Investigating Urban Landscapes: OS Maps and Databases

Combining evidence about urban places

An urban investigation needs to distinguish what land is used for from the conditions experienced by people living there. An Ordnance Survey (OS) map helps you examine the physical layout of a city. Crime and deprivation databases provide different evidence about conditions within it. Used together, they help investigate whether problems are widespread or concentrated in particular neighbourhoods.

Identifying land use from an OS map

Land-use categories include residential, commercial, industrial, transport and recreational uses. Residential land provides housing; commercial land supports activities such as shopping and offices; industrial land supports production and related activities. Transport land supports movement, while recreational land provides opportunities for leisure. A place can contain several uses rather than fitting neatly into one category.

Begin by locating the study area and checking the map's scale and key. OS Explorer maps have a scale of 1:25,000: one unit on the map represents 25,000 of the same units on the ground. For example, 1 cm represents 250 m. OS Landranger maps are at 1:50,000 and include some different symbols, so use the key belonging to the map you are examining.

Read labels and symbols alongside the arrangement of buildings, roads and open spaces. The key establishes what a symbol represents; the surrounding pattern helps you interpret the likely land use.

For example, in a fictional practice description, an area contains many small buildings arranged along local streets, another contains large buildings labelled as works, and a third contains playing fields. These clues support residential, industrial and recreational interpretations respectively. The label provides stronger evidence of industrial use than building size alone: a large building could have several different functions.

To transfer this reasoning to an actual OS map, identify the relevant features using that map's key, then record the land-use category and the evidence supporting it. Compare several parts of the city using the same categories. This makes a conclusion such as ‘residential use dominates this section’ more defensible than a list of isolated features. Where the function remains uncertain, retain that uncertainty rather than assigning a use from appearance alone.

Investigating crime using police data

Make the investigation specific: is recorded crime concentrated in a few inner-city neighbourhoods, or spread across the study area? Define the study boundary before selecting data, so that the comparison does not change halfway through.

data.police.uk publishes monthly records of crime and anti-social behaviour for England, Wales and Northern Ireland. Its custom CSV download allows you to obtain a selected portion of the data rather than downloading the whole national archive. A CSV is a table that can be opened in a spreadsheet.

Choose a common period and crime category for all the areas being compared. For example, you could investigate the category ‘Violence and sexual offences’ during one selected month. Record the category exactly: it combines offences and should not be treated as a count of just one kind of violence.

In the downloaded table, use the Month, Crime type and LSOA code fields to select and group the records. Count the records for the chosen crime category in each neighbourhood. Keep anti-social behaviour separate from crime rather than automatically adding the two together. Record the download date as well as the period investigated, because published records can be corrected or refreshed.

The published LSOA code refers to the neighbourhood containing the anonymised map point, not necessarily the exact location of the incident. Before combining crime and deprivation tables, check that their LSOA codes and boundary versions match. Neighbourhood names alone are not a reliable way to match records.

A table helps compare counts; a map helps examine their spatial distribution. Look for clusters, differences between neighbouring areas and exceptions to the overall pattern. A higher count does not automatically mean a higher risk for an individual: neighbourhoods may differ in population and in the number of people visiting them. State clearly when you are comparing counts rather than population-adjusted rates.

What a crime map can and cannot show

Published crime locations are approximate. To protect privacy, incidents are moved to nearby anonymous points, such as the centre of a street or a public or commercial place. Several incidents can therefore appear at the same point without having occurred at the same address. Use mapped concentrations to investigate a broader area, not to identify a particular household or building as the location of crime.

These data describe police-recorded incidents, not every crime that occurs. Location quality varies, some records lack usable locations, and temporary gaps can occur when forces have problems supplying data. Check the database's current changelog for issues affecting your selected place and period. These limitations affect how confidently you can interpret an apparent cluster or difference between areas.

Understanding deprivation data

Deprivation is broader than low income. England's Index of Multiple Deprivation (IMD) combines seven domains: income; employment; education, skills and training; health and disability; crime; barriers to housing and services; and living environment. These domains have different weights, so the overall index is not a simple count of problems.

The English indices of deprivation 2025 describe 33,755 Lower-layer Super Output Areas (LSOAs). These statistical neighbourhoods have broadly similar populations, averaging approximately 1,500 residents. Neighbourhood-level data can reveal differences hidden by a summary for an entire city. The English IMD should not be treated as a UK-wide ranking.

IMD ranks run from 1, the most deprived neighbourhood, to 33,755, the least deprived. This is a relative ordering within England, not a measurement of how many times more deprived one place is than another.

Deciles make the ordering easier to interpret. Decile 1 contains the most deprived 10% of neighbourhoods nationally; decile 2 contains the next 10%; decile 10 contains the least deprived 10%. There is no single definitive boundary separating all ‘deprived’ places from all other places.

Working through a neighbourhood comparison

Use the Local Deprivation Explorer to look up neighbourhoods, or filter the published spreadsheet for the relevant local authority. Record each LSOA's identifier, IMD rank and decile. Keep the release year consistent throughout the comparison. Match these results to your selected crime data using compatible neighbourhood identifiers.

The following is a practice dataset. The four fictional neighbourhoods form the whole selected study area. Crime figures are counts of records in the ‘Violence and sexual offences’ category for April 2026; the deprivation column represents IMD 2025 deciles.

A and B fall within the most deprived 20% of neighbourhoods in England. Two of the four selected neighbourhoods therefore meet that chosen threshold. This describes the extent of relative deprivation among the selected areas; it does not mean that half of their residents are individually deprived.

A has the highest recorded crime count and C the lowest. However, D has more crime records than B despite being less relatively deprived. The comparison therefore does not show a simple pattern in which every more-deprived neighbourhood has more recorded crime. Without a spatial layout, the table also cannot establish whether the higher counts form a geographical cluster.

To investigate deprivation further, examine the individual domains. An overall IMD result identifies relative multiple deprivation, whereas a domain result helps distinguish the particular dimension being investigated, such as employment or living environment.

Drawing a careful conclusion

Bring the evidence together by location: describe the land-use pattern, compare the crime figures and identify the distribution of deprivation. State whether the evidence suggests widespread problems or concentrations in particular neighbourhoods, using values and locations to support the conclusion. Keep the conclusion tied to the study boundary, crime category, period and deprivation release used.

Remember that crime is already one component of IMD, so the two measures are not completely independent. Their association cannot establish that deprivation causes crime. Likewise, a neighbourhood's result cannot describe every household within it.

For change over time, check the release and geographical boundaries carefully. The 2025 deprivation indices use changed datasets, methods and geographies, making their results less directly comparable with earlier releases. A change in relative rank alone is not proof that living conditions have improved or worsened.