Quick start guide: British Social Attitudes (BSA)

The British Social Attitudes Survey

The British Social Attitudes Survey (BSA) is the leading survey in Britain for monitoring trends and changes in public views on social and political issues. Over the past four decades, the survey has played a key role in providing authoritative statistics on important national topics, which makes the survey valuable for policy making, research, charity and media.

The BSA is a repeated cross-sectional survey that has been carried out every year since 1983 (except in 1988 and 1992 when British Election Studies were run instead). It is conducted by the largest independent social research organisation in the UK, NatCen Social Research. The data are collected from adults aged 18 or over through face-to-face interviews and self-completion questionnaires. Each year, nearly 4,000 respondents were asked around 300 questions about their views and attitudes on a range of political and social topics.

The BSA questionnaire consists of core topics that have been covered every year to examine trends over time. However, in order to keep the survey up-to-date and to reflect recent areas of interest, new questions are introduced. There are also a set of questions that are included in the survey less often. In some years, the BSA also includes questions for the International Social Survey Programme (ISSP), a collaborative project to enable cross-national analysis

The BSA seeks public opinion about a wide range of topics and covers demographics about the survey respondents. The topics change every year. Some topics are replaced with new ones to ensure that the survey is up to date, whereas others remain in the survey. For example, the 2019 BSA collected data on political party identification and Brexit, health and welfare, poverty and income, education and employment, equalities, racial equalities, climate change and cannabis legalisation, vaping, retirement and pensions.

Views on Brexit before the referendum

  • Did you know that most people were unsure about what would happen if Britain left the EU?
    The 2015 BSA survey shows that 69% were “unsure” about the implications of Brexit and 14% were “very unsure”.
  • Did you know that many minorities thought that leaving the EU will not make much difference?
    The 2015 BSA survey found that 44% of minorities thought that Britain’s influence in the world would be “not much different” if the public voted to leave the EU, and 31% thought that Brexit would not make much difference to Britain’s economy.

Each year, NatCen publish a report from the survey. These reports are rich in examples of how the data can be used. The figures above come from a pre-release of the BSA 33 report Brexit: What will it mean for Britain? (PDF).

Core questions and rotating modules

A key aim of the BSA is to track the views and opinions of the public on national issues over time. Therefore, the BSA questionnaire is designed to have several core questions that are repeated in most years. These cover different topics such as politics, welfare, poverty, health, education, equalities, and employment.

In addition, the interview questionnaire consists of various background and demographic questions. The rest of the questionnaire includes a series of non-core questions (modules) on a number of social, political, economic and moral topics, which are included in the survey less frequently.

The full questionnaire of the BSA survey used is available on the documentation section of the UK Data Service datasets for each survey year. FAQs are available.

Split sample

The sample of the BSA survey is often divided into several equally sized parts. Respondents in each part are interviewed using a different version of the questionnaire (typically called version A, B and C). Some questions are included on all versions and therefore asked of the full sample while others are asked to only a part of the sample.

Practically, this means some questions have a smaller sample, however the split is created so that each part is still nationally representative. In the dataset, those who are not asked a question due to the split sample will be recorded with a missing value code and label.

Unit of analysis

The BSA interviews one individual aged 18 or over per household.

Sample design

The BSA is designed to yield a representative sample of adults aged 18 or over. The Postcode Address File (PAF) – which is the UK’s most up-to-date and accurate address database – is the sampling frame for the survey.

The BSA sample only includes people living in private households. Individuals living in institutions and in households whose addresses are not on the PAF are excluded from the survey sample.

The BSA uses random sampling to ensure that everyone in the sample frame has a known chance of being selected in the sample. The survey follows a multi-stage design, with three separate stages of selection: the selection of postcode sectors, the selection of addresses within those sectors and the selection of individuals.

For more information about the BSA sampling procedures that were undertaken in each year of the survey, use the survey documentation section on the data catalogue. There is detailed information about the sampling procedures used in the user guide available for each year of the survey, for example, British Social Attitudes 2019 User guide (PDF).

You can also access the full reports for each year of the survey from the NatCen BSA website.

Are there survey weights?

The BSA datasets includes a weight variable named wtfactor. Applying the weight when analysing the data is important for getting results that reflect the population. The weight helps ensure that the sample is representative of the population. It adjusts for features of the sample design that make some people more likely to be selected than others (unequal selection probabilities). The weight also adjusts for non-response in the survey as not everyone selected to take part in the survey does. To adjust for non-response, the survey team uses information about non-responding households and the population. The weight also adjusts the sample to match the population in terms of age, sex, and region (which is called calibration weighting).

For more information about how the BSA weight is generated, an example is on the British Social Attitudes Survey, 2019 under documentation available via the UK Data Service. Information about the BSA weight is also available in the technical details section of the BSA reports, for example, British Social Attitudes 30 on the NatCen BSA website.

You can learn more about the BSA survey from: