The effect of motorcyclists’ age on injury severities in single-motorcycle crashes with unobserved heterogeneity

•The effects of motorcyclists' age in injury severities are studied.•Florida motorcycle crash data for 2016 is considered for three age groups.•Mixed logit models of injury severity in single-motorcycle crashes are estimated.•Significant differences among three age groups of motorcyclists are found. Introduction: Due to the myriad of unique characteristics associated with motorcycle operation, motorcycle safety is a public health concern as complex as it is serious. National crash data suggest motorcyclists are 28 times more likely to be killed when compared to passenger car occupants. In the state of Florida, motorcycle crashes are 1.5 times more likely to result in the death of the rider, placing Florida among the top deadliest states for motorcyclists in the nation. Using police-reported data from 2016, this study addresses the complex and interconnected nature of the many characteristics associated with motorcycle operation by investigating the effect of age on motorcyclists’ riding behavior as it relates to injury severity for single-motorcycle crashes in the state of Florida. Method: To account for unobserved heterogeneity in the crash data, mixed logit models with heterogeneity in means and variances were estimated to model three injury severity outcomes (non-visible, severe, and fatal) for three age groups (under 30, 30–49, and 50 and above). Results: Model results indicate that age affects motorcyclists’ safety perception and ability to assess risks, thereby influencing their involvement in risky behaviors. Characteristics unique to motorcycle operation—spatial characteristics, speed, motorcycle type, time of day, helmet usage, alcohol consumption, ejection from motorcycle, passenger presence, endorsement status, and lighting—are further complicated by their dependency on the characteristics of the individual motorcyclist. Age of motorcyclist indicates a relationship between motorcyclists’ behavior and perceived safety. Conclusion: The model results indicated that statistically significant parameters constituted different models and they were not equal across the age groups of motorcyclists: aged under 30, aged 30–49, and aged 50 and above. Through advanced econometric modeling, this study fills a gap in the existing literature and assists the safety professionals, motorcycle trainers, policymakers, law enforcement agencies, and roadway designers in developing countermeasures.