Safetlas

Methodology & sources

Safetlas shows officially recorded crime - data published by national police forces and statistics offices. No scraped news, no user guesses, no vibes.

The rules we follow

1. Rates, not raw counts. Everything is crimes per 1,000 residents, so a big city and a small town are comparable.
2. Compare within one country only. Countries define and record crime differently; colours and rankings never cross borders. The only cross-country layer is survey-based perception (Numbeo) and the Georgetown WPS women's safety index, clearly labelled.
3. Grey never means unsafe. It means no reliable official data.
4. Tourist-area caveat. Per-resident rates inflate in places with many visitors and few residents (city centres, ports, station districts). We say so wherever it applies.

Sources by country

1. England & Wales: Home Office police recorded crime (CSP open data); street-level incidents from data.police.uk for neighbourhood (LSOA) view; ONS population & boundaries.
2. Scotland: Recorded Crime in Scotland (gov.scot), council areas, own scale.
3. Netherlands: data.politie.nl registered crime per neighbourhood; CBS population & boundaries.
4. France: SSMSI communal crime base (data.gouv.fr), incl. Paris/Lyon/Marseille arrondissements; Etalab boundaries.
5. Ireland: CSO recorded crime by Garda division (statistics under reservation); Census population.
6. Denmark: Statistics Denmark STRAF11 per municipality; Dataforsyningen boundaries.
7. Czechia: Police of the Czech Republic open incident data (2.5M geocoded incidents), aggregated to districts; CZSO population.
8. Europe layer: Numbeo Safety Index (perception) and Georgetown WPS Index (women's safety), country level.

Open the atlas →