House prices are built from the latest available month in the checked-in UK House Price Index extract, then mapped from local-authority level into constituencies. England, Wales, and Scotland use Output Area population-weighted LAD-to-constituency apportionment, Northern Ireland uses postcode-share apportionment, and the published value is the resulting constituency-level weighted average price rounded to pounds. There is no standalone rent feature in the current build.

This is computed from constituency centroids and a checked-in airport catalogue filtered to medium and large UK airports only. For each constituency, HomeDive calculates the geodesic distance to every qualifying airport and keeps the nearest value in kilometres, so lower values indicate closer airport access by straight-line distance rather than a door-to-door travel-time estimate.

This metric uses constituency centroids and cached OpenStreetMap hospital points built from a single UK-wide Overpass query. HomeDive pulls features tagged as `amenity=hospital` or `healthcare=hospital`, converts mapped nodes, ways, and relations into point locations, deduplicates them, and stores the nearest geodesic distance in kilometres for each constituency, so lower values indicate closer hospital access by straight-line distance rather than hospital size, capacity, or waiting-time performance.

Safety level starts with a country-specific weighted crime index converted into constituency rates per 1,000 residents, so violent and sexual offences count more heavily than lower-severity property or nuisance offences. England and Wales use spatially joined geocoded incidents, Scotland and Northern Ireland are apportioned from area-level totals into constituencies using population or postcode-share intersections, partial-year Scottish source data is annualised to a 12-month equivalent, and HomeDive then converts the resulting rates into a UK-wide safety index and percentile-based HomeDive score so higher values indicate relatively safer constituencies.

School performance is built from checked-in national school tables for England, Scotland, Wales, and Northern Ireland that have been normalised to a shared schema of school, constituency, cohort size, and raw attainment score. HomeDive converts school results into within-nation percentile scores, aggregates them into cohort-weighted constituency averages, and fills any remaining gaps from the nearest scored constituency in the same nation.

Transport links are built from active NaPTAN access points in Great Britain and checked-in official Translink bus stop and station datasets for Northern Ireland, all spatially joined to constituency boundaries. Stop types are weighted so rail, metro, platform, coach, and interchange access count more than standard bus stops, airport and ferry nodes are kept at zero to preserve UK-wide comparability, and the final weighted total is divided by constituency area in km2.

GP access is estimated from OpenStreetMap records tagged as doctors’ surgeries, fetched via Overpass and counted from their mapped points or returned centres. Records are spatially assigned to constituencies, counted, and scaled per 10,000 residents using constituency population totals, so higher values indicate more mapped GP surgery locations per resident base.

Broadband quality is a composite index rather than a direct average wifi speed figure. HomeDive takes seven fixed-line coverage and under-service indicators at constituency level, including Gigabit, UFBB, Full Fibre, and SFBB availability plus key under-threshold measures, standardises them, extracts the first principal component, and rescales the result to 0-100 so higher values reflect stronger overall broadband availability and fewer under-served premises.

Air quality combines PM2.5, NO2, and O3 gridded measurements from the DEFRA PCM dataset. Each pollutant is spatially joined to constituencies and averaged, then HomeDive standardises the three constituency averages, extracts a single principal component, rescales it to 0-100, and inverts the score so higher values represent cleaner air conditions.

Accessible green spaces are built from cached OpenStreetMap Overpass geometry pulls covering parks, recreation grounds, public gardens, protected green land, forests, meadows, heath, grassland, and scrub. HomeDive rebuilds polygons from ways and relations, filters out polygons smaller than 2 hectares, derives one residential sample point per Output Area and constituency slice from active residential postcodes, weights those sample points by Output Area population where census estimates are available, falls back to residential postcode counts for Northern Ireland, and stores the weighted constituency median distance in metres, so lower values mean residents are, on average, closer to mapped green space.

This combines police-recorded sexual offences and violence categories into a constituency-level rate per 1,000 residents. England and Wales are built from geocoded incidents spatially joined to constituencies, while Scotland and Northern Ireland are apportioned from published area-level totals into constituencies using Output Area population shares or postcode-share splits; where the latest Scottish source is a partial year, HomeDive annualises it to a 12-month equivalent so the published rate remains comparable across the UK.

School density is now built from a cached Great Britain OpenStreetMap Overpass query for features tagged amenity=school rather than the earlier register-based approach. Nodes use their mapped coordinates, ways and relations use returned centre points, and HomeDive then counts those mapped school sites by constituency and normalises by population to produce schools per 10,000 residents.

Pharmacy density is derived from OpenStreetMap records tagged as pharmacies, fetched via Overpass. After constituency assignment by spatial join, HomeDive counts those mapped locations and normalises the total per 10,000 residents, so higher values indicate more mapped pharmacies relative to constituency population.

Dentist access is estimated from OpenStreetMap records tagged as dental practices, queried through Overpass. Locations are mapped to constituencies, counted from their points or returned centres, and normalised per 10,000 residents, so higher values indicate denser mapped dental practice availability relative to population.

Restaurant density is estimated from OpenStreetMap records tagged as restaurants using a cached Great Britain Overpass query. HomeDive counts those records by constituency and scales the result per 10,000 residents, so higher values indicate more mapped restaurants relative to the resident population. Pubs are not a standalone feature in the current build.

Live music venue density is sourced from a cached Great Britain OpenStreetMap Overpass query combining dedicated music venues, places tagged live_music=yes on bars, pubs, nightclubs, or cafes, and theatres tagged as concert halls. HomeDive counts those mapped venues by constituency and normalises the result per 100,000 residents, so higher values indicate more recorded live music venues relative to population.

Cinema density uses a cached Great Britain OpenStreetMap Overpass query for features tagged as cinemas. HomeDive maps those venues to constituency polygons, counts them from their points or returned centres, and normalises the result per 100,000 residents, so higher values indicate more cinema venues relative to constituency population rather than screen count or capacity.

Bar density uses OpenStreetMap records tagged as bars, queried through Overpass and counted from their points or returned centres. Features are spatially assigned to constituencies and converted into counts per 10,000 residents, so higher values indicate more mapped bars relative to local population.

Library density uses OpenStreetMap records tagged as libraries. HomeDive spatially joins these mapped locations to constituencies, counts them from their points or returned centres, and scales the total per 10,000 residents, so higher values indicate greater mapped library provision relative to local population.

Museum density is now built from a cached Great Britain OpenStreetMap Overpass query for tourism=museum. HomeDive clusters nearby mapped points within 250 metres to reduce double-counting of the same site, then counts those museum sites by constituency and scales the result per 100,000 residents, so higher values indicate greater mapped museum density relative to local population.

Art gallery density is sourced from OpenStreetMap via a cached Great Britain Overpass query covering records tagged as tourism galleries, arts centres, and gallery shops. Features are spatially joined to constituencies, counted from their mapped points or returned centres, and scaled per 100,000 residents using constituency populations, so higher values indicate a denser spread of mapped gallery-type venues relative to population.

Sports facility density combines OpenStreetMap records tagged as sports centres, stadiums, and sports halls. HomeDive counts these mapped records within constituency boundaries and scales them per 10,000 residents, so higher values indicate greater mapped sports infrastructure density relative to population. The current feature is broader than a gyms-only measure.

LGBT+ venue density now uses a cached Great Britain OpenStreetMap Overpass query for venues tagged lgbtq=primary or lgbtq=welcome where the amenity is a bar, pub, nightclub, or music venue. HomeDive maps those venues to constituencies, counts them, and normalises the result per 100,000 residents, so higher values indicate more explicitly tagged LGBT+ venues relative to population.

Places of worship density uses OpenStreetMap records tagged as places of worship. Constituency counts are generated through spatial joins and then normalised per 10,000 residents, so higher values indicate more mapped places of worship relative to local population rather than differences in congregation size or use.

Post office density uses OpenStreetMap records tagged as post offices. HomeDive maps these records to constituencies, counts them from their points or returned centres, and expresses the total per 10,000 residents, so higher values indicate greater mapped post office density relative to population.

Club density is based on OpenStreetMap records tagged as nightclubs. After spatial assignment to constituency polygons, HomeDive counts those mapped venues and normalises the result per 10,000 residents, so higher values indicate more mapped nightclub venues relative to population.