Evaluating Product-Related Hazards At the Consumer Product Safety Commission

Multivariate qualitative response models, such as logit regression models, are useful in evaluating product-related risk at the Consumer Product Safety Commission (CPSC). They are used to determine and quantify the factors associated with accidents, and can suggest or justify appropriate regulatory responses. This article applies the method to evaluating risks associated with all-terrain vehicles. A logit regression model is used to analyze data from surveys of users and accident victims. Both driver and vehicle characteristics are found to be associated with accidents, and conditional probabilities of injury are calculated. Policy implications are discussed and suggestions for the further application of this method are provided.