For researchers and journalists
Data and downloads
Every release is published as a flat table and as the full snapshot the site itself renders from. Nothing here is behind a form. The index adds scoring, not measurement: each indicator comes from a UK official statistics producer under the Open Government Licence, and the source is named in every row.
Scores are stress readings from 0 to 100 where higher means more stress. If you are citing a figure, please cite both the British Resilience Index and the underlying publisher: see how to cite and the methodology.
By release
Every published release
Releases are immutable once published. Where a source later revises a figure, the revision appears in the next release and the original reading stays in the archive, so a citation of a past release remains reproducible.
| Release | Score | Downloads |
|---|---|---|
| August 2026 | 52 | indicators CSVdomains CSVfull snapshot JSONreview |
| July 2026 | 51 | indicators CSVdomains CSVfull snapshot JSONreview |
| June 2026 | 52 | indicators CSVdomains CSVfull snapshot JSONreview |
The current release is August 2026, generated 15 August 2026, with a data cutoff of 13 August 2026. Also available as latest-index.json.
Reference files
Sources, history and robustness
- source-registry.json
Every registered source: publisher, dataset, type, reliability, update frequency and link.
- metric-history.json
The full back series behind every indicator, as published, one array per indicator.
- sensitivity.json
Monte Carlo results: the headline under 2,000 alternative weightings, aggregations and normalisations.
- release-history.json
One row per published release: score, band, movement and a link to the full snapshot.
- data-dictionary.json
What every column in the CSV downloads means, and its permitted values.
Data dictionary
What every column means
All scores are stress scores from 0 to 100 where HIGHER MEANS MORE STRESS. Bands: 0 to 24 stable, 25 to 49 strained, 50 to 74 fragile, 75 to 89 critical, 90 to 100 crisis.
| Column | Meaning |
|---|---|
release_period | The release this row belongs to, for example 'August 2026'. |
metric_id | Stable identifier for the indicator. Safe to join on across releases. |
metric_name | Human readable indicator name. |
domain_id | Identifier of the domain the indicator sits in. |
domain_name | Human readable domain name. |
unit | Unit of latest_value as published by the source. |
latest_value | The indicator's value as published. Not rescaled. |
latest_period | The period the value covers, in the publisher's own wording. |
previous_value | The preceding observation, used for the trend component. |
direction_of_badness | higher_is_worse or lower_is_worse: which way stress runs. |
stress_score | 0 to 100 composite for this indicator: 60% level, 30% trend, 10% volatility. |
level_normalisation | percentile = ranked within the indicator's own verified history (needs at least 8 points); bounds = linear against best and worst observed bounds, used until that history exists. |
trend_score | 0 to 100 trend component. 50 is neutral. |
trend_basis | measured = computed from a real change; no_change = the latest reading equals the previous one, so the component is a neutral 50; not_computable = the previous value is zero and no fractional change can be formed. |
volatility_score | 0 to 100 volatility component. 50 is neutral. Provisional measure. |
volatility_basis | measured = computed from at least four history points; insufficient_history = fewer than four, so the component is a neutral 50. |
confidence_score | 0 to 1 data quality score for the indicator's source. |
scored | true = enters domain and national scoring. false = a provisional placeholder with no published official figure, displayed for transparency and excluded from every score. |
verified | true = the value was transcribed or fetched from the named official source. |
geographic_coverage | The geography the indicator is published for: uk, gb, england, england_wales, scotland, wales, ni. A devolved measure is never blended into a UK figure to fill a gap. |
source_id | Identifier into the source registry (source-registry.json). |
source_organisation | The publishing body. |
source_url | The dataset or release page the figure comes from. |
weight_pct | The domain's weight in the national score, in percent. Weights sum to 100. |
score | Domain stress score: the median of its indicators' stress scores. |
average_score | The mean of the same indicators, published for comparison with the median. |
risk_band | Band label for the score. |
previous_score | The domain's score at the previous published release. |
delta | Signed change since the previous release. Positive means more stress. |
metric_count | How many indicators the domain's score was computed from. |
Licence and reuse
Free to use, with attribution
The underlying figures are UK official statistics, released by their publishers under the Open Government Licence v3. The scoring, aggregation and written analysis are the British Resilience Index’s own. Reuse is welcome; attribution to both is expected.
The scoring engine is open source under the MIT licence, so any figure here can be recomputed from the published inputs rather than taken on trust.
Spotted something wrong? Corrections are published rather than quietly fixed: get in touch.