Social and Political Change in Britain: Gallup Polls 1945-1991

How have public attitudes on key political, social and economic issues changed since the Second World War? How do those changes vary across different groups in society? From the 1930s to the early 2000s, the survey organisation Gallup conducted around three thousand surveys of the social, political and economic attitudes of the British public. At the time, these opinion polls provided valuable insights on how the public thought about key issues, personalities and events of the day - the government, the party leaders, international crises, support for specific policies, and so on - with national level results reported in monthly Gallup reports and in the news media. To date, only a small fraction of the original surveys have seen the light of day. A recent discovery by the Roper Center for Public Opinion Research at the University of Cornell revealed a veritable treasure trove of important historical Gallup survey data - around 2,500 surveys covering the period between 1945 and 1991 - stored in 'column binary format' based on antiquated IBM punch cards, rendering them inaccessible to most researchers today. Processing and digitising this data presents a major technical challenge, but offers the prospect of enhancing understanding of social and political change in Britain between the 1940s and 1990s.

This project will develop a unique dataset that enables researchers to analyse how the major political, economic, and social shifts between the 1940s and 1990s influenced the dynamics of public opinion in Britain. It will firstly digitise all the codebooks of all Gallup surveys over the period between 1945 and 1991 (making them searchable online by researchers). It will further digitise around 700 surveys that contain questions relating to social, political and economic issues (and will specifically select 100 of these surveys through a user-led exercise enabling researchers to nominate those surveys to be prioritised for digitisation). It will also create a 'merged' dataset of individual surveys that combines repeated cross-sectional measures of public attitudes and demographics - enabling researchers to track changes in public opinion by subgroup over time. Using this new data resource, it will explore long-term trends in social and political attitudes in Britain, their reaction to key events and how they vary across different cross-sections of society.

The objectives of the project are as follows:

* To digitise codebooks of all surveys conducted by the UK Gallup Poll between 1945 and 1991.
* To convert 700 individual surveys conducted by the UK Gallup Poll between 1945 and 1991 from column binary data to current software and non-proprietary formats.
* To create a merged time series dataset based on individual surveys that offers a repeated cross-sectional measures of public opinion.
* To facilitate and promote the use of this new data resource by other researchers.
* To enable educational and public use of longitudinal data on public opinion through the development of web-based data visualisation tools.
* To use this new data infrastructure to track and understand long-term dynamics of public opinion in Britain, especially across different groups (e.g. partisans, by age).

The project has been designed with an emphasis on the end user, with the broader aim of creating a data infrastructure that builds research capacity as well as providing an accessible educational resource for the study of public opinion. Prospective beneficiaries from the project include academic researchers (from undergraduate level upwards) across many disciplines, survey researchers in commercial and non-governmental organisations, researchers in think tanks and political parties, teachers at schools and universities, wider publics and media with an interest in public opinion.

Our engagement and dissemination strategies will target key audiences in a range of ways. We have already engaged with potential users via recruitment of our prospective advisory board - drawn from a number of academic disciplines and fields. Further, the design of the project builds in user engagement at different stages (such as in the selection of individual surveys for digitisation, beta-testing of datasets, and training workshops), to ensure that the impact of the research is maximised based on guidance of stakeholders. The project is additionally expected to facilitate the production of research outputs that contribute to important current debates in the study of public opinion in Britain.

We believe that the research project will benefit the following groups:

* Data users: the outputs from the project primarily relate to the construction of data infrastructure based on the archive of Gallup UK polls, and the building of research capacity through the delivery of training workshops for data users. The repository of around 700 individual-level survey datasets and merged dataset of repeated survey measures (as described in the case for support) will provide a valuable resource for researchers across a wide range of disciplines; including history, sociology, economics, political science, geography and criminology. It may also be useful to survey researchers based in think tanks and political parties, and non-specialists with an interest in public opinion.

* Researchers using advanced quantitative methodologies: the repeated cross-sectional format of the new data source will enable methodological innovation by other researchers analysing public opinion over time. For example, it will enable researchers to conduct time series analysis of cross-sectional data, regression analyses of individual-level survey data, multi-level modelling of individual attitudes nested within regions or years, or age-period-cohort analyses of attitudes on taxation and public expenditure.

* Wider industry and public/media: the data infrastructure and associated research is also expected to be of interest to those in the survey industry in the UK and US, and elsewhere. Findings from the project will also be of interest to media and wider publics, in enhancing understanding of political change in Britain over half a century, and specifically dynamics of public opinion during this period.

In order to ensure that potential beneficiaries have the opportunity to benefit from this research project, we have developed an impact plan that includes the following:

* Engagement of research users in the project via its advisory board and the selection of 100 individual surveys for digitisation.

* Presentation of information regarding new data, tools and analysis at a number of national and international conferences, and via workshops and events across the UK.

* Promotion of the availability of this new data resource and tools via professional bodies and research/user communities (through blogs/articles and other social media).

* Promotion of the data resource and findings via national media and industry publications.

* Delivery of two training workshops for prospective users.