Ipsos conducted a large survey via the UK online random probability KnowledgePanel. Panellists are recruited using random probability unclustered address-based sampling, the gold-standard in UK survey research, meaning that every household in the UK has a known chance of being selected to join the panel. Crucially, members of the public who are digitally excluded are given a tablet and provided with an email address.

Ipsos interviewed 20,835 adults aged 16+ living in the UK, including oversamples in Scotland and Northern Ireland. Data collection took place between 19-25 September 2024. Data was weighted by gender, age, region, education, ethnicity, Index of Multiple Deprivation quintile, and number of adults in the household, to reflect the profile of the UK population.

Multi-level regression and post stratification (MRP) is an advanced modelling technique that estimates the likelihood of a survey response in each small geographical unit by using a national survey with a large sample size. In this case, we have produced MRP estimates at the lower tier local authority level. Data on satisfaction with public services and local circumstances is analysed by a wide range of factors to see how different types of people, in different areas, are likely to feel. For example, it estimates the probability that a woman, aged 25-34, with a degree, living in a highly deprived area in the North East of England is satisfied with a public service. These estimates are then applied to the differing profiles of each local authority to estimate how many people in each local authority are satisfied.

Estimates are subject to a wide range of errors related to the survey data and the modelling and should be interpreted with caution.

The models are based on differences by demographics such as age, gender, ethnicity, working status, housing tenure, children in the household, education, and local and regional characteristics such as ONS output area classification. Other factors which might also have an impact on local perceptions are not taken into account (for example, the models do not include any ‘real world’ service outcome inputs or political variables and are only intended to estimate local perceptions of public services based on demographics and which area the authority is in).

We have relied on census data from the 2021 Census in England, Wales and Northern Ireland, and the 2022 Census in Scotland.

As the modelling makes use of a national survey, caution should be taken when looking at estimates for individual local authorities. While MRP is good at taking into account the different demographic profiles of each local authority, with relatively few respondents per local authority, it is unlikely to capture the full local context, e.g. where a public service is over or underperforming substantially due to specific local factors, the model will not be able to take all of these into account.

Furthermore, as with any survey approach there will be margins of error in each estimate. These will be wider than an equivalent survey as there is uncertainty from the modelling as well as the normal survey-based margins of error. The full data, including margins of error, can be downloaded here. An average margin of error is 12 points.

We would encourage readers to not place too much certainty into specific point estimates.