Patterns in perceived crash risk among male and female drivers with and without substantial cycling experience

• Published in: Transportation research. Part F, Traffic psychology and behaviour Vol. 69; pp. 1 - 12
• Main Authors: Griffin, Wanda, Haworth, Narelle, Twisk, Divera
• Format: Journal Article
• Language: English
• Published: Oxford Elsevier Ltd 01.02.2020; Elsevier Science Ltd
• Subjects: Accidents; Bicycles; Bicycling; Configurations; Crashes; Dependent variables; Drivers; Experience; Gender; Gender aspects; Gender differences; Hypotheses; Men; Opinions; Perceived risk, perceived skill; Risk analysis; Risk assessment; Risk perception; Traffic accidents & safety; Variance analysis; Women; Perceived risk, perceived skill; Experience; Opinions; Gender; Bicycles
• ISSN: 1369-8478; 1873-5517
• DOI: 10.1016/j.trf.2019.12.013

•Australian drivers and cyclists differ in perceived crash risk in identical situations.•Cyclist’s perceived risk reflects vulnerability in collision with cars.•Driver’s perceived risk reflects awareness of own injury potential.•Female drivers and cyclists do not differ in perceived risk and skill, and opinions.•Skill and opinions are weak predictors of perceived risk, especially for women. Perceived risk of crashes is generally higher in female than in male cyclists. No study has assessed whether these gender differences only apply to riding bicycles. Using an online risk assessment task and survey, this study (a) tested the hypothesis that compared to men, women perceive higher risks, irrespective of traffic mode and (b) explored gender differences in the predictors of perceived risk. Data were analysed from 595 driver licence holders in Australia: 444 who cycled regularly, termed ‘cyclists’ from here on (31% women) and 151 regular drivers who did not (63% women), who are referred to as ‘drivers’. Cyclists in their role as cyclists and drivers in their role as drivers (traffic role) were instructed to imagine being in a specified traffic scenario and to assess crash likelihood. Each of 6 traffic scenarios was presented twice but differing in configuration: a violating car and a violating bicycle. A mixed ANOVA, with rated probability of a crash as the dependent variable, gender and traffic role as grouping factors, and configuration as the repeated measure, resulted in a main effect of configuration, showing that both drivers and cyclists perceived higher levels of risk of violating cars than of violating bicycles (ηp2 = 0.34). The interaction of traffic role and configuration showed this difference to be smaller for drivers than for cyclists (ηp2 = 0.18). This suggests that drivers are aware of the car’s high injury potential in crashes with cyclists. The main effect of gender showed that irrespective of traffic role, women perceived higher risk than men, which supported the hypothesis that gender differences are not limited to cycling (ηp2 = 0.03). The main effect of traffic role was not significant. For the predictors of perceived risk, ANOVA showed main effects for traffic role and gender, with small effect sizes (from ηp2 = 0.02 to ηp2 = 0.03). Cyclists were more confident, less inattentive and more in favour of stricter laws. Women felt less skilled and were more concerned about risks. None of the interactions between gender and traffic role were significant, which lent further support to the hypothesis that gender differences were a general rather than a cycling specific characteristic. Thus, perceived risk does not differ between women who regularly cycle and those who do not. These findings may inform interventions about the role of perceived risk in the uptake of cycling.