A multimodal data framework for motorcyclist injury severity on rural undivided roads.

Motorcycle crashes on rural undivided roads remain a significant safety concern due to the high incidence of severe injuries and fatalities. This study analyzes a comprehensive dataset of 12,753 motorcycle crashes from rural undivided roads in Texas. Employing a multi-method approach, the research first applies Cluster Correspondence Analysis (CCA) to identify underlying patterns in crash characteristics, followed by the estimation of cluster-based Random Parameter Logit Models. Latent Dirichlet Allocation (LDA) topic modeling of crash narratives further supplements the analysis by uncovering key crash scenarios and thematic trends. The findings indicate that severe and fatal motorcycle injuries on rural undivided roads are primarily driven by high-speed loss-of-control events, particularly run-off-road and overturn crashes occurring on straight and curved segments. Crashes at intersections represent a severity mechanism, where inadequate lighting and turning or yielding conflicts combine to increase injury risk. Additionally, nighttime crashes on rural segments, especially those involving fixed objects or animals, emerge as a distinct high-risk scenario, reflecting the compounded effects of limited visibility, high operating speeds, and reduced reaction time on motorcyclist injury severity. The findings inform a suite of policy interventions grounded in the Safe System Approach (SSA), recommending context-sensitive speed management, rural infrastructure upgrades, helmet use promotion, and improved emergency and wildlife response as essential strategies.