Documentation   Keywords  Publications  Related Studies

Title:
Content Analysis of Newspaper Coverage, British General Election, 2001

Series: (British Election Study Series)

Subject Categories:
British Election Studies - Major studies
Election and campaign studies - Politics

Depositor(s):
Whiteley, P.F., University of Essex. Department of Government

Principal Investigator(s):
Clarke, H., University of Texas (Dallas). School of Social Sciences
Sanders, D., University of Essex. Department of Government
Stewart, M., University of Texas (Dallas). School of Social Sciences
Whiteley, P.F., University of Essex. Department of Government

Sponsor(s):
Economic and Social Research Council

Abstract:
The British Election Study (BES/BGES) series constitutes the longest academic series of nationally representative probability sample surveys in the country. Its broad aim is to explore the changing determinants of electoral behaviour in contemporary Britain. The surveys have taken place immediately after every general election since 1964. Besides the main election surveys, other datasets have also been produced. For example, some studies have included separate sub-samples for ethnic minorities and areas such as Scotland and Northern Ireland (held at the UK Data Archive (UKDA) under SNs 681, 3171, 3889, 3891, and 4622), and several inter-election panel studies have been undertaken between 1969 and 2001 that follow the same individuals interviewed in the cross-sectional surveys (see SNs 422, 2983, 3888, 4000 4028 and 4620).

Since the election series was originated in 1963 by David Butler and Donald Stokes under the name of Political Change in Britain, 1963-1970 (see under GN 33099), it has been under the direction of a number of academics over time, as detailed below.

British Election Study, 1974-1983 (BES):
The BES was started as a research project at the University of Essex in 1974, to continue the series started by Butler and Stokes. The BES conducted interview surveys following the general elections of February 1974, October 1974 and May 1979. This series also includes a questionnaire survey conducted after the referendum on Britain's membership of the European Economic Community (EEC) in 1975 (see SN 830).

British General Election Study, 1983-1997 (BGES):
From 1983, the series continued under a slightly changed name, directed by Anthony Heath, Roger Jowell and John Curtice, and conducted jointly by Social and Community Planning Research (SCPR - now the National Centre for Social Research, or NatCen) and Nuffield College Oxford. The 1992 BGES surveys were carried out as part of the activities of the Joint Unit for the Study of Social Trends (JUSST), funded by the Economic and Social Research Council (ESRC). A grant by the ESRC to the University of Strathclyde enabled the representation of Scottish electors in the sample to be boosted substantially. This 'oversampling' of the Scots was undertaken in order to permit more detailed investigation of voting behaviour in Scotland than had usually been the case with the BGES. Prior to 1992 BGES surveys excluded electors living in the five constituencies in the Scottish highlands and islands north of the Caledonian Canal because the small and scattered electorate there could not be interviewed cost-effectively. In the 1992 BGES survey however, these constituencies were included in the sampling frame. Users should note that the Scottish element of the 1992 survey has been down-weighted to form a representative British sample. Researchers wishing to analyse the Scottish sample separately should order SN 3171 General Election in Scotland, 1992.

The 1997 studies were carried out by JUSST's successor, the Centre for Research into Elections and Social Trends (CREST) (an ESRC-funded research centre linking NatCen with the Department of Sociology at the University of Oxford) in collaboration with Pippa Norris of Harvard University, the Gatsby Charitable Foundation and the Commission for Racial Equality. The 1997 BGES included a greater number of separate but interlinked studies than ever before, for example the separate Scottish and ethnic minority boost sample datasets (SNs 3889 and 3891 respectively) mentioned above.

British Election Study, 2001:
The BES for the 2001 general election (the name was changed again), funded by the ESRC, was once again based at the University of Essex, under the control of David Sanders, Paul Whiteley and Harold Clarke. The 2001 survey comprised three main components and two supplementary ones. The first main component comprised a pre- and post-election cross-section survey, and a panel survey (held under SNs 4619 and 4620). The second consisted of a free-standing 4,500-person 'rolling' election campaign survey, conducted during the general election campaign (held under SN 4621), and the third component comprised a separate election survey conducted in Northern Ireland (held under SN 4622). The supplementary components included the construction of an aggregate, 'ecological' data file, based on census and other data, which describes the enumeration district-level, ward-level and constituency-level context for each panel survey respondent, and a content analysis of national press coverage of the election campaign (see SN 4623). This analysis resulted in the creation of a 'campaign dataset' which enabled movements in opinion during the campaign to be related to press coverage of the campaign.

British Election Study, 2005:
The BES for 2005 was also based at the University of Essex. Its principal objectives were to study long-term trends in British voting behaviour; explain the election outcome, party choice and turnout; and examine the consequences of elections for the operation of democracy more generally. The 2005 BES again comprised three main components: the face-to-face cross-section survey and a follow-up conducted one year later (all held under SN 5494), the British component of an international comparative study of electoral systems (SN 5495), and an internet-based pre- and post-election 'rolling' campaign panel survey (SN 5496).

Further information about BES 2001 and 2005 may be found on the British Election Study at the University of Essex web site.

BESIS:
British Election Study data and documentation from 1966-2005 are also available online via the British Election Studies Information Site (BESIS), maintained by the Centre for Comparative European Survey Data (CCESD). Facilities include online analysis, subsetting of data and mapping of election results by constituency. Further information about and links to the BESIS site are available under UKDA SN 5860.

SN 4623 contains the content analysis of campaign newspaper coverage component of the 2001 BES. The aim of the project was to conduct a high-level content analysis of the press coverage of the 2001 election campaign, resulting in the creation of a campaign dataset that will enable movements in opinion during the campaign to be related to press coverage of the campaign. A simplified version of the 1997 Scammel/Semetko coding schema was devised for this purpose. Articles relating to the election campaign were identified from eight daily newspapers for the election campaign period. Articles were manually coded according to the set coding frame and coding was examined after completion to ensure consistent results. As an addition to the datasets of coded articles, corresponding datasets of article headlines were compiled to facilitate further research.


Main Topics:
Articles relating to the election campaign were identified from eight daily newspapers (Monday-Friday): The Guardian, The Times, The Daily Telegraph, The Independent, The Sun, The Mirror, The Daily Mail and The Express. This included articles from within the home news pages of the newspapers and in addition, leaders, editorial and comment articles referring to the election campaign. No photographs, paid-for campaign advertisements, diary columns or articles of less than 50 words (with the exception of front page articles) were included. Campaign articles from other pages or separate sections to the main newspapers were not included - for example, international, finance, sports and letters pages, The Guardian's 'G2' section, etc. The exceptions to this were leaders, editorial and comment articles from The Independent which all appear in the supplementary 'Review' section.
Five files are available:
file 1: a dataset of 1440 campaign articles content-coded using the coding schema;
file 2: a dataset of 269 non-campaign articles partially content-coded using the coding schema;
file 3: a dataset of 200 front-page campaign headlines;
file 4: a dataset of 269 front-page non-campaign headlines;
file 5: a dataset of 1240 campaign headlines from a sample taken from the main section of newspapers.

Standard Measures
A simplified version of the 1997 Scammel/Semetko coding frame was used. Please see documentation for further details.

Coverage:
Time Period Covered: 09 May 2001-07 June 2001
Dates of Fieldwork: June 2001-February 2002
Country: United Kingdom
Spatial Units: No spatial unit
Observation Units: Text units (documents/chapters/words)
Kind of Data: Textual data; Numeric data; coded from textual data

Universe Sampled:
Location of Units of Observation:National
Population:Election-related articles from national newspapers published in Great Britain between 9 May - 7 June 2001.

Methodology:
Time Dimensions: Cross-sectional (one-time) study
Sampling Procedures: For full details of the sampling and coding procedures used, please see documentation.
Number of Units: 2643 (obtained) Election campaign newspaper articles
Method of Data Collection: Compilation or synthesis of existing material

Language(s) of Written Materials:
Study Description: English
Study Documentation: English

Access:
Access Conditions: The depositor has specified that registration is required and standard conditions of use apply. The depositor may be informed about usage. See terms and conditions for further information.
Availability: ESDS Access and Preservation, UK Data Archive
Contact: Help desk: help@esds.ac.uk

Date of Release:
First Edition: 18 March 2003

READ File

Download Acrobat PDF Reader