of pairs of items with similar contents within the same factor

Pair of items Error correlation

2 and 13 0.1154 and 18 0.1246 and 25 0.1739 and 10 0.13514 and 15 0.20819 and 26 0.16119 and 33 0.09922 and 32 0.20423 and 29 0.28226 and 31 0.14030 and 31 0.26732 and 33 0.143

Table 4Summary of goodness of fit indices for the CFA models

Model X2 (Satorra-Bentler) NNFI SMRM RMSEA C.I. 95%

First model fitted* 1207.9 0.957 0.0839 0.0682 0.063:0.073m = 480 p < 0.0005

Second model fitted** 849.6 0.977 0.0767 0.05 0.045:0.055m = 468 p = 0.954

* 6 factors with all the errors uncorrelated.** 6 factors with 12 errors covariance added.


duced a second model which had satisfactory fit indices (see Table 4). Only error correlations be-tween similar items within the same factor were allowed. The correlated version of the six factormodel was therefore chosen as the best one and is shown in Fig. 1.

Table 6 shows the factor loadings, found as a result of the CFA, along with the Construct Reli-ability Coefficients (Hair, Anderson, Tatham, & Black, 1999) for the second model. All factorshad good construct reliability (>0.7) and factor loadings were all above 0.5.

3.2.2. Anger by genderTable 7 shows the relative order of the six anger inducing factors was the same for both males

and females. However, females reported significantly more anger overall, along with anger pro-voked by traffic obstructions (p < 0.05), illegal driving (p < 0.001) and discourteous driving(p < 0.001).

3.2.3. Anger by age groupAs shown in Table 8, there were significant differences overall and in two of the driving anger

factors (discourtesy and police presence). However, only overall anger showed the same clear de-crease with age found amongst New Zealand drivers.

SLOWDRIVING

ILLEGALDRIVING

TRAFFICOBSTRUCTIONS

POLICEPRESENCE

DISCOURTESY

HOSTILEGESTURE

16 23 2911

e11 e16 e23 e29

3 4 9 101 18

e1 e18e3 e4 e9

25

2

6

13

e2

e6

e13

e25

21 24 27

e21 e24 e27

22 26 30 3119 32 33

e19 e22 e26 e30 e31 e32 e33

5

7

27

8

12

14

15

17

20

e5

e7

e8

e12

e14

e15

e16

e20

e27

e10

Fig. 1. Confirmatory factor analysis of the Driving Anger Scale.


3.2.4. Anger by descriptive variablesThe correlations between driving anger and the remaining descriptive variables were examined

by calculating Pearson’s Correlation Coefficients (Table 9). The overall level of anger wassignificantly lower for those participants who had held their drivers licenses longer (�0.136,p < 0.05). More experienced drivers were also less likely to be angered by discourteous drivingbehaviour (�.202, p < 0.001). Not unexpectedly, drivers who preferred to drive faster also re-ported higher levels of anger evoked by slow driving (.238, p < 0.001) and discourteous driving(.122, p < 0.05).

3.2.5. Anger by crash involvementT-tests were used to see whether crash involved drivers had higher means on the six anger fac-

tors and overall anger. Drivers with less than five years experience were excluded (n = 346). Themeans were higher in all cases, but only police presence (t = �3.266, p < 0.01), slow driving(t = �2.344, p < 0.05) and overall anger (t = �2.105, p < 0.05) were significant. Three individualhierarchical logistic regressions were used to test whether these factors were able to predict crashinvolvement over and above that of the background variables. In each regression age, gender, an-nual mileage and speed were entered in the first block, followed by the independent variable. This

Table 6Factor loadings from the confirmatory factor analysis

Item No. Item Factor loading

Factor 1 – Traffic obstructions construct reliability coefficient = 0.84

32 Behind a large truck you cannot see around .78231 Truck kicking up gravel .71833 Road works or detours .68030 Behind a badly smoking vehicle .63726 Behind a truck with material flapping around .59919 Stuck in a traffic jam .59022 Hit unmarked deep pothole .568

Factor 2 – Illegal driving construct reliability coefficient = 0.84

25 Someone driving well above the speed limit .8822 Someone driving too fast for the road conditions .7906 Someone weaving in and out of traffic .65913 Someone runs a red light/stop sign .646

Factor 3 – Slow driving construct reliability coefficient = 0.83

18 Someone slow in parking .8184 Someone driving too slowly in the outside lane .6931 Someone does not move straight away when the lights green .65010 Slow vehicle will not pull over .6299 Someone driving more slowly than is reasonable .6203 Pedestrian walks slowly across the street .571

Factor 4 – Hostile gestures construct reliability coefficient = 0.94

27 Someone shouts at you .95024 Someone beeps at you .90821 Someone makes an obscene gesture towards you .898

Factor 5 – Discourtesy construct reliability coefficient = 0.85

17 Someone speeds up when you try to pass them .70520 Someone pulls out right in front of you when there is no-one behind you .67628 Cyclist in the middle of the lane .67415 At night someone is driving right behind you with bright lights on .63312 Someone backs out right in front of you .62414 Someone coming towards you does not dim their headlights at night .6105 Someone driving very close to your rear bumper .5867 Someone cuts in front of you on motorway .5568 Someone takes the parking spot you were waiting for .538

Factor 6 – Police presence construct reliability coefficient = 0.84

16 Pass a speed camera .86711 Police car watching traffic from a hidden position .74829 Police officer pulls you over .73823 Police driving close by .648


found that even after the contribution of the background variables had been partialled out, allthree were still predictive of crash involvement (p < 0.05). In the interests of parsimony onlythe regression of overall anger is shown (Table 10).

Table 7Anger by Gender

Males mean (SD) Females mean (SD) t P

Discourtesy 3.23 (0.64) 3.46 (0.68) �3.28 0.001

n = 165 n = 184Traffic obstructions 2.69 (0.68) 2.84 (0.79) �1.97 0.05

n = 172 n = 190Hostile gestures 2.60 (1.02) 2.76 (1.12) �1.47 0.14

n = 170 n = 192Slow driving 2.30 (0.59) 2.34 (0.62) �0.74 0.46

n = 168 n = 192Police presence 1.96 (0.79) 1.97 (0.83) �0.18 0.86

n = 164 n = 184Illegal driving 3.30 (0.87) 3.62 (0.85) �3.56 0.001

n = 171 n = 191Overall anger 2.74 (0.51) 2.90 (0.59) �2.86 0.01

n = 156 n = 171

Table 8Anger by age

<30 mean (SD) 30–44 mean (SD) 45–59 mean (SD) 60+ mean (SD) P

Discourtesy 3.49 (0.65) 3.40 (0.64) 3.16 (0.68) 3.19 (0.61) <0.05

n = 68 n = 198 n = 69 n = 10Traffic obstructions 2.94 (0.74) 2.74 (0.73) 2.69 (0.70) 2.81 (0.80) ns

n = 73 n = 204 n = 69 n = 12Hostile gestures 2.97 (1.04) 2.64 (1.10) 2.59 (1.04) 2.42 (0.75) ns

n = 73 n = 205 n = 69 n = 11Slow driving 2.37 (0.64) 2.35 (0.62) 2.24 (0.53) 2.24 (0.49) ns

n = 71 n = 206 n = 68 n = 11Police presence 2.27 (0.91) 1.87 (0.76) 1.93 (0.78) 2.00 (0.62) <0.01

n = 71 n = 197 n = 65 n = 11Illegal driving 3.41 (0.87) 3.53 (0.84) 3.45 (0.92) 3.02 (1.14) ns

n = 72 n = 205 n = 70 n = 11Overall anger 3.00 (0.54) 2.84 (0.55) 2.71 (0.57) 2.60 (0.46) <0.05

n = 66 n = 186 n = 62 n = 9


4. Discussion

This research found that confirmatory factor analysis of the Driving Anger Scale using theSpanish data supported the original six factor solution found in the American sample (Deffenb-acher et al., 1994). However, correlations between some pairs of items suggested that some of theitems were redundant. Future research should review these items and test whether the fit of the sixfactor model could be improved by removing one of each pair, thereby also reducing the size ofthe scale.

One obvious reason for the large difference between the factor structures is methodological dif-ferences, with both Lajunen et al. (1998) and Sullman (2006a) using exploratory factor analysis

Table 9Correlation coefficients for driver anger and descriptive variables

Mileage Experience Speed

Discourtesy �.021 �.202*** .122*

Traffic obstructions �.034 �.057 .072Hostile gestures �.051 �.097 �.017Slow driving .061 �.085 .238***

Police presence .039 �.027 .080Illegal driving �.039 �.052 �.110Overall anger �.018 �.136* .072

* p < 0.05.** p < 0.01.*** p < 0.001.

Table 10Prediction of crash involvement

Block Model chi-square improvement %Correctly classified B Wald Exp (B)

1 Gender 4.141 53.9 �.082 .086 0.0861 Mileage .000 .114 1.001 Age �.002 .012 .9141 Speed .091 3.719 1.102 Overall anger 4.617* 60.4 .551 4.495* 1.74

* p < 0.05.


and choosing the number of factors to keep based upon the results of parallel analysis. There wasalso the fact that Lajunen et al. (1998) excluded six items from their factor analysis. In this re-search confirmatory factor analysis was undertaken on the six factors Deffenbacher et al.(1994) found, which proved to be a good fit, after some minor changes. The hostile gestures factoris the most reliable factor, as it was found in all four samples (Deffenbacher et al., 1994; Lajunenet al., 1998; Sullman, 2006a).

Interestingly, the items which did not cause anger amongst UK drivers (means were <1.5) didcause anger amongst Spanish drivers, as they did amongst American (Deffenbacher et al., 1994)and NZ drivers (Sullman, 2006a). Thus, the similarities Parker et al. (2002) found in a cross-cul-tural comparison of England, Finland and the Netherlands may be limited, as a large number ofthe original DAS items (12) were not included in their instrument. Therefore, it would be inter-esting to investigate whether the items excluded from the original scale would have caused angerto be reported amongst the Finnish and Dutch drivers, as they did in the Spanish, American, andNew Zealand samples.

The level of driver anger reported here was mostly around the same level as reported in NewZealand, with two subscales being higher, two lower and no difference being found betweentwo. These results also appeared to be higher than Lajunen et al.’s (1998) research, but weremostly lower than that found amongst American drivers (Deffenbacher et al., 1994). Unfortu-nately it was not possible to test whether the means found here were different from the US orUK means, as neither study reported standard deviations. Interestingly, Spanish drivers reported


an exceptionally high level of anger evoked by illegal driving. This appears to be at odds with re-search which has shown Spanish drivers report a relatively high level of illegal driving behaviours(Gras et al., 2006), as we would expect those who engage more frequently in illegal driving behav-iours to be the least angered by these situations. However, the difference may be partly due to theconstrained age range in the US study, which studied first year university students. Support forthis argument can be drawn from the fact that anger at illegal driving was highest in the ‘‘middle’’age groups here and also in a recent New Zealand study (Sullman, 2006b).

Female drivers reported significantly more anger overall and in three of the six categories ofdriving anger (illegal driving, discourtesy and traffic obstructions). These findings are in line withprevious research, which has also found that females report higher overall levels of anger (Laj-unen & Parker, 2001; Sullman, 2006a) and on some of the types of driving anger (Sullman,2006a; Parker et al., 2002).

In terms of age, there was a similar pattern in the ordering of the six categories of anger acrossthe four age groups. However, reported anger did not clearly decline with age, with the oldest agegroup (60+) reporting the second highest level of anger on two factors. However, the overall levelof anger did show a significant decline with age, which supports previous research (Lajunen et al.,1998; Sullman, 2006a).

This research also found drivers who had been crash involved had significantly higher means onoverall anger and on the police presence and slow driving factors. This pattern remained evenafter the effects of the background variables had been partialled out. This may mean that angrydrivers (and in particular those angered by police presence and slow driving) are more likely tocrash. Although this is in contrast to Deffenbacher et al.’s (2001) results, some support can bedrawn from simulator research and research which has found angry drivers report engaging moreoften in risky and dangerous driving behaviours (Dahlen et al., 2005; Deffenbacher et al., 1994,2001, 2002). An alternative explanation would be that those who had been crash involved weremore angered by police presence and slow driving. However, although the former explanationseems to have more intuitive appeal, future research should attempt to clarify this relationship.

The study reported here clearly had a number of methodological limitations. There is obviouslythe possibility of sampling bias. As the participants all worked at the University of Girona, it ispossible that they differ significantly from the general population in some way. However, confi-dence in this data can be drawn from the fact that the gender distribution (47.2% male) andage range (22–75 years) in the present study were similar to that of the population of Spanish driv-ers (48.9% male and age range 18–74+; see Gras et al., 2006). Further confidence can be drawnfrom the fact that a large number of the results were similar to those found by overseasresearchers.

References

Dahlen, E. R., Martin, R. C., Ragan, K., & Kuhlman, M. M. (2005). Driving anger, sensation seeking, impulsiveness,and boredom proneness in the prediction of unsafe driving. Accident Analysis and Prevention, 37, 341–348.

Deffenbacher, J. L., Deffenbacher, D. M., Lynch, R. S., & Richards, T. L. (2003). Anger, aggression, and riskybehaviour: a comparison of high and low anger drivers. Behaviour Research and Therapy, 41, 701–718.

Deffenbacher, J. L., Lynch, R. S., & Oetting, E. R. (2002). The Driving Anger Expression Inventory: a measure of howpeople express their anger on the road. Behaviour Research and Therapy, 40, 717–737.


Deffenbacher, J. L., Lynch, R. S., Oetting, E. R., & Yingling, D. A. (2001). Driving anger: correlates of a test of state-trait theory. Personality and Individual Differences, 31, 1321–1331.

Deffenbacher, J. L., Oetting, E. R., & Lynch, R. S. (1994). Development of a driver anger scale. Psychological Reports,

74, 83–91.Gras, M. E., Sullman, M. J. M., Cunill, M., Planes, M., Aymerich, M., & Font-Mayolas, S. (2006). Spanish drivers and

their aberrant driving behaviours. Transportation Research Part F, 9, 129–137.Hair, J. F., Anderson, R. E., Tatham, R. L., & Black, W. C. (1999). Analisis Multivariante. Madrid: Prentice-Hall.Krohne, H. W., Schmukle, S. C., Spaderna, H., & Spielberger, C. D. (2002). The state-trait depression scales: an

international comparison. Anxiety, Stress and Coping: An International Journal, 15, 105–122.Lajunen, T., & Parker, D. (2001). Are aggressive people aggressive drivers? A study of the relationship between self-

reported general aggressiveness, driver anger and aggressive driving. Accident Analysis and Prevention, 33, 243–255.Lajunen, T., Parker, D., & Stradling, S. G. (1998). Dimensions of driver anger, aggressive and highway code violations

and their mediation by safety orientation in UK drivers. Transportation Research Part F, 1, 107–121.Maxwell, J. P., Sukhodolsky, D. G., Chow, C. C. F., & Wong, C. F. C. (2005). Anger rumination in Hong Kong and

great Britain: validation of the scale and a cross-cultural comparison. Personality and Individual Differences, 39,1147–1157.

Parker, D., Lajunen, T., & Summala, H. (2002). Anger and aggression among drivers in three European countries.Accident Analysis and Prevention, 34, 229–235.

Smart, R. G., Mann, R. E., & Zhao, J. (2005). Is road rage increasing? Results of a repeated survey. Journal of Safety

Research, 36, 195–201.Sullman, M. J. M. (2006a). Driving anger amongst New Zealand drivers. Transportation Research Part F, 9, 173–184.Sullman, M. J. M. (2006b, July). Driving anger amongst New Zealand drivers: Is it increasing? In: Oral presentation at

the 26th International Congress of Applied Psychology, Athens, Greece.Sullman, M. J. M., Meadows, M., & Pajo, K. B. (2002). Aberrant driving behaviour amongst New Zealand truck

drivers. Transportation Research Part F, 5, 217–232.Underwood, G., Chapman, P., Wright, S., & Crundall, D. (1999). Anger while driving. Transportation Research Part F,

2, 55–68.

driving anger in spain

Documents