Skip to main content

GeraLearn / SEN Provision / Methodology

Gera Special Educational Needs Provision Index — Methodology

How GeraLearn computes the GSNEPI from real, publicly available Department for Education data. Fully reproducible. No numbers are modelled or invented.

The short answer

The Gera Special Educational Needs Provision Index (GSNEPI) combines three real DfE measures — EHCP rate, SEN support rate, and specialist school places per 1,000 pupils — each indexed to the England national average and averaged. Knowsley ranks highest at GSNEPI 120.5. Source: DfE SEN in England, 2024/25, OGL v3.0.

Data as of Academic year 2024/25. Licensed under Open Government Licence v3.0.

Data source

Publication
DfE — Special educational needs in England, Academic year 2024/25
Publisher
Department for Education (DfE)
Coverage
All 153 English local authorities (state-funded schools, Academic year 2024/25)
File used
data/sen_phase_type_.csv — rows where geographic_level = Local authority, time_period = 202425, hospital_school = No, type_of_establishment = Total
Licence
Open Government Licence v3.0 (OGL v3.0) — free to reuse with attribution

Index computation (GSNEPI)

The Gera Special Educational Needs Provision Index (GSNEPI) combines three DfE-published measures into a single number, where 100 represents England average levels across two of the three components. The index is computed separately for each of England's 153 local authorities.

Step 1 — EHCP Rate Component

ehcp_comp = min( (LA_ehcp_pct / 5.339) × 100, 200 )

LA_ehcp_pct is the percentage of pupils in the local authority with an Education, Health and Care Plan (EHCP) in 2024/25 (DfE sen_phase_type_.csv, ehc_plan_percent). The England national figure is 5.34%. A cap of 200 prevents extreme outliers from dominating (no LA exceeds this cap in practice).

Step 2 — SEN Support Rate Component

sen_comp = min( (LA_sen_pct / 14.218) × 100, 200 )

LA_sen_pct is the percentage of pupils receiving SEN support (without an EHCP) in the LA (sen_support_percent). England national average: 14.22%.

Step 3 — Specialist Places Component

spec_comp = (LA_specialist_per_1000 / 34.997) × 100

LA_specialist_per_1000 is the number of pupils enrolled in state-funded special schools in the LA, divided by total LA pupils, multiplied by 1,000 (a per-1,000 rate). The divisor 34.997 is the maximum value across all 153 LAs (Knowsley), so this component is on a 0–100 scale. LAs with no state-funded special school returns score 0 on this component.

Final score

GSNEPI = round( (ehcp_comp + sen_comp + spec_comp) / 3, 1 )

The three components are averaged and rounded to one decimal place. GSNEPI above 100 means the LA has higher combined SEN provision intensity than the England average across EHCP rate and SEN support rate. GSNEPI below 100 means lower.

Important: The GSNEPI measures SEN provision intensity (how many pupils are identified as having SEN and/or placed in specialist settings), not SEN need or educational outcomes. Higher GSNEPI reflects more intensive formal SEN identification and specialist provision, which can reflect both greater need and more active identification policy.

Reproducibility

Every number on this site is derived directly from the DfE Special educational needs in England publication. To reproduce any figure:

  1. Download the data from explore-education-statistics.service.gov.uk
  2. Open data/sen_phase_type_.csv and filter for time_period = 202425, geographic_level = Local authority, hospital_school = No, type_of_establishment = Total
  3. Apply the GSNEPI formula above using the national averages: EHCP = 5.34%, SEN support = 14.22%

Explore SEN provision by local authority

See the GSNEPI and underlying DfE figures for all 153 English local authorities.

View all local authorities →

Source: DfE — Special educational needs in England, Academic year 2024/25. Licensed under the Open Government Licence v3.0. Contains public sector information licensed under the Open Government Licence v3.0. The Gera Special Educational Needs Provision Index is computed by GeraLearn from these figures; no figure is modelled, estimated, or interpolated.