No abstract

In Australia, the number of collisions at level crossings (LCs) remains stable and even increased between 2003 and 2011. A consistent number of crashes between trains and pedestrians compared to a decreasing number of crashes between trains and vehicles calls for a better understanding of the factors underpinning pedestrians' decision-making at such risky intersections. This study reports on an innovative method used to measure pedestrians' crossing intentions and risk perceptions in crossing scenarios recorded at three representative LCs for Queensland (Australia). Predictors of crossing intentions were examined and compared between five scenarios. An online survey featuring the footages of the crossing scenarios was completed by 222 participants. The results revealed that pedestrians' intentions to transgress depend on the situation, justifying the use of such methodology. Pedestrians perceived the lowest risk and reported the highest likelihood of risk-taking after the pedestrian lights are active, but before the pedestrian gates have started moving, confirming that pedestrian gates are an efficient safety measure. It was found that while pedestrians perceive others as more likely to take risks at LCs, they do not associate others' behaviour with higher risk of fatal consequences, suggesting that the risk of being hit by a train at LCs as a pedestrian is generally perceived as low. Past unsafe behaviour, descriptive norms and perceived risk of being involved in a crash were found to be consistent predictors of the likelihood to engage in risk-taking behaviour at LCs across the different scenarios.Implications for future research and strategies to reduce habitual unsafe behaviour are discussed.

No abstract

The number of pedestrian victims at Australian and foreign level crossings has remained stable over the past decade and it continues to be a significant problem. To examine the factors contributing to pedestrians' unsafe crossing behaviours, direct observations were conducted at three black spot urban level crossings in Brisbane for a total of 45 h during morning and afternoon peak. In total, 129 pedestrians transgressed the active controls. More transgressions were observed at the crossings located in more populated suburbs in close proximity to large shopping centres and school zones, whereas the smallest number of transgressions were observed at the least populated locations. In addition to characteristics associated with the larger socio-economic area, the patterns of transgression could be associated with the properties of the existing safety equipment and the design of each level crossing (i.e. location of the platforms, number of rail tracks). Indeed, the largest number of crossed unoccupied but ''at risk'' rail tracks (where a train could have passed), was observed at the crossing with the least transgressions. Contrary to previous findings, younger adults were the most frequent transgressors. School children and elderly were most likely to transgress in groups. Potential directions for future research and more effective measures are discussed.