When someone complains about their council and the council fails to sort it out, an independent watchdog (the LGSCO) can investigate and rule on it — this counts how many times the watchdog found the council at fault in a year. It is the core 'proven failure' count for a council. A high number means the watchdog repeatedly confirmed residents were let down; a low number can mean good service, but also simply that few cases were investigated.
| name | lgsco_upheld |
|---|---|
| dataset | 09-ombudsman-lgsco |
| kind | extracted |
| type | count |
| unit | complaints/year |
| grain | council-year |
| role | outcome |
| source | data/lgsco/2-Complaints-Decided-{yy}-{yy}.csv, column `Upheld` (under `Investigated` group; `Investigations` in 23-24) |
| period | 2023-24 and 2024-25 editions |
| missingness | Rows exist only for councils in scope of LGSCO; `Upheld decisions per 100,000 residents` and rate columns are blank for zero-decision councils. Percent columns in file 2 are strings like `67%`. Join to Housing Ombudsman data is by name and lossy (ALMO/association names don't match council names). |
| score_note | Core per-council fault count — strong outcome, though duplicated by upheld_n and the rate family. |
| peer_group | council class (district / county / unitary / metropolitan / London) — not yet scored on the leaderboard |
Number of detailed investigations in which the LGSCO found the council at fault. The core adverse-finding count for a council. Only investigated cases can be upheld, so low volume can reflect few investigations rather than good service.