Journal of Safety Research 50 (2014) 99–107

Contents lists available at ScienceDirect

Journal of Safety Research journal homepage: www.elsevier.com/locate/jsr

The Driver Behaviour Questionnaire: A North American analysis Sheila T.D. Cordazzo, Charles T. Scialfa ⁎, Katherine Bubric, Rachel Jones Ross Department of Psychology, University of Calgary, 2500 University Drive, N.W. Calgary, Alberta, T2N 1N4, Canada

a r t i c l e

i n f o

Article history: Received 27 August 2013 Received in revised form 17 April 2014 Accepted 1 May 2014 Available online 16 May 2014 Keywords: Lapses Errors Violations Driver behavior Collisions

a b s t r a c t Introduction: The Driver Behaviour Questionnaire (DBQ), originally developed in Britain by Reason et al. [Reason, J., Manstead, A., Stradling, S., Baxter, J., & Campbell, K. (1990). Errors and violations on the road: A real distinction? Ergonomics, 33, 1315–1332] is one of the most widely used instruments for measuring driver behaviors linked to collision risk. Method: The goals of the study were to adapt the DBQ for a North American driving population, assess the component structure of the items, and to determine whether scores on the DBQ could predict selfreported traffic collisions. Results: Of the original Reason et al. items, our data indicate a two-component solution involving errors and violations. Evidence for a Lapses component was not found. The 20 items most closely resembling those of Parker et al. [Parker, D., Reason, J. T., Manstead, A. S. R., & Stradling, S. G. (1995). Driving errors, driving violations and accident involvement. Ergonomics, 38, 1036–1048] yielded a solution with 3 orthogonal components that reflect errors, lapses, and violations. Although violations and Lapses were positively and significantly correlated with self-reported collision involvement, the classification accuracy of the resulting models was quite poor. Practical applications: A North American DBQ has the same component structure as reported previously, but has limited ability to predict self-reported collisions. © 2014 National Safety Council and Elsevier Ltd. All rights reserved.

1. Introduction Drivers often engage in behaviors that pose a risk to both themselves and to other road users. While many of these unsafe actions are active, conscious rule violations, others are the result of errors due to inexperience, momentary mistakes or inattention. Intentional or not, both rule violations and deficiencies in memory, judgment, or situational awareness can and do contribute to traffic collisions (Stanton & Salmon, 2009; Wierwille et al., 2002). Because of this, there is a need for tools that can measure these behaviors and the frequency with which they are committed, and can determine which specific actions predict traffic collision involvement. For more than two decades, there has been a body of research published regarding the creation, modification, and evaluation of one such tool. In 1990, Reason, Manstead, Stradling, Baxter, and Campbell, introduced a 50-item, self-report Driver Behaviour Questionnaire, in which drivers rated the frequency of risky behaviors executed while driving. Winter and Dodou (2010) identified almost 200 studies that have since used the DBQ in part or in its entirety. af Wåhlberg, Dorn, and Kline (2011) concur that the DBQ is one of the most widely used instruments for measuring driving behavior. The original publication (Reason, Manstead, Stradling, Baxter, & Campbell, 1990) involved a sample of over 500 drivers at least 20 years of age. Principal components analysis identified three factors that ⁎ Corresponding author. Tel.: +1 403 220 4951; fax: +1 403 282 8249. E-mail addresses: [email protected] (S.T.D. Cordazzo), [email protected] (C.T. Scialfa), [email protected] (K. Bubric), [email protected] (R.J. Ross).

http://dx.doi.org/10.1016/j.jsr.2014.05.002 0022-4375/© 2014 National Safety Council and Elsevier Ltd. All rights reserved.

accounted for 33% of the variance in responses; these were identified as violations, errors, and lapses. Violations were defined as “deliberate deviations from those practices believed necessary to maintain the safe operation of a potentially hazardous system” (pp. 1316), in other words, a violation is the breach of a law or socially accepted code of behavior. Common examples are speeding and driving while under the influence of drugs or alcohol. In contrast, errors were defined as “the failure of planned actions to achieve their intended consequences” (p. 1315). Errors have been divided into slips/lapses and mistakes, both of which are unintended deviations of action from intention. Slips are actions that were not planned (e.g., turn on the headlights instead of wipers), and lapses are related to memory failures (e.g., forget the route one is driving or locking one's keys in the vehicle). Mistakes are faults of judgment or decisionmaking, such as underestimating the speed of an oncoming vehicle. Following on Reason et al.'s work, Parker, Reason, Manstead, and Stradling (1995) selected the 8 items that had the highest component loadings for violations, errors, and lapses and administered this abbreviated instrument to a sample of 1,656 British drivers, with an age range of 17–70 years. The same three-component solution reported by Reason et al. (1990) was obtained. Since its creation, the DBQ has been modified, updated, and adapted for a variety of driving environments and populations (e.g., Aberg & Rimmo, 1998; Bener, Ozkan, & Lajunen, 2008; Gabaude, Marquie, & Obriot-Claudel, 2010; Kontogiannis, Kossiavelou, & Marmaras, 2002). For example, Blockey and Hartley (1995) gave an Australian version to a sample of 135 predominantly young drivers. They also obtained a three-component solution that explained 27.7% of the total variance,

100

S.T.D. Cordazzo et al. / Journal of Safety Research 50 (2014) 99–107

but the loadings suggested a different composition. The components were identified as general errors, dangerous errors, and dangerous violations. General Errors were a mixture of slips, mistakes, and unintentional violations with a variable level of risk of a collision. Dangerous Errors were also defined as a combination of slips, mistakes, and unintentional violations, with a greater level of risk of a collision. The last component, named dangerous violations consisted of those violations with a high risk of collision. Blockey and Hartley suggested that the differences between their study and the original work likely were due to the age and gender differences in the samples, as well as the socioeconomic and cultural differences between the two countries. Despite attempting a virtually literal French translation of the Reason et al. instrument, Gabaude et al. (2010) also reported that their components were more accurately defined by risk level rather than by behavior type. In general, however, more recent investigations using the DBQ have replicated the three-component structure of violations, errors and lapses. For example, in 1997, Lawton, Parker, Manstead and Stradling refined the Violations scale by distinguishing between ordinary and aggressive violations. Ordinary violations were actions that involved disregarding traffic laws, such as running red lights, while aggressive violations held an emotional component, and included behaviors like getting angry and chasing another vehicle. A structure comprised of slips/lapses, mistakes, ordinary violations, and aggressive violations has been broadly replicated, although the distinction between ordinary and aggressive violations is not always obtained at the factor or component level (Harrison, 2009; Lajunen, Parker, & Summala, 2004; Lawton, Parker, Manstead, & Stradling, 1997; Parker, McDonald, Rabbitt, & Sutcliffe, 2000; Parker et al., 1995; Rimmo, 2002). The Swedish version of the DBQ (DBQ-SWE) was created based on the original British scales (Parker et al., 1995; Reason et al., 1990), with additional items measuring Errors, resulting in a 104-item scale (Aberg & Rimmo, 1998). The replication of the Reason et al. items confirmed the three-component solution, with 32.8% of total variance explained. Analysis with the new items added produced a four-component solution, with 44% of total variance explained. The components were identified as violations, mistakes, errors by inattention, and errors by inexperience. Based on the DBQ-SWE, Rimmo (2002) carried out a confirmatory factor analysis of a four-factor model comprised of violations, mistakes, inattention errors, and inexperience errors. There were approximately 5,000 drivers in 4 different samples divided by age. The model was an approximate fit regardless of age and gender. The author suggested that because the fit to the model was adequate for each age group, differences in the factor structure between these groups may only be of minor importance. Although Rimmo's work suggests that factor structure is independent of demographic characteristics, factor scores tend to be related to both age and gender. Males score higher on the Violations factor, and females score higher on Lapses (Parker et al., 1995; Reason et al., 1990; Winter & Dodou, 2010). Compared to their younger counterparts, older drivers report fewer violations and errors, but more slips and lapses (Blockey & Hartley, 1995; Parker et al., 2000; Winter & Dodou, 2010). Although errors tend to decrease with age, a Swedish study distinguished between errors of inexperience and errors of inattention, and found that errors of inattention are more prevalent in older drivers (Aberg & Rimmo, 1998). The DBQ has also been used as a predictor of individual differences in collision involvement, but there is considerable debate concerning its predictive utility. Davey, Wishart, Freeman, and Watson (2007), Ozkan and Lajunen (2005) and Parker et al. (1995) among others have found in their samples that a higher score on the Violations component was a significant predictor of self-reported collisions. DeLucia, Bleckley, Meyer, and Bush (2003), Freeman, Wishart, Davey, Rowland, and Williams (2009), and Sumer (2003) reported an association between collisions and the Errors component. In a meta-analysis study, Winter and Dodou (2010) concluded that Errors and Violations are both associated with self-reported traffic collisions, with an overall correlation of .10 and .13, respectively, when based on zero-order effects.

In contrast, af Wåhlberg et al. (2011) argued that the apparent correlation between DBQ responses and collision risk is an artifact of common method variance and social desirability biases (af Wåhlberg, 2010) stemming from the self-report nature of the instrument and collision history. Furthermore, they contend with justification that self-reports of collision history are prone to systematic biases that may inflate the correlation with other self-report measures. They found that the DBQ is able to predict only self-reported traffic collisions and not recorded collision data. The issue is complicated further because while self-reported collision history is by no means perfect, archival collision data are also prone to inadequacies (Arthur et al., 2005). In the current study, the DBQ was expanded by incorporating a larger sample of driving behaviors and by modifying test items to be appropriate for a North American population of drivers. One goal was to determine whether the resulting factor structure was consistent with the interpretation that the instrument measures errors, violations and lapses. A second goal was to ascertain if factor scores varied by age and gender, as previously reported. Finally, we sought to determine whether scores on the DBQ could predict self-reported traffic collisions and if so, with what magnitude. 2. Method 2.1. Materials A total of 82 items from various versions of the DBQ were considered initially for inclusion (Kline et al., 1992; Lajunen et al., 2004; Lawton et al., 1997; Ozkan, Lajunen, & Summala, 2006; Parker et al., 1995; Parker et al., 2000; Reason et al., 1990). From those items, 19 were retained verbatim and 46 were modified as needed to ensure clarity and relevance for the North American driving context. One example of a modification was the item “Attempt to drive away from traffic lights in third gear” (Lajunen et al., 2004; Ozkan et al., 2006; Parker et al., 1995; Reason et al., 1990). Due to the popularization of automatic transmissions in North America, this item was modified to, “Attempt to leave a parking space in the wrong gear.” A total of 17 items were omitted because of repetition, unclear wording, or because they were not clearly relevant to driving safety or the North American driving context (e.g., “How often do you lock yourself out of your car with the keys still inside?”). An additional 27 items were developed to capture behaviors deemed to be related to driving safety, but which were absent from earlier versions. For example, items related to distracted driving (e.g., “How often do you talk on your cell phone when you are driving?”) and situations that drivers avoid or have difficulties with (e.g., “How often do you avoid busy or complex roads and intersections?”) were added to the questionnaire. These items were added because of evidence that distraction and some kinds of impairments play an increasing role in collisions (Frittelli et al., 2009; Regan, Lee, & Young, 2009) or because they are particularly relevant for older drivers who make up a growing segment of the driving population (cf., Kline et al., 1992). Finally, 13 items with reversed wording were added to encourage respondents to read each question more carefully and to avoid the response biases in single-valence tests (Allen & Seaman, 2007). For example, corresponding to “Drive with a seatbelt on” we added, “Drive without a seatbelt on.” In total then, there were 105 items in the instrument. For this paper, only items that most clearly matched those in two common versions of the DBQ (Parker et al., 1995; Reason et al., 1990) were considered. Analysis of the entire set will form the topic of future papers. After pilot testing all items on a small sample of 20 specialists in driving, psychology, and scale construction, 5 items were added and 6 items were omitted for reasons of clarity and wording confusion. Responses were recorded on a 6-point Likert-type scale with the following anchors; never, very rarely, occasionally, often, nearly all the time, and always. A complete listing of the items and the changes is shown on Appendix 1.

S.T.D. Cordazzo et al. / Journal of Safety Research 50 (2014) 99–107

2.2. Procedures This DBQ was distributed in two separate mailings to 15,000 randomly selected, currently licensed Alberta Motor Association (AMA) members over 30 years of age. Because of the lack of younger drivers in the AMA membership roles, we collected younger drivers' data from the undergraduate student population at the University of Calgary. Participants in this sample held a valid driver's license, were enrolled in psychology courses, and were given course credit for their participation. There was no monetary compensation or inducement provided to the AMA members for responding. This study was approved by the Conjoint Faculties Research Ethics Board (CFREB) of the University of Calgary (File # 7207). 2.3. Participants Of the questionnaires distributed to AMA members, 19.35% (N = 2902) were completed and returned. Reflecting the population distribution of the province, 68% came from people in the Calgary or Edmonton areas. Those few participants from the AMA sample under the age of 30 (N = 29) or missing data on age (N = 27) and participants without a valid driver's license (N = 7) were excluded. Thus, the AMA sample contained data from 2,839 drivers. An additional 484 University of Calgary students completed and returned the survey. From the student sample, participants over the age of 30 (N = 23) or missing data on age (N = 5) were excluded from analyses, yielding 456 respondents. Table 1 provides the percentage of participants by age and gender in both samples. In the AMA sample, approximately one-half (53.9%) of the respondents were females. Drivers were, on average, 60.65 years old (SD = 13.81 years). They rated themselves as generally healthy, with an average of 4.04 (SD = .76) on a scale of 1 (very poor) to 5 (excellent). They had held a valid license for a mean duration of 41.39 years (SD = 13.77 years). Most of the participants (72.4%) reported their annual driving distance to be between 5,000 and 25,000 km/year. Some 13% indicated that they had experienced at least 1 collision in the past 2 years. In the student sample, 42.8% were female with a mean age of 20.95 years (SD = 2.16 years). They had held a driver's license for an average of 3.72 years (SD = 2.25 years). For 68.9%, their self-reported annual distance driven was between 5,000 and 25,000 km/year. By self-report, like the AMA sample, their general health was quite good (M = 4.35, SD = .69). Reflecting the greater collision risk among younger drivers, 26.7% (N = 122) reported that they had experienced at least 1 collision in the past 2 years. The sampling strategy was intended to gather data from across the age span of current drivers but produced a group that did not match the general driving population in age. Specifically, compared to U.S. (U.S. Department of Transportation, Federal Highway Administration, 2011) and Canadian (Government of Canada, Transport Canada, 2012) statistics, the current data set contains an over-representation of older drivers. 3. Results Items that were reverse-coded, when transformed for comparison, resulted in significantly different responses on 12 out of 13 pairs of

101

items that were substantively identical. The direction of the difference was positive for some of the pairs of items (N = 8) but negative for other pairs of items (N = 5). This suggested that respondents had difficulties responding to reverse-coded items. As a consequence, reversed items were eliminated (N = 13). All missing data (.8%) were replaced with the mean of each variable, based on the entire sample. The original DBQ (Reason et al., 1990) had 50 items (9 mistakes, 19 violations, and 22 slips/lapses). In our adaptation and revision, the wording of 29 of these items was modified for clarity among North American drivers and 14 items were omitted because they were repetitious, confusing, or had an unclear meaning (e.g., “Park on a double-yellow line and risk a fine”). Thus, of the original 50 items, 36 items were retained (7 mistakes, 13 violations, and 16 slips/lapses). Reason et al. (1990) analyzed their data using principal components analyses (PCA), with a Varimax rotation. For ease of comparison, the same approach was used here. Principal components analysis is a data reduction procedure that is designed to transform a matrix of correlations into a set of orthogonal variables that are linear combinations of the original scores. The first component accounts for the largest proportion of variance in the original variables. Successive components account for the maximum remaining variance with the restriction that they are uncorrelated with previously extracted components. Examination of the component loadings (often setting a criterion minimum loading for inclusion in the component) provides guidance into the interpretation of the components. Thus, if the first component consists of items that are related to trait A but not trait B, then it is labeled Trait A. The components can be rotated so as to bring about “simple structure,” where items tend to load on one and only one component, facilitating their interpretation. Reason et al. (1990) reported that their DBQ yielded three components, accounting for about 33% of the total variance; violations, errors and lapses. In contrast, our data initially produced a sixcomponent solution that explained 40.37% of the total variance. A forced, three-component solution with Varimax rotation, explained 30.96% of the variance, however, the loadings were inconsistent with the assertion that violations, errors and lapses were being measured. Examination of the scree plot suggested a two-component solution with a total variance of 27.06%. For comparison, Table 2 shows the prominent loadings with the components labeled from Reason et al. (1990). The first component is a mix of slips/lapses and mistakes that could be called Errors. As a single scale, it has considerable internal consistency, with a Cronbach's alpha of .86 and an average inter-item correlation of .22. The second component appears to be indexing the propensity for Violations. As a single-component scale, it also has considerable reliability, indicated by a Cronbach's alpha of .74 and a mean interitem correlation of .20. As noted previously, Parker et al. developed a shorter version of the DBQ by selecting 24 of the original 50 items that had high loadings on only 1 of the 3 components obtained. They reported that a PCA with a Varimax rotation produced a solution with three components accounting for 37.4% of the variance in scores, which they interpreted as measuring violations, errors and lapses. We had 20 items that closely resembled those used by Parker et al. Following their example, these items were analyzed using PCA and a Varimax rotation. A forced, three-component solution accounted for

Table 1 Participants grouped by age in decades and gender in each sample. Student sample(N = 456)

AMA sample (N = 2,839)

Age

b19

20–29

30–39

40–49

50–59

60–69

70–79

80+

Total

Male Female Total

2.4% 1.1% 3.5%

5.4% 4.8% 10.2%

1.9% 4.7% 6.6%

4.7% 7.4% 12.1%

9% 13.1% 22.1%

10.5% 10.1% 20.6%

9.6% 7.5% 17.1%

4.1% 3.8% 7.9%

47.6% 52.4% 100%

102

S.T.D. Cordazzo et al. / Journal of Safety Research 50 (2014) 99–107

Table 2 Rotated component matrix (limited to a 2-component solution). Brief items

Mean (SD)

Reason's factors⁎

Misjudge the distance between oncoming vehicles Fail to notice pedestrians crossing Turn left into the path of an oncoming vehicle Fail to notice someone stepping out from behind a bus Underestimate the speed of an oncoming vehicle Ignore a yield sign and almost collide Slam on the brakes to avoid a collision Get into the wrong lane when approaching In a line of cars nearly hit the car in front of you Switch on one thing instead of another Fail to notice someone waiting at a crosswalk Brake too hard on a slippery road Miss your exit on a highway Hit something when backing up Misjudge the space available in a parking lot Forget where you parked your car Forget that you have your high beams on Attempt to pass a vehicle that was signaling Realize you have no clear recollection of the road When turning right nearly hit a cyclist Attempt to leave a parking space in the wrong gear Intending to drive to destination A, realize you are en route to B Fail to check your mirrors before pulling Deliberately disregard the speed limit Drive especially close to or flash the car in front of you Try to pass in risky circumstances Disregard red lights or stop signs at night Drive while looking at a map or GPS Get involved in unofficial races Drive as fast along country roads at night on low beams Take a chance and run a red light Feel angered by another driver's behavior Check your speedometer and discover that you are faster Fail to yield right-of-way to a bus Drive the wrong direction down a deserted one-way street Drive over the legal blood-alcohol limit Amount of variance explained

1.34 (.51) 1.53 (.59) 1.32 (.51) 1.48 (.58) 1.63 (.64) 1.25 (.48) 1.79 (.68) 1.72 (.75) 1.39 (.59) 1.64 (.72) 1.82 (.64) 1.58 (.64) 2.03 (.69) 1.33 (.53) 1.27 (.52) 2.09 (.89) 1.71 (.64) 1.20 (.48) 1.81 (.84) 1.20 (.45) 1.18 (.44) 1.85 (.77) 1.57 (.82) 1.85 (1.07) 1.32 (.65) 1.47 (.72) 1.23 (.58) 2.39 (1.11) 1.16 (.49) 1.65 (1.07) 1.29 (.56) 1.11 (.44) 2.97 (1.08) 1.49 (.81) 1.11 (.35) 1.24 (.63)

M S M S S V S M S S V M S M M S S S S S S S S V V V V S V M V V V V V V

Component loadings .637 .597 .589 .566 .540 .540 .536 .536 .517 .517 .514 .512 .476 .459 .440 .439 .403 .393 .388 .386 .386 .386 .381 .707 .644 .642 .587 .569 .545 .493 .422 .392 .348 .310 .309 .274⁎⁎ 10.85%

16.21%

⁎ From Reason et al. (1990); Slips/Lapses (S), Mistakes (M) and Violations (V). Items are abbreviated. ⁎⁎ Because of the strong relationship between alcohol and traffic collisions we decided to keep this item, even though the loading is less than the .30 cut-off used otherwise.

35.71 % of the total variance, and the loadings we obtained were similar to those of Parker et al. (1995). Table 3 shows the results with the components labeled to compare results. Factor loadings of less than .30 are omitted for the sake of clarity.

Consistent with Parker et al. (1995), the first component reflects errors plus two items designated as lapses. Cronbach's alpha for these items is .74 and the average inter-item correlation is .30. The second and third components appear to measure lapses and violations. The

Table 3 Rotated component matrix (limited to a 3-component solution). Brief items

Mean (SD)

Parker's factors⁎

Ignore a yield sign and almost collide Fail to notice pedestrians crossing Underestimate the speed of an oncoming vehicle Get into the wrong lane when approaching Fail to check your mirrors before pulling In a line of cars nearly hit the car in front of you Brake too hard on a slippery road When turning right nearly hit a cyclist Hit something when backing up Attempt to pass a vehicle that you hadn't noticed was signaling Intending to drive to destination A, realize you are en route to B Forget where you parked your car Realize you have no clear recollection of the road Switch on one thing instead of another Attempt to leave a parking space in the wrong gear Drive especially close to or flash the car in front of you Get involved in unofficial races Deliberately disregard the speed limit Feel angered by another driver's behavior Drive over the legal blood-alcohol limit Amount of variance explained

1.25 (.48) 1.53 (.59) 1.63 (.64) 1.72 (.75) 1.57 (.82) 1.39 (.59) 1.58 (.64) 1.20 (.45) 1.33 (.53) 1.20 (.48) 1.85 (.77) 2.09 (.89) 1.81 (.84) 1.64 (.72) 1.18 (.44) 1.32 (.65) 1.16 (.49) 1.85 (1.07) 1.11 (.44) 1.24 (.63)

E E E L E E E E L E L L L L L V V V V V

Component loadings .614 .589 .582 .545 .511 .503 .483 .479 .473 .412 .686 .659 .595 .560 .402

15.03%

10.42%

.722 .649 .645 .549 .277⁎⁎ 10.26%

⁎ From Parker et al. (1995); Errors (E), Lapses (L) and Violations (V). Items are abbreviated. ⁎⁎ Because of the strong relationship between alcohol and traffic collisions we decided to keep this item, even though the loading is less than the .30 cut-off used otherwise.

S.T.D. Cordazzo et al. / Journal of Safety Research 50 (2014) 99–107

Lapses component has an internal consistency of .61 (inter-item correlation average = .24). The Violations component yielded a Cronbach's alpha of .55 and an average inter-item correlation of .21. Regression-type component scores (DiStefano, Zhu, & Mindrila, 2009) from the Reason et al. (1990) and Parker et al. (1995) items were regressed on age, gender, and annual distance driven. Table 4 summarizes these results. For the Parker et al. solution, violations and lapses decreased significantly with age, while there was no age effect for Errors. Similarly, for the Reason et al. solution, older adults had lower Violations scores, while there was no association between age and Errors. Males had higher scores on the violations component under both solutions, but there was no systematic gender effect on Errors. Lapses, calculated only for the Parker et al. solution, were of greater magnitude for females. Interestingly, as exposure increased, drivers reported fewer Lapses and Errors, but more Violations on Parker et al. items. Similar results were found for the comparable Reason et al. components. Binary logistic regression, often used to predict a dichotomous outcome variable from a set of continuous or categorical predictors, was used to determine the ability of the DBQ component scores to predict self-reported collision involvement. Table 5 shows that most DBQ components are positively correlated with self-reported collisions and, in the case of violations and lapses, the association is significant. However, the low r-squared values indicate that less than 1% of the variance in collisions can be accounted for by these predictors. Additionally, the technique allows one to determine the accuracy with which a person can be classified into the collision or no-collision group, with perfect accuracy reflecting both good sensitivity and specificity. Unfortunately, while the classification accuracy of the resulting model was 85%, no one was predicted to have a collision. That is, because most individuals were collision-free and the association with component scores was so low, the best prediction for an individual is that they did not report a collision. 4. Discussion Since its introduction (Reason et al., 1990), the Driver Behaviour Questionnaire has been used extensively in the attempt to measure risk-increasing driver behaviors and to predict collision risk. While the underlying variables measured by the DBQ vary, sometimes substantially across implementations and populations (e.g., Blockey & Hartley, 1995; Gabaude et al., 2010), there is some agreement that it provides an index of violations, mistakes or errors, as well as lapses in attention (Aberg & Rimmo, 1998; Lawton et al., 1997; Parker et al., 1995; Rimmo, 2002). Some components are associated with demographic variables such as age and gender (Blockey & Hartley, 1995; Parker et al., 2000). Additionally, there is some evidence that DBQ scores can predict

Table 4 Demographic predictors of component scores. Scale

Dependent

Parker et al. items Errors

Lapses

Violations Reason et al. items

Errors

Violations

⁎ (Male = 1, Female = 2).

Predictor

R2

Beta

p

Age Gender⁎ Distance Age Gender⁎ Distance Age Gender⁎ Distance Age Gender⁎ Distance Age Gender⁎ Distance

.001 b.001 .003 .032 .031 .009 .159 .046 .026 b.001 .010 .012 .250 .044 .037

.030 b.001 −.054 −.178 .177 −.094 −.399 −.213 .160 .006 .102 −.108 −.500 −.209 .192

.088 .985 .002 b.001 b.001 b.001 b.001 b.001 b.001 .743 b.001 b.001 b.001 b.001 b.001

103

Table 5 Component predictors of self-reported collisions using logistic regressions. Scale

Dependent

Predictor

R2

Beta

p

Parker et al. items

Collisions

Reason et al. items

Collisions

Errors Lapses Violations Errors Violations

b.001 .007 .010 .003 .017

.061 .171 .196 .106 .254

.203 b.001 b.001 .027 b.001

collision risk (Davey et al., 2007; Freeman et al., 2009; Ozkan & Lajunen, 2005; Winter & Dodou, 2010), although this association does not obtain uniformly and has been challenged on methodological grounds (af Wåhlberg et al., 2011). In the current study, items from the original scale developed by Reason et al. (1990) revealed a structure of two components that can be interpreted as violations and errors. These results are not consistent with Reason et al.'s data that indicated a Lapses component. In some ways, this should not be surprising because our instrument was a modification of the original DBQ both in number and content. Fourteen items had to be excluded from the original scale because they were not related clearly to driving safety (i.e., “Lock yourself out of your car with the keys still inside”), or because they concerned behaviors encountered infrequently in North America (i.e., “Forget which gear you are currently in and have to check with your hand”). Still, the current implementation had an adequate sample of items from all three components and the failure of a Lapses component to emerge suggests that these items are in need of revision. At the measurement model level, our results are in broad agreement with, and provide an independent, North American assessment of Parker et al. (1995) with respect to the component structure of the DBQ. Similar to their work, the 20 items most closely resembling their instrument yielded a solution with three orthogonal components that appear to reflect errors, lapses and violations. There are several other ways in which our data are consistent with the extant literature (Winter & Dodou, 2010) and particularly with that of Parker et al. (1995). In both studies, males were more likely than females to report violations and age was negatively correlated with Violations scores. In both cases, females scored higher on the Lapses component. As well, non-significant effects of gender were found for the Errors component. The only substantive difference between these studies was that while we found age to be negatively associated with Lapses, Parker et al. (1995) found no association between these variables. Can the DBQ be used to predict collision risk? Like Parker et al., Violations scores were associated with self-reported collisions. In contrast, however, we found that Lapses (using the Parker et al. component scores) and Errors (using the Reason et al. component scores) were also significant predictors of collisions. Both, Parker et al. (1995) and Reason et al. (1990) suggested that violations and errors are risky driver behaviors and that lapses are relatively harmless. It is no doubt true that violations (e.g., excessive speed) and errors/mistakes (e.g., underestimating speed of an oncoming vehicle) are related to collision risk, especially in young adult drivers (McGwin & Brown, 1999). However, it has become increasingly clear that lapses, broadly construed as inattention, are an important contributor to collisions, including “looked-failed-to-see” incidents (Hills, 1980; Koustanai, Boloix, Elslande, & Bastien, 2008). In fact, in their meta-analysis, Winter and Dodou (2010) included lapses of several types in their Errors category and found that Errors so defined were positively correlated with self-reported collision involvement. The associations found between DBQ component scores and selfreported collisions suggest some criterion-based validity to the instrument. The correlations, in the expected direction, are quite small in magnitude. In fact, the associations are so small that adding information about a person's component scores does not influence whether they are predicted to be collision-involved. Thus, from a practical point of view, the current DBQ is of no predictive utility for collision risk.

104

S.T.D. Cordazzo et al. / Journal of Safety Research 50 (2014) 99–107

Even if there is poor classification accuracy of self-reported collisions based on the DBQ, such an instrument has the potential to provide valuable information relevant to driver training and evaluation. Indeed, there is a significant association between self-reported collisions and behaviors such as violations. These results are consistent with the findings of previous studies (Davey et al., 2007; Ozkan & Lajunen, 2005; Parker et al., 1995), in that a higher score on the Violations component is a significant predictor of self-reported collisions. In our data, both the Violations and the Lapses component were significant predictors of collision. Given the size of the driving population, even a small relationship may have important economic and safety consequences. If, for example, 1% of the collisions can be predicted and there are around 10.8 million collisions in a given year (U.S. Census Bureau, 2012) at an average cost of $17,000 for property and liability, then there is the potential to save approximately $1.8 B. Additionally, normative data, were it to become available, could be used by drivers to gauge their relative standing in respect of behaviors that are amenable to remediation (e.g., inadequate mirror checking or propensity toward distraction). An easily accessed version of the DBQ could also be used to assess longitudinal change in global driving difficulties or specific driving behaviors that may signal the need for more extensive evaluation. This may be particularly valuable to older drivers, for whom a variety of medical conditions (e.g., dementia) impair driving performance (Dobbs & Schopflocher, 2010). Future research must incorporate more items to address the kinds of behaviors and contexts that are not sampled adequately in the current version. Two areas of specific concern are distractions (e.g., texting, talking at cell phone) and impairment in perception and cognition, the latter relevant not only to the aging population, but also to those driving with deficits in attention (Poulsen, Horswill, Wetton, Hill, & Lim, 2010), operating vehicles under the influence of drugs that influence performance (Drummer et al., 2004), and shift workers. We have already begun this research and intend to report the results of this effort in the near future. Another direction for future research would be the replication of these same items in other samples from the North America, as the sample limitations in this study, especially with the younger drivers, limit the generalizability of our results. Indeed, the younger drivers were not recruited from the random sample because the host organization has relatively few younger members. Rather, participants were all recruited from the same institution (University of Calgary). The other age groups were not completely random either, as they were all recruited from a list of AMA members. Therefore, the replication of these scales in a truly random sample would help to establish the reliability and generalizability of the DBQ NA version. However, even if the present sample is not a perfect fit with the North American population, the reported DBQ serves as a strong foundation for establishing a North American version of the instrument. Acknowledgments This research was funded by a grant from the Alberta Motor Association. We would like to express our sincere gratitude to the staff of the Alberta Motor Association, whose efforts were indispensable in the data collection phase of this study. References Aberg, L., & Rimmo, P. A. (1998). Dimensions of aberrant behavior. Ergonomics, 41, 39–56. af Wåhlberg, A. E. (2010). Social desirability effects in driver behavior inventories. Journal of Safety Research, 41, 99–106. af Wåhlberg, A., Dorn, L., & Kline, T. (2011). The Manchester Driver Behaviour Questionnaire as a predictor of road traffic accidents. Theoretical Issues in Ergonomics Science, 12, 66–86. Allen, I. E., & Seaman, C. A. (2007). Likert Scales and data analyses. Quality Progress, 40, 64–65. Arthur, W., Bell, S. T., Edwards, B. K., Day, E. A., Tubre, T. C., & Tubre, A. H. (2005). Convergence of self-report an archival crash involvement data: A two-year longitudinal follow-up. Human Factors: The Journal of the Human Factors and Ergonomics Society, 47, 303–313.

Bener, A., Ozkan, T., & Lajunen, T. (2008). The Driver Behaviour Questionnaire in Arab Gulf countries: Qatar and United Arab Emirates. Accident Analysis and Prevention, 40, 1411–1417. Blockey, P. N., & Hartley, L. R. (1995). Aberrant driving behavior: Errors and violations. Ergonomics, 38, 1759–1771. Davey, J., Wishart, D., Freeman, J., & Watson, B. (2007). An application of the Driver Behaviour Questionnaire in an Australian organisation fleet setting. Transportation Research Part F, 10, 11–21. DeLucia, P. R., Bleckley, M. K., Meyer, L. E., & Bush, J. M. (2003). Judgements about collision in younger and older drivers. Transportation Research Part F, 6, 63–80. DiStefano, C., Zhu, M., & Mindrila, D. (2009). Understanding and using factor scores: Considerations for the applied researcher. Practical Assessment, Research & Evaluation, 14, 1–11. Dobbs, B. M., & Schopflocher, D. (2010). The introduction of a new screening tool for the identification of cognitively impaired medically at-risk drivers: The SIMARD a modification of the DemTect. Journal of Primary Care & Community Health, 1, 119–127. Drummer, O. H., Gerostamoulos, J., Batziris, H., Chu, M., Caplehorn, J., Robertson, M. D., et al. (2004). The involvement of drugs in drivers of motor vehicles killed in Australian road traffic crashes. Accident Analysis and Prevention, 36, 239–248. Freeman, J., Wishart, D., Davey, J., Rowland, B., & Williams, R. (2009). Utilising the Driver Behaviour Questionnaire in an Australian organisational fleet setting: Can it identify risky drivers? Journal of the Australasian College of Road Safety, 20, 38–45. Frittelli, C., Borghetti, D., Ludice, G., Bonanini, E., Maestri, M., Tognoni, G., et al. (2009). Effects of Alzheimer's disease and mild cognitive impairment on driving ability: A controlled clinical study by simulated driving test. International Journal of Geriatric Psychiatry, 24, 232–238. Gabaude, C., Marquie, J. -C., & Obriot-Claudel, F. (2010). Self-regulatory driving behaviour in the elderly: Relationships with aberrant driving behaviours and perceived abilities. Le travail humain, 73, 31–52. Government of Canada, Transport Canada (2012). Number of licensed drivers by gender and by age. http://www.tc.gc.ca/eng/roadsafety/tp-tp3322-2004-page11-701.htm (Retrieved from) Harrison, W. (2009). Reliability of the driver behaviour questionnaire in a sample of novice drivers. Australasian Road Safety Research, Policing and Education Conference, Sydney, New South Wales. Hills, B. L. (1980). Vision, visibility and perception in driving. Perception, 9, 183–216. Kline, D. W., Kline, T. J., Fozard, J. L., Kosnik, W., Schieber, F., & Sekuler, R. (1992). Vision, aging and driving: The problems of older drivers. Journals of Gerontology, 47, 27–34. 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. Koustanai, A., Boloix, E., Elslande, P. V., & Bastien, C. (2008). Statistical analysis of “lookedbut-failed-to-see” accidents: Highlighting the involvement of two distinct mechanisms. Accident Analysis and Prevention, 40, 461–469. Lajunen, T., Parker, D., & Summala, H. (2004). The Manchester Driver Behaviour Questionnaire: A cross-cultural study. Accident Analysis and Prevention, 36, 231–238. Lawton, R., Parker, D., Manstead, A. S. R., & Stradling, S. G. (1997). The role of affect in predicting social behaviors: The case of road traffic violations. Journal of Applied Social Psychology, 27, 1258–1276. McGwin, G., & Brown, D. B. (1999). Characteristics of traffic crashes among young, middle-aged, and older drivers. Accident Analysis and Prevention, 31, 181–198. Ozkan, T., & Lajunen, T. (2005). A new addition to DBQ: Positive driver behaviours scale. Transportation Research Part F, 8, 355–368. Ozkan, T., Lajunen, T., & Summala, H. (2006). Driver Behaviour Questionnaire: A follow up study. Accident Analysis and Prevention, 38, 386–395. Parker, D., McDonald, L., Rabbitt, P., & Sutcliffe, P. (2000). Elderly drivers and their accidents: The aging driver questionnaire. Accident Analysis and Prevention, 32, 751–759. Parker, D., Reason, J. T., Manstead, A. S. R., & Stradling, S. G. (1995). Driving errors, driving violations and accident involvement. Ergonomics, 38, 1036–1048. Poulsen, A. A., Horswill, M. S., Wetton, M. A., Hill, A., & Lim, S. M. (2010). A brief officebased hazard perception intervention for drivers with ADHD symptoms. Australian and New Zealand Journal of Psychiatry, 44, 528–534. Reason, J., Manstead, A., Stradling, S., Baxter, J., & Campbell, K. (1990). Errors and violations on the road: A real distinction? Ergonomics, 33, 1315–1332. Regan, M. A., Lee, J. D., & Young, K. L. (2009). Driver distraction: Theory, effects and mitigation. Boca Raton, FL: CRC Press, Taylor & Francis Group. Rimmo, P. -A. (2002). Aberrant driving behaviour: Homogeneity of a four-factor structure in samples differing in age and gender. Ergonomics, 45, 569–582. Stanton, N. A., & Salmon, P. M. (2009). Human error taxonomies applied to driving: A generic driver error taxonomy and its implications for intelligent transport systems. Safety Science, 47, 227–237. Sumer, N. (2003). Personality and behavioral predictors of traffic accidents: Testing a contextual mediated model. Accident Analysis and Prevention, 35, 949–964. U.S. Census Bureau (2012). Motor Vehicle Accidents and Fatalities. http://www.census. gov/compendia/statab/2012/tables/12s1103.pdf ([Data file]. Retrieved from) U.S. Department of Transportation, Federal Highway Administration (2011). Highway statistics 2011: Distribution of licensed drivers-2011 by sex and percentage in each age group and relation to population. http://www.fhwa.dot.gov/policyinformation/ statistics/2011/dl20.cfm ([Data file]. Retrieved from) Wierwille, W. W., Hanowski, R. J., Hankey, J. M., Kieliszewski, C. A., Lee, S. E., Medina, A., et al. (2002). Identification and evaluation of driver errors: overview and recommendations. Final Report for Federal Highway Administration contract DTFH 61-97-C-00051. Washington, DC: Federal Highway Administration. Winter, J. C. F., & Dodou, D. (2010). The Driver Behaviour Questionnaire as a predictor of accidents: A meta-analysis. Journal of Safety Research, 41, 463–470.

S.T.D. Cordazzo et al. / Journal of Safety Research 50 (2014) 99–107

105

Appendix 1. List of items and its changes from the originals.

Reason et al. (1990) wording

Parker et al. (1995) wording

Final item

Reason for the changes

Attempt to drive away from traffic lights in third gear Check your speedometer and discover that you are unknowingly traveling faster than the legal limit Lock yourself out of your car with the keys still inside Become impatient with a slow driver in the outer lane and overtake on the inside

Attempt to drive away from traffic lights in third gear

Attempt to leave a parking space in the wrong gear Check your speedometer and discover that you are traveling faster than the posted speed limit

Revised for clarity (most people drive an automatic transmission) Changed wording slightly for clarity

Become impatient with a slow driver in the outer lane and overtake on the inside

Drive as fast along country roads at night on dipped lights as on full beam Attempt to drive away without first having switched on the ignition Drive especially close or ‘flash’ the car in front as a signal for that driver to go faster or get out of your way Forget where you left your car in a multilevel car park Distracted or preoccupied, realize belatedly that the vehicle ahead has slowed, and have to slam on the brakes to avoid a collision Intend to switch on the windscreen wipers, but switch on the lights instead, or vise versa Turn left onto a main road into the path of an oncoming vehicle that you hadn't seen, or whose speed you had misjudged Misjudge your gap in a car park and nearly (or actually) hit adjoining vehicle ‘Wake up’ to realize that you have no clear recollection of the road along which you have just traveled Miss your exit on a motorway and have to make a lengthy detour Forget which gear you are currently in and have to check with your hand Stuck behind a slow-moving vehicle on a two-lane highway, you are driven by frustration to try to overtake in risky circumstances Intending to drive to destination A, you ‘wake up’ to find yourself en route to B, where the latter is the more usual journey Take a chance and cross on lights that have turned red Angered by another driver's behavior, you give chase with the intention of giving him/her a piece of your mind Try to overtake without first checking your mirror, and then get hooted at by the car behind which has already begun it's overtaking maneuver Deliberately disregard the speed limits late at night or very early in the morning Forget when your road tax/insurance expires and discover that you are driving illegally Lost in thought, you forget that your lights are on full beam until ‘flashed’ by other motorists On turning left, nearly hit a cyclist who has come up on your inside In a queue of vehicles turning left on to a main road, pay such close attention to the traffic approaching from the right that you nearly hit the car in front

Omitted because not related to driving safety. Omitted for repetition

Drive as fast along country roads at night on low beams as you would on high beams

Revised wording for North America (NA) context Omitted for repetition

Drive especially close to the car in front as a signal for that driver to go faster or get out of your way Forget where you left your car in a car park

Switch on one thing, such as the headlights, when you meant to turn on something else, such as the wipers

Realize you have no clear recollection of the road along which you have just been traveling

Drive especially close to or ‘flash’ the car in front of you to try and get them to go faster or get out of your way Forget where you parked your car

Altered the original wording for clarity

Realize that the vehicle ahead has slowed, and have to slam on the brakes to avoid a collision because you were distracted or preoccupied Switch on one thing, such as the headlights, when you meant to turn on something else, such as the wipers Turn left into the path of an oncoming vehicle that you hadn't seen

Changed original wording for clarity

Misjudge the space available in a parking lot and nearly (or actually) hit another vehicle Realize you have no clear recollection of the road along which you have just been traveling Miss your exit on a highway and have to make a detour

Try to pass in risky circumstances when stuck behind a slow-moving vehicle on a two-lane highway Intending to drive to destination A, you ‘wake up’ to find yourself en route to B, perhaps because the latter is your more usual destination

Angered by another driver's behavior, you give chase with the intention of giving him/her a piece of your mind

Revised for NA context and clarity

Wording from Parker et al. (1995)

Revised wording to avoid repetition of other items Revised wording for NA context

Wording from Parker et al. (1995)

Revised wording for NA context Omit because few drivers have manual transmission Revised wording for brevity and clarity

Intending to drive to destination A, you realize that you are actually en route to destination B, perhaps because destination B is your more usual destination Take a chance and run a red light

Revised for clarity

Feel angered by another driver's behavior and chase after him/her with the intention of giving him/her a piece of your mind

Revised wording for clarity

Revised wording for NA context

Omitted for repetition

Disregard the speed limits late at night or early in the morning

Deliberately disregard the speed limit late at night or very early in the morning

Original wording

Omitted because it does not happen often

On turning left, nearly hit a cyclist who has come up on your inside Queuing to turn left onto a main road, you pay such close attention to the main stream of traffic that you nearly hit the car in front

Forget that you have your high beams on until ‘flashed’ by other motorists

Revised for NA context and clarity

When turning right, nearly hit a cyclist who has come up beside you In a line of cars turning left onto a main road, pay such close attention to the main stream of traffic that you nearly hit the car in front

Revised wording for NA context

(continued on next page)

Reworded for clarity

106

S.T.D. Cordazzo et al. / Journal of Safety Research 50 (2014) 99–107

Appendix (continued) 1 (continued) Reason et al. (1990) wording

Parker et al. (1995) wording

Final item

Reason for the changes

Drive back from a party, restaurant, or pub, even though you realize that you may be over the legal blood-alcohol limit Have an aversion to a particular class of road user, and indicate your hostility by whatever means you can Lost in thought or distracted, you fail to notice someone waiting at a zebra crossing, or a pelican crossing light that has just turned red Park on a double-yellow line and risk a fine Misjudge speed of oncoming vehicle when overtaking

Drive even though you realize that you may be over the legal blood-alcohol limit Have an aversion to a particular class of road user, and indicate your hostility by whatever means you can

Drive even though you realize that you may be over the legal blood-alcohol limit

Parker et al. (1995) wording

Hit something when reversing that you had not previously seen Fail to notice someone stepping out from behind a bus or parked vehicle until it is nearly too late Plan your route badly, so that you meet traffic congestion you could have avoided Overtake a single line of stationary or slow-moving vehicles, only to discover that they were queuing to get through a one-lane gap or roadwork lights Overtake a slow-moving vehicle on the inside lane or hard shoulder of a motorway Cut the corner on a right-hand turn and have to swerve violently to avoid an oncoming vehicle Get into the wrong lane at a roundabout or approaching a road junction

Hit something reversing that you had not previously seen

Fail to read the signs correctly, and exit a roundabout on the wrong road

Misread the signs and exit a roundabout on the wrong road

Fail to give way when a bus is signaling its intention to pull out Ignore ‘give way’ signs, and narrowly avoid colliding with traffic having the right of way Fail to check your mirror before pulling out, changing lanes, turning, etc. Attempt to overtake a vehicle that you hadn't noticed was signaling its intention to turn right Deliberately drive the wrong way down a deserted one-way street Disregard red lights when driving late at night along empty roads Drive with only ‘half-an-eye’ on the road while looking at a map, changing a cassette or radio channel, etc. Fail to notice pedestrians crossing when turning into a side-street from a main road Get involved in unofficial ‘races’ with other drivers ‘Race’ oncoming vehicles for a one-car gap on a narrow or obstructed road Brake too quickly on a slippery road and/or steer the wrong way in a skid Misjudge your crossing interval when turning right and narrowly miss collision

Omitted after pilot testing

Fail to notice someone waiting at a crosswalk

Underestimate the speed of an oncoming vehicle when overtaking

Underestimate the speed of an oncoming vehicle when passing on a two-lane highway Hit something when backing up that you did not see Fail to notice someone stepping out from behind a bus or parked vehicle until it is nearly too late

Reworded to accommodate NA context and for clarity

Omit for NA context Changed wording for clarity

Changed wording for clarity Original wording

Omitted because irrelevant to driving safety Omitted because irrelevant to driving safety

Omitted after pilot testing

Omitted after pilot testing

Get into the wrong lane at a roundabout or approaching a junction

Miss ‘give way’ signs, and narrowly avoid colliding with traffic having the right of way Fail to check your rearview mirror before pulling out, changing lanes, etc. Attempt to overtake someone that you hadn't noticed to be signaling a right turn

Fail to notice pedestrians are crossing when turning into a side-street from a main road Get involved in unofficial ‘races’ with other drivers

Brake too quickly on a slippery road and/or steer the wrong way in a skid

Cross a junction knowing that the traffic lights have already turned against you

Get into the wrong lane when approaching an intersection or roundabout

Fail to yield right-of-way to a bus that is signaling its intention to pull out Ignore a yield sign and almost collide with traffic having the right-of-way Fail to check your mirrors before pulling out, changing lanes, turning, etc. On a two-lane road, attempt to pass a vehicle that you hadn't noticed was signaling its intention to turn left Drive the wrong direction down a deserted one-way street Disregard red lights or stop signs when driving late at night along empty roads Drive while looking at a map or GPS device, changing the radio station, etc. Fail to notice pedestrians crossing when turning into a side-street from a main road Get involved in unofficial ‘races’ with other drivers

Brake too hard on a slippery road or steer the wrong way in a skid Misjudge the distance between oncoming vehicles when turning left and narrowly miss a collision

Revised for clarity

Omitted because roundabouts are infrequent in NA and redundant with respect to other items Revised wording for clarity Revised wording for clarity and NA context Original wording (Reason et al., 1990)

Reworded for clarity and NA context

Revised wording to make a S/L/M – would be very rare if deliberate Added wording to make more comprehensive Revised wording for clarity and NA context Original wording (Reason et al., 1990)

Original wording (Reason et al., 1990) Omitted because it is confusing and unclear Revised wording for clarity Revised wording for clarity and NA context Omit for repetition

S.T.D. Cordazzo et al. / Journal of Safety Research 50 (2014) 99–107

107

Sheila T. D. Cordazzo is a Senior Researcher Assistant in the Perceptual and Cognitive Aging Laboratory at the University of Calgary. She obtained her Ph.D. in Developmental Psychology in 2008 from the Federal University of Santa Catarina, Brazil. Currently she is been conducting research concerned with developing and analyzing screening techniques to predict driving performance.

Katherine Bubric is a graduate student at Cornell University, currently pursuing a M.S. in Human Environment Relations with a concentration in Human Factors and Ergonomics. Her current research focuses on the ways in which college students use laptop computers in non-traditional configurations, and the associated postural concerns. She also holds a B.Sc. in Psychology from the University of Calgary, where she was involved in research related to vision, aging and driving.

Charles (Chip) Scialfa has been a professor at the University of Calgary since 1989, where his research program has focused on visual and cognitive aging. He has examined the control of eye movements, acuity and contrast sensitivity, visual search and visual attention. Applied work has examined the behavioral consequences of distraction while driving, Web navigation, automation and driving performance. Since 2007, research has been directed to the prediction of driving performance using brief screening tests, including a North American Hazard Perception Test.

Rachel Jones Ross is a Ph.D. student in Experimental Psychology at the University of Calgary. Her research interests began in Social Psychology, with a focus on unforgiveness. Some of the projects she has worked on include the development and validation of a measure to assess unforgiveness, and developing a theoretical model of how unforgiveness is experienced. More recently her interest in research brought her to the Perceptual and Cognitive Aging Laboratory, where she is researching various aspects of driving, including predicting on-road performance in both healthy and cognitively impaired older adults.