Abstract
When accident data is collected by many people or organizations separately, the collectors may differ systematically in the way they assess injury severity. A statistical method of allowing for this is proposed and applied to a set of road accident data classified according to four sets of circumstances, five police force areas, and three severities of injury. It is concluded that appreciable improvements in the sensitivity of analysis can be obtained by allowing for possible differences in the grading of injury severity.

This publication has 0 references indexed in Scilit: