How to Interpret Local Crime Statistics

Crime statistics are collected and published by police services and government agencies. They are a useful starting point, but reading them correctly requires understanding what they do and do not measure.

Reported vs actual crime

Published figures reflect crimes reported to and recorded by police. Some crime types (for example certain sexual offences and domestic incidents) are known to be under-reported relative to their actual occurrence. Comparing raw totals across crime types can be misleading for this reason.

Absolute numbers vs rates

A community with a larger population or higher footfall (through commercial areas, transit hubs, or event venues) will often show higher absolute crime counts without being meaningfully less safe on a per-resident basis. Where available, a rate per population or per household is generally more comparable across areas than a raw total.

Geographic boundaries

Crime is usually reported against defined boundaries (a police beat, a postcode, a census area) that may not match how residents think of a neighbourhood. A boundary can include a busy retail strip alongside a quiet residential street, blending very different experiences into a single average.

Trends over time

A single year's figure can be affected by a small number of unusual incidents or by changes in how crimes are classified or recorded. A multi-year trend is generally more informative than a single year in isolation, and a genuine change in classification methodology (which does happen periodically) can create a step change in the data that has nothing to do with actual crime levels.

Questions worth asking

• What time period does this figure cover?
• Is this a total count or a rate?
• Has the reporting boundary or methodology changed recently?
• How does this compare to the wider city or region, not just to a single neighbouring area?