Gera Liveability Score — Methodology

1. The six inputs

• HM Land Registry — HM Land Registry — UK House Price Index (April 2026).
• Ministry of Housing, Communities and Local Government — MHCLG — Council Tax levels set by local authorities in England (2026-27).
• Ofcom — Ofcom — Connected Nations (July 2024).
• Ofsted — Ofsted — State-funded school inspections and outcomes (31 May 2026).
• Home Office — Home Office — Police recorded crime open data (year ending December 2025).
• NHS England — NHS England — Appointments in General Practice (April 2026).

All inputs are published under the Open Government Licence v3.0.

2. How the datasets are joined

• House price, council tax, broadband join directly on the ONS GSS local-authority code (E06/E07/E08/E09).
• Ofsted joins on the normalised local-authority name (Ofsted publishes no GSS code).
• Crime is published at Police Force Area (PFA) level. Each local authority inherits the recorded-crime rate of its containing PFA via a fixed geographic crosswalk — a real published figure for that force, applied at the coarser PFA granularity. This is stated on every area page.
• GP access is published at NHS Integrated Care Board (ICB) level. Each local authority inherits the Gera GP Access Score of its containing ICB via a fixed geographic crosswalk, also stated on every page.

A local authority is included in the index only when all six inputs are present for it. No missing value is silently treated as zero.

3. Coverage and exclusions

The index covers 130 single-tier English local authorities — unitary authorities, metropolitan districts and London boroughs — where every input exists at a 1:1 local-authority granularity. Two-tier county areas (for example Surrey, Kent, Essex) are deliberately excluded: their house-price and council-tax figures are published only at the lower-tier district level and cannot be joined 1:1 to the upper-tier Ofsted, crime and GP figures without aggregation. Excluding them keeps every score directly comparable rather than mixing geographic levels.

4. Normalisation

Each pillar is normalised 0-100 using min-max scaling across the 130 included local authorities. For higher-is-better inputs (broadband, GP access):

sub = 100 × (value − min) / (max − min)

For lower-is-better inputs (house price, council tax, crime rate) the scale is inverted:

sub = 100 × (max − value) / (max − min)

The schools sub-score is already a 0-100 percentage (% of state schools rated Good or Outstanding by Ofsted) and is used directly. Affordability blends house price (60%) and Band D council tax (40%), both lower-is-better.

5. Weights

GLS = 0.3·affordability + 0.2·safety + 0.2·schools
      + 0.15·connectivity + 0.15·healthcare

These are the default weights. On any index page you can re-weight the five pillars yourself with the personalisation tool to get a ranking specific to your priorities.

6. Worked example — Stoke-on-Trent

Stoke-on-Trent (West Midlands, ONS code E06000021) has these real source figures:

• Average house price £152,101 · Band D council tax £2,183/yr → affordability sub-score 72.2
• Recorded crime 73.3 per 1,000 (Staffordshire police) → safety sub-score 96.0
• 88.1% of schools Good+ (42 graded) → schools sub-score 88.1
• Gera Broadband Index 87.0 → connectivity sub-score 79.5
• Gera GP Access Score 94.0 (Staffordshire and Stoke-on-Trent ICB) → healthcare sub-score 94.0

GLS = 0.3×72.2 + 0.2×96.0 + 0.2×88.1
      + 0.15×79.5 + 0.15×94.0
    = 84.5 / 100  (rank 1 of 130)

7. Freshness and limitations

• Re-computed quarterly as the underlying datasets refresh; each page carries the as-of date and last-computed date.
• Crime and GP access are area-level statistics applied at PFA / ICB granularity — they describe the wider force or ICB area, not the individual local authority in isolation.
• Recorded-crime figures are neutral official statistics that reflect reporting and recording practice and an area’s daytime / visitor population; they are presented as context, never as a measure of how safe a place is.