At present, data collected for the project includes:
The Speech from the Throne (the Queen’s Speech), 1911-2012.
Acts of UK Parliament, 1911-2015.
Prime Minister’s Questions, 1997-2008.
Front page headlines of The Times of London, 1960-2008.
Public expenditure by function of UK government, 1910-2007.
Bills of the Scottish Parliament, 1999-2008.
Hearings of the Scottish Parliament, 1999-2007.
Public opinion about the ‘most important problem facing the country’, 1946-2015.
UK Statutory Instruments, 1987-2022.
Party Manifestos, 1983-2015.
Media
The UK media data measures the policy content of the front page of The Times of London. Front page news stories are sampled on the Wednesday of each week between 1960 and 2008, with the headlines blind-coded by two researchers according to major topic code, generating a database of 21,854 front page headlines. Due to important changes in the formatting of The Times during period between 1960 and 2008, along with a strike that stopped production for almost a year we strongly suggest that those interested in the media data refer to the dataset codebook.
The short and long titles of Acts of the UK Parliament were blind-coded by two researchers; assigning a major topic code and sub-topic code to each piece of legislation. This procedure led to eighty-five percent inter-coder reliability for major topics. The remaining differences were resolved through discussion by the project leaders.
Prime Minister’s Questions (PMQs) was first introduced in 1961 and has become a weekly centrepiece of British political life. Our data contrains questions from 1997 to 2008 that were blind-coded by two researchers; assigning a major topic and sub-topic code to each question. These are coded based on the question asked and not the answer given hence their place alongside other parliamentary datasets.
The short and long titles of Bills of the Scottish Parliament were blind-coded by two researchers; assigning a majortopic and subtopic code to each. This procedure led to eighty-five percent inter-coder reliability for major topics. The remaining differences were resolved through discussion by the project leaders.
A short description of the content and purpose of Hearings of the Scottish Parliament were blind-coded by two researchers; assigning a majortopic and subtopic code to each. This procedure led to eighty-five percent inter-coder reliability for major topics. The remaining differences were resolved through discussion by the project leaders.
This was gathered by Shaun Bevan. It contains all Scottish statutory instruments from 1999 to 2014 coded according to the Comparative Agendas Project based on their subject.
Please cite the following article when using this data:
Bevan, Shaun. “Devolution is Secondary: What Drives Scottish Secondary Legislation?” Public Administration, forthcoming, 2021. (DOI: 10.1111/padm.12694).
This dataset was gathered by Shaun Bevan with the support of the Mannheim Centre of European Social Research (MZES), University of Mannheim (1987-2008) and later extended (2009-2022) and updated with additional variables. It contains all national level UK statutory instruments from 1987 to 2022 coded according to the Comparative Agendas Project based on their subject.
Please cite the following articles when using this data:
Bevan, Shaun. 2015. “Bureaucratic Responsiveness: The Effects of Government, Public and European Attention on the UK Bureaucracy.” Public Administration, 93(1): 139-158. (doi: 10.1111/padm.12113).
Bevan, Shaun and and Zoe Greene. 2026 “Administrating Crisis is Just a Transition: Interventions on Bureaucratic Activity in the United Kingdom, 1987-2022.” Journal of European Public Policy, 33: 11-39. (DOI: 10.1080/13501763.2025.2514827).
The Speech from the Throne, the King’s or the Queen’s Speech, sets out executive and legislative priorities of British government in the forthcoming session of parliament. The full text of the speech is coded at the ‘quasi-sentence’ level (a quasi-sentence constitutes an expression of a single policy idea or issue). Transcripts of the speech were blind-coded by two researchers who compared and reconciled their responses; first to ascertain whether each quasi-sentence contained any policy content and then to assign a major topic and sub-topic code to each quasi-sentence. This procedure led to ninety per cent inter-coder reliability for major topics for most years. The coders resolved remaining differences through discussion and project leaders made the final decision in the few cases where coders could not agree.
The creation of the UK Party Manifestos data was led by Caterina Froio for 1983-2010. Christoffer Green-Pedersen and Henrik Seeberg led on the 2015 data. Complete details of these efforts as well as the codebook and notes from both periods is included at the end of the variable codebook.
Budgetary data between 1910 and 2007 is compiled using a combination of the UK Blue Book on Government Expenditure and expenditure data reported in Alan T. Peacock and Jack Wiseman (1961, Princeton University Press) The Growth of Public Expenditure in the United Kingdom.* Budget data from 1951 to 2007 was taken directly from the UK Blue Book on government expenditure recoded to match the policy agendas coding system. Each Blue Book contains several years of data updated to account for inaccuracies and additional expenditures not initially included upon first publication. To obtain a time series of the annual expenditure of British government from 1950 to 2007, data from the Blue Book for 1962, 1971, 1980, 1989, 1997, 2002 and 2008 were used. For more information about working with budget data, generally, there is a page with resources.
This dataset contains responses to Gallup UK's survey question about the "Most Important Problem" aggregated at the annual level and coded by major topic. Contact us for monthly or quarterly MIP data if needed.
This dataset contains responses to Ipsos-MORI's survey question about the "Most Important Issue" aggregated at the annual level and coded by major topic. Contact us for monthly or quarterly MII data if needed.
