Abstract
The current methodology for handling repeat victimisation in the Crime Survey for England and Wales (CSEW) is capping. Repeat victimisations for any series of crimes are capped at maximum of 5 crimes before the data is used to produce crime rate estimates for England and Wales. The new proposals merely shift the level of this cap. Capping produces inaccurate estimates of crime which are systematically biased in specific ways, no matter what level the cap is set at.
It is possible to increase the accuracy of crime estimates from the CSEW by deriving them from all reported crimes, and without increasing volatility by utilising three-year rolling averages. A move away from capping to deriving crime estimates based on all reported crimes would increase: relevance, accuracy, clarity, coherence and comparability of crime statistics and would better conform to ONS quality principles. A capping methodology does not conform to these ONS quality principles.
It is possible to increase the accuracy of crime estimates from the CSEW by deriving them from all reported crimes, and without increasing volatility by utilising three-year rolling averages. A move away from capping to deriving crime estimates based on all reported crimes would increase: relevance, accuracy, clarity, coherence and comparability of crime statistics and would better conform to ONS quality principles. A capping methodology does not conform to these ONS quality principles.
| Original language | English |
|---|---|
| Publication status | Published - 31 Aug 2016 |
