---
category: literaturenote
citekey: ozkannewadditiondbqpositive2005
title: "A new addition to DBQ: Positive Driver Behaviours Scale"
authors: "Özkan, Türker; Lajunen, Timo"
year: 2005
date: 2005-00-00 2005
doi: 10.1016/j.trf.2005.04.018
publication: "Transportation Research Part F: Traffic Psychology and Behaviour"
url: "https://doi.org/10.1016/j.trf.2005.04.018"
zotero_key: WPSUS7ES
zotero_storage: U55Q6C86
collections: transpordi_planeerimine
folder: 001_artiklid
firstAuthor: "Özkan, Türker"
status: converted
---
![](_page_0_Picture_0.jpeg)

![](_page_0_Picture_2.jpeg)

Transportation Research Part F 8 (2005) 355–368

www.elsevier.com/locate/trf

# A new addition to DBQ: Positive Driver Behaviours Scale

Tu¨rker O¨ zkan a,b, Timo Lajunen b,\*

<sup>a</sup> Department of Psychology, Traffic Research Unit, P.O. Box 9, University of Helsinki, 00014 Helsinki, Finland <sup>b</sup> Department of Psychology, Middle East Technical University (ODTU¨ ), 06531 Ankara, Turkey

Received 11 October 2004; received in revised form 28 April 2005; accepted 30 April 2005

#### Abstract

The aims of the present study were to develop an instrument for measuring ''positive'' driver behaviours and to investigate the relationship between these behaviours, DBQ scales (violations and errors), aggression, traffic offences, and accidents. Participants were 306 Turkish drivers (194 male and 112 female) who completed a questionnaire including the newly developed Positive Driver Behaviour Scale, Driver Behaviour Questionnaire (DBQ), Driver Aggression Indicators Scale (DAIS), and items related to drivers driving records and demographics. Factor analysis resulted in a clear three-factor structure (violations, positive driver behaviours, and errors), high item loadings, and acceptable internal consistency coefficients. Negative binomial regression analyses showed that violations were related to the number of accidents and penalties. Multiple regression analyses indicated that violations were also associated with all other dependent variables except hostile aggression and revenge committed by the respondent, measured by the DAIS ''self'' scale. Errors were related to hostile aggression and revenge committed by the respondent and other drivers, measured by the DAIS ''other drivers'' scale, and aggressive warnings measured by the DAIS ''Self'' scale. Positive driver behaviours were negatively related to hostile aggression and revenge measured by the DAIS ''Self'' and ''Other drivers'' scales, but not to traffic offences or accidents. 2005 Elsevier Ltd. All rights reserved.

Keywords: Driver Behaviour Questionnaire (DBQ); Aggression; Positive driver behaviours; Traffic offences and accidents

<sup>\*</sup> Corresponding author. Tel.: +90 312 2105113; fax: +90 312 2101288. E-mail address: [timo@metu.edu.tr](mailto:timo@metu.edu.tr) (T. Lajunen).

## 1. Introduction

Driving style refers to the ways drivers choose to drive or habitually drive, including, for example, the choice of driving speed, habitual level of general attentiveness, and gap acceptance [\(Elander, West, & French, 1993](#page-12-0)). Because of the evident relevance of driving style to accident risk, dozens of self-report instruments have been developed for measuring driving style. While most of the scales focus on self-reported behaviours (e.g., Driver Behaviour Questionnaire by [Reason,](#page-13-0) [Manstead, Stradling, Baxter, & Campbell, 1990](#page-13-0)), some of the scales measure other aspects of driving like a drivers decision making (e.g., Driving Style Questionnaire by [French, West, Elander, &](#page-12-0) [Wilding, 1993](#page-12-0)) or driver stress (Driver Behaviour Inventory by [Glendon et al., 1993; Gulian,](#page-12-0) [Matthews, Glendon, Davies, & Debney, 1989](#page-12-0)). Although all these ''driving style'' scales show high inter-correlations ([Westerman & Haigney, 2000](#page-13-0)) and a similar pattern in predicting different types of accidents [\(Parker, West, Stradling, & Manstead, 1995\)](#page-13-0), these measures are in fact, conceptually fairly distinct.

The Driver Behaviour Questionnaire (DBQ) ([Reason et al., 1990](#page-13-0)) seemed to be a possible turning point in constructing a comprehensive model for everyday driving behaviour ([Ranney,](#page-13-0) [1994](#page-13-0)). In the DBQ, driver behaviours could be categorized and evaluated under a theoretically sound framework, which also seemed to have practical importance. The DBQ was based on the main distinction between errors and violations, which were assumed to have different psychological origins and demand different modes of remediation ([Reason et al., 1990\)](#page-13-0). Errors were taken as a ''generic term to encompass all those occasions in which a planned sequence of mental or physical activities fails to achieve its intended outcome, and when these failures cannot be attributed to the intervention of some chance agency'' [\(Reason, 1990, p. 9](#page-13-0)). Errors were differentiated into slips and lapses (resulting from action), and mistakes (errors of intention). While lapses largely involve failures of memory, slips are related to attention deficits. Mistakes were further divided into two subcategories: rule-based mistakes and knowledge-based mistakes [\(Reason, 1999](#page-13-0)). In fact, Rasmussens ''skill-rule-knowledge'' taxonomy of human performance levels was the base for this further subdivision (for comprehensive review, see [Rasmussen,](#page-13-0) [1980](#page-13-0)). Violations can be located in the ''rule-based mistakes'' category and are closely associated with the misapplication of normally good rules, or the application of bad rules, or a failure to apply a good rule, or erroneous performance in a no rules situation. Contextual and motivational demands influence violations rather than slips and lapses. Knowledge-based mistakes (i.e., errors) emerge when pre-existing solutions do not work and a trial-and-error learning process is needed for finding new feasible solutions (see [Reason, 1999\)](#page-13-0). On the other hands, it is still possible that a driver may commit violations without errors or may commit errors without violations.

[Reason et al. \(1990\)](#page-13-0) proved the theoretical taxonomy in their first study using the DBQ. They found that driver errors and violations are two empirically distinct classes of behaviour containing three factors (deliberate violations, dangerous errors, and ''silly'' errors). Later studies have shown that the main distinction between errors (slips, lapses, and errors) and violations seems to occur in different populations in the UK ([Lawton, Parker, & Stradling, 1997; Parker,](#page-12-0) [Manstead, Stradling, & Reason, 1992; Reason et al., 1990\)](#page-12-0) as well as in other countries (in Australia by [Blockley & Hartley, 1995](#page-12-0); in Brazil by [Bianchi & Summala, 2002](#page-12-0); in China by [Xie & Parker, 2002](#page-13-0); in Greece by [Kontogiannis, Kossiavelou, & Marmaras, 2002;](#page-12-0) in Finland and Netherlands by [Lajunen, Parker, & Summala, 1999;](#page-12-0) in New Zealand by [Sullman, Meadows,](#page-13-0) [& Pajo, 2002](#page-13-0); in Sweden by A˚ [berg & Rimmo¨, 1998;](#page-12-0) and in Turkey by [Lajunen & O¨](#page-12-0) zkan, 2004). In a recent study by [Lajunen, Parker, and Summala \(2004\),](#page-12-0) the DBQ factor structure was studied among British, Dutch and Finnish drivers. The results of this study supported the idea of two second-order factors that could be named as errors and violations although some instability was found in the first-order factor structures reported in various studies.

While the DBQ is focused on aberrant driver behaviours because of their relevance to traffic safety, [Ben-Ari, Mikulincer, and Gillath \(2003\)](#page-12-0) aimed at developing a multi-dimensional omnibus measure of driver behaviour. They hypothesized that most of the driving-specific factors can be grouped into four broad facets: (a) a reckless and careless driving style, (b) an anxious driving style, (c) an angry and hostile driving style, and (d) a patient and careful driving style. The patient and careful driving style was described as ''a well-adjusted driving style that has received less attention in previous studies''. In their study, [Ben-Ari et al. \(2003\)](#page-12-0) obtained eight main factors representing one of the each driving style facets. Items included in these factors were classified as either describing ''maladaptive'' or ''adaptive'' behaviours, attitudes and emotions. It should be noted, however, that some types of violations (i.e., ''correct violations'' defined as correct performance achieved by deviating from inappropriate rules or procedures, see [Reason, 1999](#page-13-0)) may actually be highly adaptive but are still violations, whereas strict obedience to the rule might actually be maladaptive in some situations. In addition to this conceptual difference, [Ben-Ari et al.](#page-12-0)s [\(2003\)](#page-12-0) new scale differed from the DBQ in how the measurement was conducted. While the DBQ is strictly a behavioural scale, with respondents indicating how often they commit behaviours described (from ''never'' to ''all the time''), [Ben-Ari et al.](#page-12-0)s (2003) scale measured behaviours, attitudes and emotions by asking drivers to indicate how well each item ''fits to their feelings, thoughts and behaviours'' (from ''not at all'' to ''very much''). The present study attempts to extend the DBQ towards an omnibus scale while preserving the theoretical taxonomy of the DBQ and its characteristic as a behavioural scale.

In the DBQ, intention is the key factor for classifying the aberrant driver behaviours into errors (no intention to make an error) and violations (intention to violate). Furthermore, errors were divided to slips and lapses, and mistakes according to the potential consequence of the error. The common feature for both violations and errors were, however, that they were both labelled as aberrant, and therefore negative, behaviours. Although focusing on negative behaviours is well justified in terms of traffic safety interventions, everyday driving involves other behaviours that cannot be classified as errors or violations. These behaviours do not have to be based on coded rules and regulations, nor primarily take safety into account. The main intention in these behaviours is to take care of the traffic environment or other road users, to help and to be polite with or without safety concerns. Since intention is the main predictor of our behaviours [\(Fishbein &](#page-12-0) [Ajzen, 1975\)](#page-12-0), we can suppose that not only targeting driver behaviours but also focusing on ''what to intend to do in traffic'' (positive or negative) could be proactive and helpful intervention for improving traffic safety.

These ''positive'' behaviours can be committed without a violation or an error if the action and the plan were adequate for reaching their desired end. In contrast, ''positive'' behaviours may sometimes include errors or violations when the aforementioned requirements are not fulfilled. For instance, a driver may cross the barrier line in order to avoid a puddle that might splash water on pedestrians. In the worst case, this basically positive but unexpected action may result in a small accident despite the good intention. Since every day driving is full of such behaviours in which the main intention is not to violate rules but facilitate smooth driving, the first aim of this study was to develop a reliable scale for measuring ''positive'' driver behaviours. The second aim was to investigate the relationships between positive driver behaviours, violations, errors, aggression, traffic offences, and accidents.

## 2. Method

## 2.1. Participants

The data reported in the present study were collected from 312 Turkish drivers in Ankara. All participants had driving licenses. Students of the traffic psychology course at the Middle East Technical University collected the data among drivers in Ankara as part of their course work. The participants were assured of anonymity and confidentiality. The participants filled out the DBQ, ''Positive Driver Behaviours Scale'', Driver Aggression Indicators Scale (DAIS), and items related to drivers driving records and demographic variables. Six drivers were excluded because they were either multivariate or univariate outliers on the major variables (p < .001) leaving 306 drivers (194 males and 112 females) for the main analyses.

Participants were aged from 18 to 67 years (mean 28.99, SD 11.44), with a range of 0–34 years of driving experience (mean 7.62, SD 8.36), and lifetime mileage ranging from 0 to 1,320,000 km (mean 86,034, SD 203,184). Participants lifetime accident involvement ranged from 0 to 14 (mean 2.08, SD 2.46), and their lifetime offences ranged from 0 to 60 (mean 2.37, SD 6.28).

## 2.2. Measures

## 2.2.1. Aberrant driver behaviours

DBQ with extended violations was used to measure violations and errors [\(Reason et al., 1990](#page-13-0)). The Turkish translation and the factor structure of the DBQ have been validated in studies conducted among both professional ([Su¨mer & O¨](#page-13-0) zkan, 2002) and non-professional drivers ([Lajunen &](#page-12-0) O¨ [zkan, 2004](#page-12-0)). The extended version of DBQ includes two types of violations; aggressive violations (4 items, e.g., ''to sound your horn to indicate your annoyance'') and ordinary violations (8 items, e.g., ''pull out of junction so far that the driver with right of way has to stop and let you pass''), and errors (8 items, e.g., ''misjudging the speed of another vehicle when overtaking''). Four items were added to measure aggressive violations. Two of those items (21 and 23) were taken from Krahe´ [and Fenske \(2002\)](#page-12-0) and two of them (22 and 24) were taken from [Macmillan](#page-13-0) [\(1975\).](#page-13-0) These items were reworded to fit the scale format. Participants were asked to indicate how often they committed each of the 24 behaviours in the previous year on a six-point scale (1 = never and 6 = very often). The means and standard deviations of subscales for male and female drivers are shown in [Table 1](#page-4-0).

## 2.2.2. Positive driver behaviours scale

In the DBQ, violations are behaviours that are deliberately done and deviations from the normal practices, which seem to be necessary for maintaining the safe operation of a potentially

<span id="page-4-0"></span>Table 1 Means and standard deviations for DBQ subscales, positive driver behaviours and DAIS subscales for female and male drivers

| Variables                              | Females |      | Males |      |
|----------------------------------------|---------|------|-------|------|
|                                        | Mean    | SD   | Mean  | SD   |
| Violations                             | 1.74    | 0.59 | 2.13  | 0.78 |
| Errors                                 | 1.61    | 0.56 | 1.60  | 0.49 |
| Positive driver behaviours             | 4.37    | 0.88 | 4.37  | 0.85 |
| Hostile aggression and revenge (self)  | 1.53    | 0.83 | 1.53  | 0.71 |
| Aggressive warning (self)              | 2.78    | 0.89 | 2.83  | 0.83 |
| Hostile aggression and revenge (other) | 1.19    | 0.51 | 1.31  | 0.56 |
| Aggressive warning (other)             | 2.14    | 0.76 | 2.43  | 0.79 |

hazardous system. On the other hand, errors are unintentionally behaviours and closely related to the deficiencies of cognitive capacity or psychomotor skills. As observed in traffic, there are some behaviours, which are deliberately done but not associated with any written traffic code or safety practices. The focus of these behaviours is on ''taking care of smooth traffic flow or paying attention to other road users'' or simply ''politeness''. In the present study, a new scale was developed to measure these behaviours. To ensure content validity of the instrument, an extensive literature review was conducted and instruments measuring driver behaviours were collected. However, no study tapping exactly ''positive'' driver behaviours was found. Then, researchers interviewed with a group of drivers (n = 5) to collect ''positive'' driving episodes in traffic. An item pool was developed and the items were screened for redundancy and clarity. A jury of drivers (N = 3) reviewed the set of items by rating the applicability and the prevalence of the behaviours described in the items. The final instrument included 14 items. The participants were asked to indicate on a six-point scale (1 = never and 6 = very often) how often they committed each of the 14 behaviours.

## 2.2.3. Driver aggression indicators scale (DAIS)

The DAIS was developed by T.J. Lajunen (personal communication, 30 August 2004) and D. Parker and used in [Ra¨met \(2003\)](#page-13-0). The DAIS was used for the first time in a Turkish sample in a study by [Su¨mer, Lajunen, and O¨](#page-13-0) zkan (2003). The factor structure of the DAIS has been validated in a study by [Lajunen and O¨](#page-12-0) zkan (2004), whose translation was used in the present study. The DAIS consists of 13 items and assesses drivers possible aggressive behaviours on the road. Drivers rate these items twice (once for themselves and once for other drivers). Drivers were asked to indicate how often they have done each of the given behaviours/episodes on the list (e.g., ''verbally threatening other drivers'', ''showing the anger by honking'') and how often another driver has done those behaviours to them using a five-point scale from ''Never'' to ''Nearly all the time''.

## 2.2.4. Demographic variables

Participants were asked to indicate their age, sex, number of lifetime active and passive accidents and offences, years of driving experience and lifetime mileage.

## 3. Results

## 3.1. Factor structure of driver behaviours

Factor analyses were administered on 38 items. Items of the DBQ and ''positive'' driver behaviours were subjected to principal axis factor analysis with direct oblimin rotation. The criteria used to determine the number of factors were the Kaiser criterion of eigenvalues over 1.0, the Cattell scree plot test, parallel analysis, and the interpretability of factors. Initially, nine factors had eigenvalues over 1.0. However, the scree plot and parallel analysis suggested the three-factor solution to be the most interpretable one. Two original DBQ items (4 and 5), as well as one positive driver behaviour item (29) did not load on any factors. A DBQ item, ''shooting lights'' loaded both on violation and error factors. Also, a positive driver behaviour item, ''thank another driver for helping me by waving my hand'' loaded both on violations and positive driver behaviours. These ambiguous items were not used in further analyses. The first factor included 13 items that measured violations, accounting for 20% of the variance. The second factor included 13 items that tapped positive driver behaviours and accounted for 9% of the variance. This factor was named as the ''Positive Driver Behaviours'' scale. The third factor included eight items that measured errors, accounting for 5% of the variance. Alpha reliabilities for those subscales were 0.86, 0.84, and 0.79, respectively. The results indicated that the DBQ with positive driver behaviours items had a clear three-factor structure, high item loadings, and acceptable internal consistency coefficients [\(Table 2\)](#page-6-0).

## 3.2. Factor structure of DAIS

A principle axis factoring analysis with oblimin rotation on the 13 items of the DAIS yielded two interpretable factors for both ''self'' (a scale measuring how often a respondent committed behaviours described in items) and ''other drivers'' (a scale measuring how often a respondent was treated by other drivers as described). The first component included six items describing a drivers very hostile and direct aggressive actions. This component was labelled as ''hostile aggression and revenge'' (e.g., ''physical attack''). The second component consisted of seven items reflecting mostly aggressive warnings on the road, and thus, was labelled ''aggressive warnings'' (e.g., ''sounding horn''). For the ''self'' part of the scale, these factors together explained 54% of the total variance (45% for ''hostile aggression and revenge'' and 9% for ''aggressive warnings''). Alpha reliabilities for those subscales were 0.89 and 0.84, respectively. Similarly, the ''hostile aggression and revenge'' factor included six items that explained 36% of the variance for the ''other'' part of the scale. The ''aggressive warnings'' factor included seven items that explained 9% of the variance. The result of factor analyses showed that the previous factor structure of DAIS ([Lajunen & O¨](#page-12-0) zkan, 2004) was replicated. In the present study, item 12 loaded both on hostile aggression and revenge and aggressive warnings factors in both ''self'' and ''other'' part of the scale. Also, item 11 loaded both on hostile aggression and revenge and aggressive warnings factors in only ''other'' part of the scale. These ambiguous items were not used in further analyses. For the ''other'' part of the scale, alpha reliabilities for hostile aggression and revenge and for aggressive warnings subscales were 0.86 and 0.75, respectively. The results indicated that the DAIS had a clear two-factor structure, high item loadings, and acceptable internal consistency coefficients (see [Table 3\)](#page-7-0).

<span id="page-6-0"></span>Table 2 Means and standard deviations for DBQ and positive behaviour items and the three-factor solution

| (Mean/SD)           | Violations | Positive<br>behaviours | Errors | Items (number)                                                                                     |
|---------------------|------------|------------------------|--------|----------------------------------------------------------------------------------------------------|
| (2.37/1.28)         | 0.74       |                        |        | Overtake a slow driver on the inside (14)                                                          |
| (2.48/1.31)         | 0.71       |                        |        | Disregard the speed limit on a residential road (8)                                                |
| (2.40/1.45)         | 0.71       |                        |        | Using long lights to disturb oncoming car (24)                                                     |
| (2.05/1.12)         | 0.65       |                        |        | Aversion, indicate hostility (18)                                                                  |
| (1.82/1.22)         | 0.64       |                        |        | Race from lights (15)                                                                              |
| (2.28/1.46)         | 0.62       |                        |        | Let driver coming with long lights behind pass and<br>then let him/her taste the same feeling (22) |
| (2.35/1.44)         | 0.57       |                        |        | Disregard the speed limit on motorway (20)                                                         |
| (1.80/0.96)         | 0.48       |                        |        | Drink and drive (1)                                                                                |
| (1.47/0.83)         | 0.48       |                        |        | Close following (16)                                                                               |
| (1.57/0.99)         | 0.47       |                        |        | Accelerating your speed while driver overtaking you (23)                                           |
| (1.50/0.92)         | 0.46       |                        |        | Get angry, give chase (12)                                                                         |
| (1.71/1.00)         | 0.45       |                        |        | Push in at last minute (13)                                                                        |
| (1.96/1.17)         | 0.42       |                        |        | Reducing your speed to indicate your annoyance (21)                                                |
| (4.76/1.33)         |            | 0.71                   |        | Avoid close following not to disturb                                                               |
|                     |            |                        |        | the car driver in front (31)                                                                       |
| (4.82/1.42)         |            | 0.68                   |        | Less frequently use long lights to help                                                            |
|                     |            |                        |        | the oncoming driver (27)                                                                           |
| (5.12/1.19)         |            | 0.61                   |        | Parking car by taking into other road                                                              |
|                     |            |                        |        | users free movement (36)                                                                           |
| (4.58/1.48)         |            | 0.61                   |        | Pay attention to puddle not to splash                                                              |
|                     |            |                        |        | water on pedestrians or other road users (38)                                                      |
| (4.79/1.21)         |            | 0.61                   |        | Arrange my speed to help the driver trying to overtake (32)                                        |
| (4.34/1.56)         |            | 0.60                   |        | No sounding horn to avoid noise (28)                                                               |
| (4.20/1.52)         |            | 0.60                   |        | Return to your place not to block coming car behind (33)                                           |
| (4.76/1.34)         |            | 0.55                   |        | Avoid using the left lane to facilitate                                                            |
|                     |            |                        |        | the speed of traffic flow (30)                                                                     |
| (3.77/1.46)         |            | 0.48                   |        | Let pedestrians cross even it is my right to pass (35)                                             |
| (4.43/1.61)         | 0.36       | 0.48                   |        | Thank the driver helping me by waving my hand, etc. (34)                                           |
| (4.52/1.68)         |            | 0.45                   |        | Do my best not to be obstacle for other drivers (25)                                               |
| (4.08/1.62)         |            | 0.36                   |        | No sounding horn to disturb the driver in front                                                    |
|                     |            |                        |        | waiting even after green light (37)                                                                |
| (2.66/1.13)         |            | 0.34                   |        | Give my right of way to other drivers (26)                                                         |
| (1.38/0.71)         |            |                        | 0.67   | Turning right nearly hit cyclist (9)                                                               |
| (1.63/0.78)         |            |                        | 0.65   | Fail to see pedestrians crossing (3)                                                               |
| (1.75/0.90)         |            |                        | 0.56   | Underestimate the speed of an oncoming vehicle (19)                                                |
| (1.60/0.89)         |            |                        | 0.55   | Pull out, force your way out (7)                                                                   |
| (1.40/0.68)         |            |                        | 0.51   | Attempt to overtake someone turning left (11)                                                      |
| (1.48/0.72)         |            |                        | 0.50   | Miss ''Give Way'' signs (10)                                                                       |
| (1.81/0.90)         |            |                        | 0.43   | Brake too quickly on a slippery road (6)                                                           |
| (1.69/0.82)         |            |                        | 0.40   | Queuing, nearly hit car in front (2)                                                               |
| (1.70/0.90)         | 0.35       |                        | 0.36   | Shooting lights (17)<br>Sound horn to indicate your annoyance (4)                                  |
| Initial eigenvalues | 8.15       | 4.08                   | 2.63   | Fail to check your rear-view mirror (5)                                                            |
| Variance explained  | 19.83      | 9.13                   | 5.26   | Using indicator to help the driver behind<br>me whose view is not good enough to overtake (29)     |

Note: Factor loadings below 0.30 were omitted for the sake of clarity.

<span id="page-7-0"></span>Table 3 Two-factor solution of DAIS

| Aggressive behaviours (self) |                                      |                        | Aggressive behaviours (other drivers) |                                      |                        |                                                                 |  |  |
|------------------------------|--------------------------------------|------------------------|---------------------------------------|--------------------------------------|------------------------|-----------------------------------------------------------------|--|--|
| (Mean/SD)                    | Hostile<br>aggression<br>and revenge | Aggressive<br>warnings | (Mean/SD)                             | Hostile<br>aggression<br>and revenge | Aggressive<br>warnings | Items (number)                                                  |  |  |
| (1.40/0.87)                  | 0.84                                 |                        | (1.20/0.56)                           | 0.47                                 |                        | Hugged the rear bump (10)                                       |  |  |
| (1.39/0.83)                  | 0.82                                 |                        | (1.19/0.57)                           | 0.85                                 |                        | Threatened physically (9)                                       |  |  |
| (1.51/0.96)                  | 0.81                                 |                        | (1.36/0.73)                           | 0.81                                 |                        | Physically attacked (6)                                         |  |  |
| (1.32/0.81)                  | 0.80                                 |                        | (1.16/0.59)                           | 0.78                                 |                        | Rammed a vehicle (5)                                            |  |  |
| (1.78/1.15)                  | 0.58                                 |                        | (1.39/0.82)                           | 0.66                                 |                        | Chased (7)                                                      |  |  |
| (1.80/109)                   | 0.52                                 |                        | (1.37/0.85)                           | 0.58                                 |                        | Threatened verbally (4)                                         |  |  |
| (3.20/1.14)                  |                                      | 0.76                   | (2.76/1.16)                           |                                      | 0.62                   | Flashed lights (8)                                              |  |  |
| (2.74/1.14)                  |                                      | 0.71                   | (2.12/1.13)                           |                                      | 0.72                   | Made a hand gesture (3)                                         |  |  |
| (3.21/1.11)                  |                                      | 0.70                   | (2.68/1.15)                           |                                      | 0.57                   | Sounded horn (13)                                               |  |  |
| (2.94/1.19)                  |                                      | 0.63                   | (1.92/0.96)                           |                                      | 0.49                   | Cut up (1)                                                      |  |  |
| (2.53/1.15)                  |                                      | 0.61                   | (2.12/1.19)                           |                                      | 0.53                   | Swore/verbally abused (2)                                       |  |  |
| (2.25/1.13)                  |                                      | 0.53                   | (1.54/0.87)                           | 0.33                                 | 0.34                   | Prevented or obstructed<br>from manoeuvring<br>the vehicle (11) |  |  |
| (2.12/1.12)                  | 0.31                                 | 0.43                   | (1.60/0.95)                           | 0.41                                 | 0.39                   | Drove slowly in order to<br>annoy the driver behind (12)        |  |  |
| Initial<br>eigenvalues       | 6.34                                 | 1.65                   | Initial<br>eigenvalues                | 5.23                                 | 1.67                   |                                                                 |  |  |
| Variance<br>explained        | 45.42                                | 9.27                   | Variance<br>explained                 | 36.36                                | 8.70                   |                                                                 |  |  |

Note: Factor loadings below 0.30 were omitted for the sake of clarity.

## 3.3. Multiple correlation coefficients

[Table 4](#page-8-0) lists correlations between background variables, errors, violations, positive driver behaviours, hostile aggression and revenge committed by self or other drivers, aggressive warnings committed by self or other drivers, and the number of traffic accidents and offences. Age, sex, and lifetime mileage correlated with the number of accidents and offences. Age was negatively significantly related to violations, errors, hostile aggression and revenge and aggressive warnings committed by self or other drivers, whereas it was positively associated with positive driver behaviours. Being male was related to the high number of violations and aggressive warnings committed by other drivers. Lifetime mileage was positively related to positive driver behaviours while it was negatively associated with errors and hostile aggression and revenge committed by self.

Self-reported accident involvement was associated only with violations, but most strongly with offences. The number of offences correlated significantly only violations but not errors and positive behaviours.

Positive driver behaviours were negatively associated with errors, violations, hostile aggression and revenge, and aggressive warnings committed by self or other drivers, whereas errors and violations correlated positively with all aggression scales.

<span id="page-8-0"></span>Table 4 Correlations among DBQ scales, positive driver behaviours, DAIS subscales, demographic variables, and the number of accidents and offences

| Variables                                       | 1          | 2                 | 3      | 4             | 5             | 6      | 7      | 8      | 9     | 10           | 11     |
|-------------------------------------------------|------------|-------------------|--------|---------------|---------------|--------|--------|--------|-------|--------------|--------|
| 1. Age                                          | –          |                   |        |               |               |        |        |        |       |              |        |
| 2. Sex (1 = male,<br>2 = female)                | .18*** –   |                   |        |               |               |        |        |        |       |              |        |
| 3. Mileage                                      |            | .49*** .20*** –   |        |               |               |        |        |        |       |              |        |
| 4. Violations                                   |            | .25*** .26*** .08 |        | (.86)         |               |        |        |        |       |              |        |
| 5. Positive driver<br>behaviours                |            | .34*** .00        |        | .23*** .31*** | (.84)         |        |        |        |       |              |        |
| 6. Errors                                       | .27***     | .01               | .20**  |               | .46*** .29*** | (.79)  |        |        |       |              |        |
| 7. Hostile aggression<br>and revenge (self)     | .17**      | .00               | .14*   |               | .23*** .30*** | .32*** | (.89)  |        |       |              |        |
| 8. Aggressive<br>warning (self)                 | .26***     | .03               | .08    |               | .27*** .19*** | .26*** | .60*** | (.84)  |       |              |        |
| 9. Hostile<br>aggression and<br>revenge (other) | .18*** .10 |                   | .10    |               | .53*** .38*** | .42*** | .47*** | .19*** | (.86) |              |        |
| 10. Aggressive<br>warning (other)               | .13*       | .18***            | .02    |               | .52*** .19*** | .26*** | .17**  | .36*** |       | .47*** (.75) |        |
| 11. Number of accidents                         |            | .40*** .31***     | .38*** | .12*          | .11           | .05    | .02    | .04    | .01   | .06          | –      |
| 12. Number of offences                          |            | .27*** .22***     | .42*** | .13*          | .07           | .02    | .01    | .03    | .02   | .06          | .38*** |

Note: Alpha reliability coefficients are given in brackets. \* <sup>p</sup> < .05. \*\* <sup>p</sup> < .01. \*\*\* <sup>p</sup> < .001.

## 3.4. Poisson and negative binomial regression models

The distributions of accidents and offences did not follow normal distribution. Both an accident and a traffic penalty rare events too. Thus, Poisson and Poisson-gamma (or negative binomial) regression analyses (see [Lord, Washington, & Ivan, 2005](#page-13-0)) were performed. In each of the analyses, age, sex, and lifetime mileage were forced into the model to control their effect. In the second step, errors, violations, and positive driver behaviours were entered by using forward selection (p-value for F-to-enter was 0.05).

The appropriateness of the regression models based on Poisson distribution was tested by using goodness-of-fit statistics. If chi-square statistics indicated statistically significant misfit, then models were constructed by using negative binomial distribution. Goodness-of-fit statistics indicated misfit of Poisson distribution for both accidents (v<sup>2</sup> (272) = 540.62, p < .001) and traffic offences (v2 (270) = 1096.99, p < .001). Therefore, models based on negative binomial distribution were constructed.

As shown in [Table 5,](#page-9-0) age, sex, lifetime mileage, and violations were significantly related to the number of accidents and penalties. Older drivers reported more accidents and offences. Sex (being female) and violations were the strongest predictors of accident involvement and penalties. According to incidence rate ratios, female drivers had 46% less accidents and 71% less offences

<span id="page-9-0"></span>Table 5 Negative binomial regression analyses on number of accidents and offences

| Variables                                   | Incidence rate ratios (IRR) | Std. Err. | Z-value     | (95% Conf. interval) |
|---------------------------------------------|-----------------------------|-----------|-------------|----------------------|
| Number of accidents as a dependent variable |                             |           | Pseudo-R2 = | .07                  |
| Age                                         | 1.02                        | 0.01      | 4.10***     | 1.01–1.04            |
| Sex (1 = male, 2 = female)                  | 0.54                        | 0.08      | 4.23***     | 0.40–0.72            |
| Lifetime mileage                            | 1.00                        | 0.00      | 2.75**      | 1.00–1.00            |
| Violations                                  | 1.27                        | 0.11      | 2.65**      | 1.06–1.51            |
| Number of offences as a dependent variable  |                             |           | Pseudo-R2 = | .12                  |
| Age                                         | 1.05                        | 0.01      | 4.94***     | 1.03–1.07            |
| Sex (1 = male, 2 = female)                  | 0.29                        | 0.07      | 4.98***     | 0.18–0.48            |
| Lifetime mileage                            | 1.00                        | 0.00      | 3.55***     | 1.00–1.00            |
| Violations                                  | 2.37                        | 0.35      | 5.79***     | 1.77–3.17            |

Table 6 Hierarchical regression analyses on hostile aggression and revenge (self and other) and aggressive warnings (self and other)

| Step | Independent variables                                          | R2   | ADJ-R2 | Change in R2 | F        | df    | b       |
|------|----------------------------------------------------------------|------|--------|--------------|----------|-------|---------|
|      | Hostile aggression and revenge (self) as a dependent variable  |      |        |              |          |       |         |
| 1.   | Age                                                            | 0.04 | 0.03   | 0.04         | 3.60*    | 3,278 | 0.15*   |
| 2.   | Positive driver behaviours                                     | 0.18 | 0.16   | 0.14         | 9.88***  | 6,275 | 0.22*** |
|      | Errors                                                         |      |        |              |          |       | 0.24*** |
|      | Aggressive warning (self) as a dependent variable              |      |        |              |          |       |         |
| 1.   | Age                                                            | 0.08 | 0.07   | 0.08         | 8.39***  | 3,279 | 0.32*** |
| 2.   | Violations                                                     | 0.14 | 0.12   | 0.06         | 7.68***  | 6,276 | 0.15*   |
|      | Errors                                                         |      |        |              |          |       |         |
|      | Hostile aggression and revenge (other) as a dependent variable |      |        |              |          |       |         |
| 1.   | Age                                                            | 0.06 | 0.05   | 0.06         | 5.38***  | 3,276 | 0.18**  |
|      | Sex (1 = male, 2 = female)                                     |      |        |              |          |       | 0.16**  |
| 2.   | Violations                                                     | 0.40 | 0.39   | 0.34         | 30.07*** | 6,273 | 0.43*** |
|      | Positive driver behaviours                                     |      |        |              |          |       | 0.23*** |
|      | Errors                                                         |      |        |              |          |       | 0.18*** |
|      |                                                                |      |        |              |          |       | 0.14*   |
|      | Aggressive warning (other) as a dependent variable             |      |        |              |          |       |         |
| 1.   | Age                                                            | 0.07 | 0.06   | 0.07         | 6.79***  | 3,272 | 0.18**  |
|      | Sex (1 = male, 2 = female)                                     |      |        |              |          |       | 0.23*** |
| 2.   | Violations                                                     | 0.30 | 0.28   | 0.23         | 18.93*** | 6,269 | 0.48*** |

than male drivers. Besides, accident rate increased 1.27 times (increase of 27%) and traffic offences rate becomes 2.37 times higher (increase of 137%) with increase of one unit in violations.

<sup>\*</sup> <sup>p</sup> < .05. \*\* <sup>p</sup> < .01. \*\*\* <sup>p</sup> < .001.

<sup>\*</sup> <sup>p</sup> < .05. \*\* <sup>p</sup> < .01. \*\*\* <sup>p</sup> < .001.

## 3.5. Hierarchical regression analyses

Four separate hierarchical regression analyses were performed on each of the outcome variables (hostile aggression and revenge, and aggressive warnings committed by self or other drivers). In each of the regression analyses, age, sex, and lifetime mileage were forced into the model in the first step to control for their effect. Errors, violations, and positive driver behaviours were entered in the second step by using stepwise procedure.

As shown in [Table 6](#page-9-0), the age of the drivers was related to all dependent variables whereas sex (being male) was significantly associated with hostile aggression and revenge, and aggressive warnings committed by other drivers.

After controlling the age, sex, and lifetime mileage, the results of the regression analyses in the second steps showed that violations were related to all dependent variables except hostile aggression and revenge committed by self. This result might indicate that violations in DBQ (both ordinary and aggressive violations) do not capture aggression in traffic well. Errors were significantly associated with the hostile aggression and revenge committed by self and other drivers, and aggressive warnings committed by self. Positive driver behaviours were negatively related to the hostile aggression and revenge committed by self and other drivers. Those who reported a high number of positive driver behaviours showed a low number of hostile aggression and revenge, and they were exposed to a low number of hostile aggression.

## 4. Discussion

The results of the present study indicated that the DBQ with positive driver behaviours items had a clear factor structure, high item loadings, and acceptable internal consistency. It is clear that driving style includes both ''negative'' and ''positive'' driver behaviours at the same time. It was found that positive driver behaviours were negatively associated with errors and violations. Obviously, being a polite driver requires a driver to search the traffic environment well and be aware of the situation in order to avoid behaviours that might annoy other road users. It is also possible that positive driver behaviours are ''secondary'' behaviours, which are expressed in less-demanding driving situations. Positive driver behaviours were found to have a positive relationship to age and mileage. Obviously, novice drivers need to pay more attention to the traffic situation and vehicle handling, and have less resources available for secondary behaviours like politeness to other road users. It is also possible that drivers learn the ''unofficial behaviour code'' of driving during time spent in traffic. In the beginning of their driving career, novice drivers driving behaviour can be expected to be more based on formal rules. Within the first years of their driving, drivers learn that some rules are more important than the others. At the same time, the socialization process of driving teaches drivers the behavioural code of driving. Positive driver behaviours are ''good manners of driving'' which are learnt by experience in traffic.

The present study is consistent with the earlier findings that male drivers were involved in more accidents and committed offences and violations more frequently than female drivers (A˚ [berg &](#page-12-0) [Rimmo¨, 1998; Blockley & Hartley, 1995; Lawton, Parker, Manstead, & Stradling, 1997; Mesken,](#page-12-0) [Lajunen, & Summala, 2002; Parker et al., 1992](#page-12-0)). Similarly, male drivers seemed to be exposed to more aggressive warnings committed by other drivers than female drivers. Being male together with their gender roles (e.g., high masculinity) might influence a drivers interpretation of the traffic situation and other drivers behaviours. It is possible that they may be more likely to interpret other drivers acts as aggressive, than other drivers do (O¨ [zkan & Lajunen, in press\)](#page-13-0).

Regression analysis revealed that violations were related to the number of accidents and penalties. These findings are consistent with [Parker, Reason, Manstead, and Stradling \(1995\),](#page-13-0) [Parker,](#page-13-0) [West, et al. \(1995\),](#page-13-0) and [Mesken et al. \(2002\)](#page-13-0). It was also found that violations were related to all aggression scales except hostile aggression and revenge committed by the respondents themselves. Interestingly, errors were related to hostile aggression and revenge committed by the respondents and other drivers, and aggressive warnings committed by the respondents. [Su¨mer et al. \(2003\)](#page-13-0) found that drivers with low levels of self-reported perceptual-motor and safety skills reported the highest levels of hostile aggression and revenge. It seems that feelings of frustration and disappointment because of low perceptual-motor skills may result in direct aggression toward other drivers if these feelings are not buffered by safety skills. Similarly, respondents seemed to express their anger and get a similar reaction from other road users when they reported a high number of errors and a low orientation to positive driver behaviours. Since positive driver behaviours were negatively related to hostile aggression and revenge committed by drivers themselves and other drivers, positive driver behaviours can be said to be good for the traffic climate and driving atmosphere. The perception of which, in turn, may positively influence drivers driving style. Safety oriented driving style may buffer the effects of errors and violations on aggression, especially at places in which interpersonal contacts are frequent.

Traffic safety in the literature is normally seen as the lack of accidents, damage or injury, and hence, is mainly defined by using negatives [\(Lehtima¨ki, 2001](#page-13-0)). In addition to the lack of safety training in the driving school curriculum, there is no training for stimulating ''positive'' driving style because the main emphasis of driving courses is to improve only driving skills and make candidates to drive autonomously. However, it is obvious that behaviours, which could potentially cause accidents or create a traffic atmosphere in which drivers are more likely to commit more violations, must also be considered. We may also need to extend the definition of traffic safety by emphasizing the positive side of driving too. Traffic should remind drivers of a lack of accidents and conflicts, i.e., ''Positive Traffic Culture'', rather than the fear of a danger.

The present study has some methodological limitations. The data were based solely on drivers self-reports of behaviour and no observations were made. However, several studies have indicated that self-reports of driving correspond well to actual driving behaviour. [West, French, Kemp, and](#page-13-0) [Elander \(1993\)](#page-13-0) reported a correlation of 0.65 between observed driving speed and responses on the driving speed subscale of their Driving Style Questionnaire. In addition, [Ingham \(1991\)](#page-12-0) found high correlations between recorded driving on a 40-km test route and self-reported driver behaviour. The measurement of accident involvement was based on a self-report of all past accidents and was not restricted to a shorter time period like the last three years. Simply because of forgetting to report some accidents (approximately 30% per year) ([Maycock, Lockwood, & Lester,](#page-13-0) [1991](#page-13-0)), some respondents may have underestimated the number of accidents in which they had been involved. It is also possible that some respondents embellished their answers about positive and aggressive driving. It should be noted, however, the respondents completed the questionnaires anonymously and could not gain anything by giving embellished responses. In their experimental study about DBQ and social desirability bias, [Lajunen and Summala \(2003\)](#page-12-0) concluded that the bias caused by socially desirable responding is very small in DBQ responses. Results of this study <span id="page-12-0"></span>indicated that positive driver behaviours are a part of driving style. This finding together with the scale should be tested in other countries where the traffic environment is safer, more stable, and more established than in Turkey.

## Acknowledgements

We would like to thank David Lamble for valuable comments. This study was supported by the Turkish Academy of Sciences, in the framework of the Young Scientist Award Program (TL/ TUBA-GEBIP/2001-2-14), and by the Graduate School of Psychology in Finland.

## References

- A˚ berg, L., & Rimmo¨, P.-A. (1998). Dimensions of aberrant driver behaviour. Ergonomics, 41, 39–56.
- Ben-Ari, O. T., Mikulincer, M., & Gillath, O. (2003). The multidimensional driving style inventory-scale construct and validation. Accident Analysis and Prevention, 952, 1–10.
- Bianchi, A., & Summala, H. (2002). Moral judgment and driver behavior among Brazilian students. Psychological Reports, 91(3), 759–766.
- Blockley, P. N., & Hartley, L. R. (1995). Aberrant driving behaviour: Errors and violations. Ergonomics, 38, 1759–1771.
- Elander, J., West, R., & French, D. (1993). Behavioral correlates of individual differences in road traffic crash risk: An examination of methods and findings. Psychological Bulletin, 113, 279–294.
- Fishbein, M., & Ajzen, I. (1975). Belief, attitude, intention and behaviour: An introduction to theory and research. Reading, MA: Addison-Wesley.
- French, D. J., West, R. J., Elander, J., & Wilding, J. M. (1993). Decision-making style, driving style, and self-reported involvement in road traffic accidents. Ergonomics, 36, 627–644.
- Glendon, A. I., Dorn, L., Matthews, G., Gulian, E., Davies, D. R., & Debney, L. (1993). Reliability of the Driving Behaviour Inventory. Ergonomics, 36, 719–726.
- Gulian, F., Matthews, G., Glendon, A. I., Davies, D. R., & Debney, L. M. (1989). Dimensions of driver stress. Ergonomics, 32, 585–602.
- Ingham, R. (1991). The young driver. Paper presented to the conference ''New Insights Into Driver Behaviour'' organized by Parliamentary Advisory Council for Transport Safety (PACTS), London, 21 October 1991.
- Kontogiannis, T., Kossiavelou, Z., & Marmaras, N. (2002). Self-reports of aberrant behaviour on the roads: Errors and violations in a sample of Greek drivers. Accident Analysis and Prevention, 34, 381–399.
- Krahe´, B., & Fenske, I. (2002). Predicting aggressive driving behavior: The role of Macho personality, age, and power of car. Aggressive Behavior, 28, 21–29.
- Lajunen, T., & O¨ zkan, T. (2004). Culture, safety culture, and traffic safety in Turkey and in Europe. The Turkish Driver Behaviour Questionnaire (T-DBQ): Validity and norms. Report no: SBB-3023. The Scientific and Technical Research Council of Turkey (TU¨ B\_ ITAK), June 2004, Ankara, Turkey.
- Lajunen, T., Parker, D., & Summala, H. (1999). Does traffic congestion increase driver aggression? Transportation Research Part F, 2, 225–236.
- Lajunen, T., Parker, D., & Summala, H. (2004). The Manchester Driver Behaviour Questionnaire: A cross-cultural study. Accident Analysis and Prevention, 36, 231–238.
- Lajunen, T., & Summala, H. (2003). Can we trust self-reports of driving? Effects of impression management on driver behaviour questionnaire responses. Transportation Research Part F, 6, 97–107.
- Lawton, R., Parker, D., & Stradling, S. G. (1997). Predicting road traffic accidents: The role of social deviance and violations. British Journal of Psychology, 88, 249–262.
- Lawton, R., Parker, D., Manstead, A. S. R., & Stradling, S. G. (1997). The role of affect in predicting social behaviours: The case of road traffic violations. Journal of Applied Social Psychology, 27, 1258–1276.

- <span id="page-13-0"></span>Lehtima¨ki, R. The conceptions of traffic safety among young male drivers. Academic Dissertation, University of Helsinki, August 2001 (ISBN 952-9763-04-2).
- Lord, D., Washington, S. P., & Ivan, J. N. (2005). Poisson, Poisson-gamma and zero-inflated regression models of motor vehicle crashes: Balancing statistical fit and theory. Accident Analysis and Prevention, 37, 35–46.
- Macmillan, J. (1975). Deviant drivers. Westmead, UK: Saxon House.
- Maycock, J., Lockwood, C. R., & Lester, J. F. (1991). The accident liability of car drivers. Report no: 315. Crowthorne: Transport and Road Research Laboratory.
- Mesken, J., Lajunen, T., & Summala, H. (2002). Interpersonal violations, speeding violations and their relation to accident involvement in Finland. Ergonomics, 45(7), 469–483.
- O¨ zkan, T., & Lajunen, T., in press. Why are young men risky drivers? The effects of sex and gender-role on aggressive driving, traffic offences and accident involvement among young men and women Turkish drivers. Aggressive Behavior.
- Parker, D., Manstead, A. S. R., Stradling, S. G., & Reason, J. T. (1992). Determinants of intention to commit driving violations. Accident Analysis and Prevention, 24, 117–131.
- Parker, D., Reason, J. T., Manstead, A. S. R., & Stradling, S. G. (1995). Driving errors, driving violations and accident involvement. Ergonomics, 38, 1036–1048.
- Parker, D., West, R., Stradling, S. G., & Manstead, A. S. R. (1995). Behavioural characteristics and involvement in different types of traffic accident. Accident Analysis and Prevention, 27, 571–581.
- Ra¨met, T. (2003). Nuorten kuljettajien ajotavat ja-asenteet: Periytyva¨tko¨ vanhempien ajotavat? (Young drivers driving style and attitudes: Do the young drivers ''inherit'' the driving style of their parents?). Unpublished Master Thesis, University of Helsinki, Finland.
- Ranney, T. A. (1994). Models of driving behavior: A review of their evolution. Accident Analysis and Prevention, 26, 733–750.
- Rasmussen, J. (1980). What can be learned from human error reports? In K. Duncan, M. Gruneberg, & D. Wallis (Eds.), Changes in working life. London: Wiley.
- Reason, J. T. (1990). Human error. New York: Cambridge University Press.
- Reason, J. T. (1999). Managing the risks of organizational accidents. Aldershot: Ashgate Publishing Company.
- Reason, J. T., Manstead, A. S. R., Stradling, S., Baxter, J., & Campbell, K. (1990). Errors and violations on the roads. Ergonomics, 33, 1315–1332.
- Sullman, M. J. M., Meadows, M. L., & Pajo, K. (2002). Aberrant driving behaviours amongst New Zealand truck drivers. Transportation Research Part F, 5(3), 217–232.
- Su¨mer, N., Lajunen, T., & O¨ zkan, T. (2003). Asymmetric relationship between driving performance and safety skills: A test on Turkish drivers. Presented at 9th Hellenic Conference of Psychological Research, 21–24 May, Rhodes, Greece.
- Su¨mer, N., & O¨ zkan, T. (2002). The role of driver behavior, skills, and personality traits in traffic accidents. Turkish Journal of Psychology, 17(50), 1–22.
- West, R., French, D., Kemp, R., & Elander, J. (1993). Direct observation of driving, self reports of driver behaviour and accident involvement. Ergonomics, 36, 557–567.
- Westerman, S. J., & Haigney, D. (2000). Individual differences in driver stress, error and violation. Personality and Individual Differences, 29, 981–998.
- Xie, C.-Q., & Parker, D. (2002). A social psychological approach to driving violations in two Chinese cities. Transportation Research Part F, 5, 293–308.