When using crash data to study the influence of roof strength on injury, it is essential to consider all factors that might affect the likelihood of injury. For example, if it is known that rural rollovers tend to be more severe than urban rollovers, then the analysis should control for rural vs. urban because some vehicles may have more rural crashes than others. The statistical study that considers these extraneous influences is more accurate than the study that ignores them. Unfortunately, the current paper by the Insurance Institute for Highway Safety (IIHS), "Roof Strength and Injury Risk in Rollover Crashes" (IRCOBI Conference, 2008) (see ITRD E144233), ignores many of the important confounding factors that would influence their results. When these errors are corrected, the IIHS data are consistent with the results of prior research: that there is no relationship between roof strength-to-weight ratio and serious injury/fatality likelihood for vehicles which comply with Federal Motor Vehicle Safety Standard (FMVSS)216. For the covering abstract see ITRD E144229.
Samenvatting