ons-nomis

name	ons-nomis
description	Query Nomis — ONS's free Census 2021 + labour-market data service — at constituency / LAD / MSOA / LSOA / OA granularity. Use when the question is about the population, employment, claimant count, occupation, qualifications, ethnicity, age, health, or any other Census 2021 / labour-market table for a specific Westminster constituency, council area, or ward, or for by-constituency comparisons across the UK. Coverage spans ~1,600 datasets including the full Census 2021 (TS-series) and the ONS labour-market series (claimant count, BRES, ASHE, APS). Anonymous SDMX-JSON; no auth.
license	Open Government Licence v3.0 (Crown copyright; ONS via Nomis at the University of Durham)
metadata	{"facility":"ons-nomis","cli-alias":"nomis","base-url":"https://www.nomisweb.co.uk/api/v01","provenance":{"tier":3,"operator":"Office for National Statistics (Nomis, University of Durham)","service":"nomisweb.co.uk","upstream-data":"ONS Census 2021 + labour-market series + DWP claimant data","citation-short":"via Nomis (ONS)","citation-formal":"Nomis, Office for National Statistics, retrieved {date}","confidence":"authoritative","confidence-notes":"Underlying tables are ONS-published official statistics. The Nomis service itself is hosted by the University of Durham under contract to ONS. The per-row 'value' is what ONS publishes."}}

Nomis — ONS Census + labour-market data

Base URL: https://www.nomisweb.co.uk/api/v01

Free SDMX-JSON API to ~1,600 datasets at Westminster-constituency (WMC) and finer granularities. The canonical UK source for "who lives here, what do they do" questions, joinable to Parliament data via ONS GSS codes.

Headline datasets

Dataset id	What it is
NM_2021_1	Census 2021 main residents table (TS001 — usual residents in households / communal establishments).
NM_2021_*	Other Census 2021 TS-series tables (TS002 = household composition, TS003 = age, TS021 = ethnicity, …). Each gets its own NM_2021_<n> id.
NM_1_1	Jobseeker's Allowance — claimant count with rates and proportions.
NM_17_5	Annual Survey of Hours and Earnings (ASHE) — workplace-based.
NM_142_1	Annual Population Survey (APS) — labour market by area.
NM_189_1	Business Register and Employment Survey (BRES).

parl nomis datasets lists every one.

The geography-filter dance (read this)

Nomis indexes data by its own internal geography ids and "type" codes, NOT by bare ONS GSS codes. A bare geography=E14001063 filter often returns zero observations, even for a valid Aldershot record. Two reliable patterns:

1. All geographies of a type (the typed query): geography=2092957699TYPE172 means "every Westminster Parliamentary Constituency 2024 within the UK parent (2092957699)". Returns up to ~650 rows.

2. One geography by Nomis internal id (NOT GSS): obtained from the geography codelist for the dataset.

To discover the right TYPE number for the boundary you want, run:

parl nomis geography-types NM_1_1

Look for the TypeCode annotation. For 2024 Westminster constituencies the answer is TYPE172. Census 2021 datasets use the same convention. The type numbers are stable across datasets that publish the same geography.

CLI

# Discovery
parl nomis datasets                                  # 1,600+ datasets
parl nomis dataset-def NM_2021_1                     # dimensions & codelists
parl nomis codelist NM_2021_1 geography              # what geographies are valid?
parl nomis geography-types NM_1_1                    # TYPE codes the dataset is indexed by

# Data
parl nomis data NM_1_1 \
    --geography 2092957699TYPE172 \                  # every WMC 2024
    --measures 20100 \                               # measure "Value"
    --date latest \
    --take 650

# CSV instead of SDMX-JSON
parl nomis data NM_2021_1 --geography 2092957699TYPE172 --measures 20100 --format csv

Worked chain

parl mapit postcode SW1P3JA              # → constituency name "Cities of London and Westminster", GSS E14001172
parl nomis geography-types NM_2021_1     # find TYPE172 = Westminster 2024
parl nomis data NM_2021_1 \
    --geography 2092957699TYPE172 \
    --measures 20100 \
    --date latest                        # → Census 2021 totals for every constituency

Common dimension filters

Dimension	Meaning
geography	The area code or <parent>TYPE<n> typed query
measures	What metric (20100 = "Value", others vary)
date	Time period — latest, 2024, 2024-12, or comma-separated
c_age, c_sex, c_industry, …	Census / survey breakdowns — vary per dataset
select	Comma-separated list of fields to project (geography_name,obs_value,…)
recordlimit (via --take)	Page size
RecordOffset (via --skip)	Page offset

Every Nomis dataset has its own list. parl nomis dataset-def <id> returns the full dimension list with codelist references.

Provenance to cite

Tier 3 — third-party (ONS via Nomis), authoritative.

• Inline cite: "(via Nomis, ONS)" — once per paragraph.
• For formal output use "Office for National Statistics, Census 2021, via Nomis, retrieved {date}" — the underlying authority is ONS; Nomis is the delivery vehicle.
• When publishing a constituency-level figure, name the dataset table number (e.g. "TS001"), the geography type used (TYPE172 = Westminster 2024), and the date.
• The bare GSS-code shortcut can silently return 0 obs; the skill body documents the typed-query workaround. Don't claim "no data for X" if you've only tried the GSS shortcut.
• See ../../docs/provenance.md for cross-skill rules.