A family compares two flats. On an online map the district of the first registers a lower number of recorded offences than the independent city of the second. The sentence quickly arises: “The first flat is safer.” However, the map has examined neither these two flats nor the family’s personal routes.
The following fact check concerns Germany. The example is fictional. It makes no statement about the safety of any actual place.
First note the exact claim
The family wants to know whether the daily route home and the neighbourhood suit them. The map found initially answers a different question: How many designated recorded cases are reported for a particular area and period?
These two questions can be related, but they are not identical. A statement about an entire district is not a measurement of the route from a particular stop to a particular front door.
As its own reading method the family therefore rewrites the original sentence: “For this area this source reports the designated figure.” Only after that does it check what of that is helpful for its actual housing question.
Distinguish recorded incidents from actual incidents
The Federal Criminal Police Office explains the limit of the Police Crime Statistics: what is recorded is police-known incidents; offences that do not come to light lie in the dark field (Dunkelfeld). Among other things, reporting behaviour, police control and statistical recording can influence the recorded figures. [PKS]
A change can therefore not be explained solely as an actual increase or decrease in all offences without further basis. That does not mean the statistics would be useless. It means keeping its specific question and its limits visible.
In the example the family does not decide on the basis of a single coloured value. It notes what was actually recorded, which period the representation covers and whether a comparison uses the same definition.
A figure per population is not a personal probability
The Deutschlandatlas displays the designated offences regionally per 100.000 inhabitants and names the Police Crime Statistics as its data basis. [ATLAS] This reference does not turn an area figure into a personal probability of becoming a victim of an offence.
An invented calculation example already shows this: 500 recorded cases with 10.000 inhabitants correspond arithmetically to 5.000 cases per 100.000 inhabitants. That does not mean that five percent of the inhabitants were personally affected. Cases and different affected persons are different units.
The total number also does not yet say which offences are relevant for the concrete question. A blanket sum is not presented as direct evidence for the likelihood of a residential burglary.
Do not unknowingly shrink the spatial reference
The designated atlas representation uses circles as the geographic basis. [ATLAS] A ranking of individual streets derived from this would claim a smaller study area that this basis does not provide.
In the fictitious situation one flat is on the edge of a large district. The other is in a particular neighbourhood of an independent city. The family recognises that the two area figures cannot be read like two measured doorstep environments.
Anyone researching further should use only actually available and suitably defined local information. Missing small-scale data must not be replaced by assumptions about people or population groups.
Keep personal orientation as your own task
The family plans an ordinary route from the stop to the house at a time suitable for their everyday life. They consider lighting, wayfinding, access and the actual surroundings. That is a practical personal observation, not a criminological statistical investigation.
Subjective unease, a pleasant viewing impression and an official statistic are noted as different pieces of information. None of these individual observations provides a guarantee for future events.
For concrete safety questions appropriate professional advice is obtained. A general article about a residential location should not pretend to provide a personal risk assessment that it cannot deliver.
An editorially justifiable location statement
In the invented case what remains in the end is limited information: the source describes recorded cases for certain areas and periods. The family uses it as one building block and answers its own housing question with further appropriate information.
A real-estate company should also keep this distance. It can explain a comprehensible regional figure. It should not turn that into an absolute safety guarantee or an unfounded devaluation of a street or its residents.