Online Tools for Analysing Religious Data: (2) Britsocat.com

Following on from the previous post, a second online resource exists at Britsocat.com, which provides an online tool for analysis of data from the British Social Attitudes survey. You need to register your e-mail address and login with a password, but access is free and it’s very easy-to-use. It’s run by the Centre for Comparative European Survey Data, which also runs a resource for the British Election Study.

This is an amazing resource: over 20,000 questions have been put to respondents over the course of the 1983-2008 surveys, most as a one-off, but some recurrently. Answers are for nationally-representative samples of the population. You can search the database of questions, or browse by category: attitude and value scales; business ethics; central government and the establishment; civil liberties; constitutional issues; crime; defence and diplomacy; the economy; education; e-society; ethnicity and race; Europe; “gender”; health; housing; identity, locality and region; labour market, employment and training; relationships; media and technology; morality and personal ethics; Northern Ireland; pensions and elder care; politics and political parties; religion and beliefs; science; class, age and “gender”; social welfare, inequality and poverty; and transport.

With regard to religion, the frequency of church attendance and religious affiliation has been asked in all survey years (namely since 1983, although not 1987 or 1992, when the survey was not run). Religious beliefs were examined in selected years since 1991. The respondent’s past own and parental attendance was asked in 1991, 1998, and 2008.

In 2008, there was round about 100 additional questions on specific aspects of religion and religious identity (described further here).

Within Britsocat, the questions are categorised for browsing purposes as relating to:

• the meaning of life;
• religions and religious organisations;
• religious affiliation;
• religious beliefs;
• religious convictions;
• religious observance;
• religious participation; and
• religious prejudice.

You can calculate frequency of responses for different questions for different years online, and export the results as CSV files for editing in Excel. Alternatively, you can also create charts of the results online, and edit the charts to make them more attractive for your work – and of publishable quality (subject to acknowledging the software creators).

You can also create cross-tabulations to break responses to questions down by age, sex, broad faith group and so on.

Here is an example of what is possible. In 1991, 1993, 1995 and 1996, respondents were asked, 

“How would you describe yourself … as very prejudiced against people of other religions, a little prejudiced, or, not prejudiced at all?”

The responses were: very prejudiced; a little prejudiced; not prejudiced at all.

I looked for breakdowns by age (18-34, 35-54 and 55 plus) and used the chart tool to help visualise trends. It’s interesting to note that the young, who are generally more tolerant towards minorities, are apparently a little more intolerant in this case than the older age groups. Whether this is an aversion to religious difference or religiosity (since a high proportion of the young are secular, and have no religious affiliation) is an open question.

I’ve also used the different colour options to show how the charts can be edited: it’s a neat little tool.

I’m a big fan of these tools, because in quantitative sociology there is a slight fetish for the less accessible statistical softwares and methods – which screen out the amateurs! But there is a great deal of value in using existing data and accessible tools to cover unexplored ground, and also to provide exploratory empirical analysis quickly, particularly for survey papers or policy-oriented papers. Many researchers do not have access to SPSS or Stata at work (they can be expensive) and so tools of this sort are extremely valuable.

An additional tool which allows simple multivariate analysis (namely linear regression) and two-way cross-tabulations exists at the ESDS via the Nesstar tool, though this analysis requires either a UK Higher Education Federation login, or registration with the UK Data Archive (follow the links provided by  the ESDS). Reports of frequencies (percentages of respondents replying to each response option) are available without a login.

For the BSA data, go to the ESDS Nesstar catalogue, and then click on ‘Research Datasets’. The BSA data is then ordered by year, with clickable headings beneath each survey year for ‘Metadata’ and ‘Variable Description’. The individual questions are listed under the latter, categorised by subject.

For those with access to statistical software, the British Social Attitudes survey microdata are available for download at the Economic and Social Data Service, again following registration, or via a UK HE Federation login. The datasets and questionnaires aren’t as ‘browsable’ as at Britsocat.com, so it’s worthwhile using Britsocat.com in tandem with the microdata.