Election coverage, Framing and political parties in the 2017 British General Election Ceri E. Taylor
11368527
MSc Communication Science½Political Communication Supervisor: Dhr. C.M. Brenes Peralta
Abstract
Looking at the case of the British election, this study uses quantitative content analysis to uncover the presence of four frames in British newspapers coverage of the 2017 British General election. The four frames included are strategic game frame, attribution of responsibility frame, human interest frame and economic consequences frame. Based on previous findings, this study looked at the use of such frames across seven British
newspapers, using Semetko and Valkenburg’s (2000) seminal content analysis of Dutch press and television news looking at the framing of European politics.
Our findings show that the dual-party system is replicated in election coverage, with the Conservatives and the Labour party being the most discussed political parties. Concerning the use of the four frames, human interest frame was the most commonly used followed closely by strategic game frame. The analysis found that there is not a significant difference between the use of strategic game frame in either tabloids and broadsheets or online articles and print articles . Other hypotheses including the use of attribution of responsibility and economic consequences being more common in broadsheets than tabloids were not
supported. Finally, the use of human interest frame was not significantly higher in tabloids compared to broadsheets but was used significantly more in online articles than print. Keywords: Framing, election coverage, strategic game frame, British election, News coverage, generic frames
British Political Parties and Framing in the 2017 General Election In communication science research there has been a growing interest in the phenomenon of framing in recent years, with particular interest in framing effects and developing the operationalization of generic frames in the media (de Vreese, 2005; Semetko & Valkenburg, 2000). The media is a tool which links the reality of what is happening in the world and the impression of it within your own mind (Lippman in Strömbäck & van
Aelst,2010). Framing intercepts this process and can change your perception of issues as a result. This study looks at the use of different frames in British newspapers during election periods.. The British media has a diverse landscape, labelled an outlier by Hallin and Mancini, sitting comfortably between the Democratic Corporatist model and Liberal model (Hallin & Mancini, 2004). As such the press and television hold very different journalistic styles and therefore their resulting news coverage varies (Scammell & Semetko, 2008). Whilst the British Broadcasting Company (BBC) is known world-wide for its balanced and bias-free publicly funded journalism, the British newspaper press is historically partisan and highly opinionated(Yaxley, 2017). This study will analyse the use of previously defined frames, specifically the concept of strategic game frame, that were commonly used in British Newspapers during the campaigning period of the 2017 British General Election and
compare the use across newspaper type and format (Scammell & Semetko, 2008). Context of Research
The eight-week campaigning period concluded with a result that did not follow either what the polls predicted or what the newspapers reported in the run up to election day. Prime Minister May was predicted to win a landslide majority when it was first called in April, with YouGov putting Westminster voting intention at 48% and Labour at 25% (Smith, 2017). The role of the media in British elections has been surrounded by increased volatility over the years given the rise of social media and other online platforms being used by political parties
as a way to overcome the partisan and heavily opinionated British newspapers (Deacon & Wring, 2011). While online platforms have become increasingly the default source for
individuals’ daily dose of news and information, academics argue that this does not mean that newspapers become irrelevant as they still have a tendency to set the agenda for both
television and online sources (Nielson, 2017). YouGov, a polling company found that during this election traditional news sources still remained influential among voters with just under half of those polled believe they had a significant impact on the outcome (Yaxley, 2017).
Previous studies on framing during election periods have found that certain
characteristics of the election are linked to the most commonly used frames in the lead up to election day. In other words, aspects such as electoral system of a country, the number of parties running as well as media system were all found to be linked to the use of certain specific frames (Lawrence, 2000; de Vreese, 2005; Patterson, 2017; Schmuck et al, 2017). Given the election was written off as a win for the Conservatives when it was called, the framing of such an election might not be a similar to previous ones (Strömbäck & van Aelst, 2010). In a dual-party system which already makes elections less competitive, the news coverage reflects this, with Labour and Conservatives making up just under 82% of airtime given to all parties in the first two weeks of the campaign, up from 67% in 2015 (Cushion, 2017). Therefore, our first hypothesis is as followed:
H1: The Conservative party and Labour party were the most discussed parties in British newspapers.
In Hallin and Mancini’s ‘Comparing Media systems’ the United Kingdom is labelled together with the United States as a ‘North Atlantic model’ which suggests a more liberal approach to the media market (Scammel & Semetko, 2008). While this is true in terms of the printed press, it diverges towards the ‘Democratic Corporatist’ model when it comes to television (2004). Lumping the U.K. together with the U.S. is evident in the Newspaper
industry which follows a free-market approach much like the U.S and has shown increasing tendencies towards American reporting characteristics such as presidentialization and
increasing use of strategy game frame- Horse-race journalism-in British newspapers as of late (Scammell & Semetko, 2008). What’s more, the United Kingdom has a majoritarian
government system whereby coalitions are rare and therefore parties rarely have to work together the same way they would have to in proportional representation systems such as in Sweden (Strömbäck & Shehata, 2007).
Moving towards the British media and how they cover such events, limited research is available on specifically the British media and their framing of elections but rather it is usually part of a greater comparative study looking across several countries (Strömbäck & Shehata, 2007). British newspapers have long since been vital to the election process and could be seen as having played roles in past politicians’ downfalls such as the case of Labour leader Neil Kinnock in the late eighties - early nineties and The Sun newspaper (Scammel & Semetko, 2008). Limited academic research is found on the subject of election coverage but Loughborough University’s Centre for Research in Communication and Culture continue to conduct media analyses of recent general election campaigns. Whilst these reports do not look at framing directly they do look at the overall coverage of political parties, its leaders and the main issues which can be helpful when looking at such issues in the 2017 general election.
Framing
Framing has increased in prominence in communication science as studies have identified that the emphasis of different aspects of an issue or event can influence how an individual perceives the event and thereby their understanding of it (de Vreese, 2005;
Entman, 1993). Many different definitions have been identified in the literature, but Entman’s 1993 definition is both heavily cited and to the point. ‘To frame is to select some aspect of a
perceived reality and make them more salient in a communicating text, in such a way to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation for the item described.’ (Entman, 1993; 52). Definitions of framing are numerous, but more generally they help in the understanding of complex issues and tap into knowledge and norms of the audience (de Vreese, 2005). Capella and Jamieson compared news frames to how a photographer frames their subjects (1997). Framing a photograph determines what is included or excluded in the frame and what is the focus of the photograph (Capella & Jamieson, 1997). The same can be said for news frames, which depending on what frame is implemented, includes and excludes certain aspects and implies a certain problem definition or solution. This is the case for most media outlets who have economic or political interests which often dictate newspaper coverage or lack thereof. Literature defines both media frames and audience frames, the former exploring how stories or issues are presented in the news and the latter how the audience perceive these issues (de Vreese, Peter & Semetko, 2010). This study focuses on the former; looking at possible frames used in the press coverage of political parties during the British election.
The frames that are being analysed in this study are derived from previous literature on news framing, with a specific focus on strategic game frame (de Vreese, Peter & Semetko, 2010; de Vreese, 2005; Lawrence, 2001; semetko & Valkenburg, 2001). This frame focuses on almost exclusively campaign tactics and electoral process (de Vreese, 2005). It is part of a growing area of research looking into the use of strategy or game frame in election coverage (Lawrence, 2000; Schmuck, Heiss, Matthess, Engesser & Esser, 2017; de Vreese, 2005). The concept of strategic game frame used in this study stems from previous research which connected the two frames- strategy and game frame- to form the frame (Strömbäck, van Aelst & de vreese, 2011).
Post-electoral analysis found several important points in the campaign which influenced the issues discussed both by the media and politicians as well as a tendency to cover the election in terms of political actors and not parties; meaning a more personalized approach to politics which has further effects on public opinion (Cushion, 2017). It seems the U.K. is following the United States in more ways than one, as their media system expands and becomes more market-driven and their election campaign was the most presidential yet (Deacon et al, 2017). Furthermore, the short notice of such an election combined with the Theresa May’s brief stint in Government meant there wasn’t enough time to establish a new narrative to contradict the anti-EU agenda which had over 20 years to develop ( Schifferes, 2017). What must be emphasised is the U.K.’s contrasting media system where both the radio and the BBC are heavily regulated whilst newspapers are exclusively market-oriented and partisan. Numerous studies have found that the use of different frames have various effects on the audience ( de Vreese, 2005; Patterson, 2017). Such an analysis on coverage of the event can further future research on the possible effect the framing of the election had on the public and its voting behaviour during this period. Something to note however, is the effect of the use of different frames has been found to be dependent on other variables, such as level of knowledge and education ( Capella & Jamieson, 1997).
There is a clear understanding in the literature that framing and their effects vary significantly depending on the social or political context, education of the receiving audience, ownership of the outlet, type of media outlet and the issue itself. Other frames analysed in this study are taken from Semetko and Valkenburg's content analysis of press and television news in framing European Politics which looked at the prevalence of four common news frames (2000). The same study found evidence to support the statement that the frequent use of attribution of responsibility, conflict and economic consequences frame is just as
Studies looking at the exposure of both issue framing and specific framing and the effect it has on political cynicism found that exposure to issue-framing did not lower cynicism but rather resulted in as if no news has been received at all comparative to strategy framed news (Capella & Jamieson, 1997). One thing clear in most of the studies conducted is that strategic framing does result in an increase in cynicism (Capella & Jamieson, 1997). With this in mind, the following research question was created:
RQ: How was the 2017 British General election framed in British Newspapers? Concepts such as the spiral of cynicism have been linked to strategy based coverage which implies politicians are acting on self-interest and not the common good; individuals who consume a high amount of this coverage are more likely to project this framing onto their own thoughts (Aalberg, Strömbäck & de Vreese, 2010, Capella & Jamieson, 1997). Within these frames they also included both offline and online media outlets and found a distinct variation in the frequent use of certain frames. In looking at online platforms versus traditional formats they found that the former implemented strategic game frame significantly more often than traditional platforms (Schmuck et al, 2017). This was linked to the vast amount of competition found online; strategic game frame is an easy option to reduce time and money spent on a story whilst attempting to make something boring more appealing to a greater audience (Stromback & van Aelst, 2010). Strategic game frame is used to lower the barrier between hard news, which includes election news, and the average citizen. The comparative work of Stromback and van Aelst concerning Swedish and Belgian newspapers found that in both countries tabloids were more likely to use game frame (2010).
Generic versus Issue-specific framing
The concept of framing is further split into two categories: issue-specific and generic frames (Schmuck et al, 2017: de Vreese, 2005). The former is event or case-specific whereas
generic frames can be applied to most issues and have very few limitations (Schmuck et al, 2017). Generic frames are the subject of this study. The strategic game frame is a commonly used generic frame in election coverage; it is synonymous with campaigning and electoral periods (Aalberg, Stromback & de Vreese, 2011). It frames politics as a strategic game, where there are always winners and losers. It focuses on the performance of politicians and parties as well as their strategies and tactics, relegating policy issues and stances to the bench (Aalberg, Stromback & de Vreese, 2011). The use of the frame appears to have increased over time and has been associated with the increased professionalization of election
campaigns in recent years (Norris, 2000 in Azrout et al, 2013; Cappella & Jamieson, 1997). It is part of emerging political journalism and now dominates US mainstream political news coverage (Aalberg, Stromback & de Vreese, 2011). The increase of this type of coverage is linked to the changing nature of the political system and the media landscape as modern campaigning requires more elaborate strategies and journalist understand it as their job to uncover these strategies; it is a balance between the journalist trying to stay objective in its reporting and pushing back on the politician’s attempt to control what the media publish (Aalberg, Stromback & de Vreese, 2011). Other research suggests that it is in fact a positive outcome as it informs citizens about politics in an entertaining way (Zaller, 1999 in Dunaway & Lawrence, 2015). The economic benefits of such a frame are clear; it is a frame which is adaptable between each election cycle and little research is involved in implementing it (Dunaway & Lawrence, 2015). This explains findings that that show it is more likely to occur in online media outlets, which often have less time and money to invest into stories
(Schmuck et al, 2017). Whilst its effect on the public is still unclear, it nonetheless plays a vital role during election periods. This frame will therefore be the focus of this study, with several hypotheses testing the use of it. From this, the following hypothesis was created:
H2: Strategic game frame is more frequently implemented in online format outlets than print formats.
Generic news frames
Other frames analysed here are taken from Semetko and Valkenburg’s study of press coverage of the 1997 Amsterdam meeting of heads of state in both the print and television press (2000). These frames include: attribution of responsibility, human interest and
economic consequences (Semetko & Valkenburg, 2000). Human interest frame attempts to personalize or emotionalize the story and is found to be implemented more in sensationalist newspapers (de Vreese, 2005; Semetko & Valkenburg, 2000). This is attributed to the fact that such titles are more likely to use such a frame to catch the audience’s attention and simplify complex matters (de Vreese, 2005; Semetko & Valkenburg, 2000). It puts a face to a policy issue and instantly reduces the distance between the audience and the story. For
example, the image of Alan Kurdi, the Syrian toddler found lying face down in the sand on a Turkish beach, put a face to the refugee crisis that was too far away. Whilst this frame has not been used with much frequency in the coverage of politics, it is included because it is more likely to be used by tabloid newspapers and the most popular newspaper in the UK are tabloids (Semetko & Valkenburg, 2000; Schmuck et al, 2017; Lawrence & Dunaway, 2015). It is possible that it will be used more in order to keep the public’s attention on an election that no one wanted. As such, the following hypotheses were formed:
H3: Human Interest frame is applied more frequently in tabloids than broadsheet newspapers.
H4: Human interest frame is implemented more frequently in articles published online than articles published in the print edition.
The above hypothesis suggesting the human interest frame is implemented more frequently in online articles than print is based on research that found the strategic game frame is used more frequently in online outlets as it requires less time and money (Schmuck et al, 2017). This hypothesis suggests that online outlets, because they lack time and money as well as facing greater competition in the online environment, are more likely to use human interest frame to gain clicks and decrease the distance between the reader and the subject, in this case elections (Schmuck et al, 2017; Strömbäck & van Aelst, 2010).
Economic consequences frame is another generic frame that is identified in Semetko and Valkenburg’s (2000) framing analysis which is characterised by reporting on an issue or event with a focus on the consequences it will have economically on a group, country or individual. Given economic consequences are often widespread, and the impact an election has on the economy could be great, the use of such a frame could be substantial within the case of the 2017 general election. Previous research found it to be significantly more used in higher quality newspapers (Semetko & Valkenburg, 2000). With this in mind, the following hypothesis is introduced;
H5: economic consequences frame is implemented more frequently in broadsheets newspapers than tabloids
The third and final generic frame that will be analysed in this study Is attribution of responsibility. This frame, whilst appearing in limited framing research, has been explored more in the United States where the media is reliant on blaming a group, an organisation or an individual in a story or problem (Semetko & Valkenburg, 2000). Given the growing trend of the U.K toward American style reporting, it is possible that such a frame will be present in British coverage. The same study that found economic consequences frame to be used more frequently in broadsheets than tabloids, also found evidence that the attribution of
responsibility frame is used more frequently in broadsheets ( Semetko & Valkenburg, 2000). As such, the following hypothesis is identified:
H6: attribution of responsibility is more frequently implemented in broadsheet newspapers than tabloid newspapers.
Strategic game frame and Election coverage
Previous research about framing in election coverage have almost exclusively found that strategy or game frame is the most commonly used in election coverage (Strömbäck & Dimitrova, 2006; Strömbäck & van Aelst, 2010). Following on from this, some academics even lump the two together thereby labelling it as the broader frame ‘strategic game’ which encompasses many aspects of either frame including horse-race journalism, focusing on strategies or tactics of political actors and obsessing over public opinion polls (Scammell & Semetko, 2008; Dimitrova & Kostadinova, 2013; de Vreese, 2005). This study focuses on the strategic game frame which has been commonly found in campaign and electoral coverage (Lawrence, 2000; Schmuck et al, 2017; Dunaway & Lawrence, 2010). This generic frame encompasses a common occurrence in electoral coverage by journalists which include reporting politics as a strategic game that politicians play to get ahead and do this through certain strategies and tactics in campaigning (Lawrence, 2000). This is problematic for the public who rely on the press for information in election periods (Lawrence, 2000). This is in conflict with the essence of journalism; which is to inform the public (Lawrence, 2000). It deprives the public of political information they can then use to formulate informed opinions (Lawrence, 2000). The use of strategic game frame has also increased over the years and fits in well with the press’s tendency to focus on drama and conflict in news (Lawrence, 2000). Furthermore, the UK has experienced a trend toward the increasing use of strategic game frame; a 2005 content analysis of media coverage during the election found 44% of it was concerned with the electoral ‘process’ and just over half of this covered political strategies
during the election (Scammell & Semetko, 2008). A convergence towards a US media model that is market driven and competitive is evident and is only exacerbating the upward trend toward strategic game frame coverage (Scammell & Semetko, 2008). Previous comparative research found that British newspapers in fact were more likely to frame politics as a strategic game than their counterparts in Sweden, where coalitions are common and therefore
competitive, sports-like election coverage is lower (Strömbäck & Shehata, 2007). More generally, strategic game frame has been consistently identified as the most commonly used frame in election coverage and has only shown signs of media outlets increasingly used it to grab the audience’s attention and save economically (Lawrence, 2000). From this we
introduce the following hypothesis:
H7: Strategic game frame is the most frequently implemented frame in electoral coverage.
This frame is most heavily researched in the US and is also the most commonly used, with strategic game frame synonymous with horse-race journalism which is much loved on their side of the Atlantic (Patterson, 2017). Furthermore, a clear increase has been seen in the journalist’s reliance on polls since the 60s (Patterson, 2017). This study will look at whether this trend continues in this election.
Framing and Media characteristics
Previous scholars have also delved into how media characteristics such as the format and type of media outlets relates to the use of different frames (Strömbäck & van Aelst, 2010; Schmuck et al, 2017; Strömbäck & Shehata, 2007). This includes studies that research
framing in different types of newspaper outlets as well as commercial versus public broadcasting television channels (Schmuck et al, 2017). Several studies have looked at the public broadcasting versus commercial and upmarket or broadsheet versus tabloid
newspapers. However, findings have been mixed with some studies supporting the idea that tabloids use strategic game frame more frequently than broadsheets and other studies failing to find such proof (Schmuck et al, 2017; Stromback & van Aelst, 2010; Lawrence &
Dunaway, 2015). Studies have suggested commercialism is one of the driving forces behind the rise in strategic game frame (Schmuck et al, 2017; Stromback & van Aelst, 2010). This claim was first linked to commercial TV news and Public Broadcasting news but the same was expanded to include tabloid and broadsheet newspapers (Strömbäck & van Aelst, 2010). This is based on the clarification that while both tabloids and broadsheets are commercially run, the former relies almost exclusively on newsstand sales for revenue thereby having to hold the audience’s attention every day (Strömbäck & van Aelst, 2010). A distinction was also made in analysis of the strategic game frame in news outlets concerning economic gains and the market place, suggesting that the market penetration of a newspaper or ownership has an influence on their use of game frame in coverage (Dunaway & Lawrence, 2015). Whilst findings on ownership and competition in economic terms have been limited and conflicting, there has been evidence found to support this (Stromback & van Aelst, 2010; Dunaway & Lawrence, 2015). With this in mind, the following hypothesis is introduced:
H8: Strategic Game frame is implemented more frequently in tabloids than broadsheet newspapers.
The same research suggests, much like previous studies, that highly competitive elections are more than twice as likely to produce game frame coverage of elections (Stromback & van Aelst, 2010). Similar findings were found concerning election day; as it nears strategic game frame coverage increases (Dunaway & Lawrence, 2015, Sinclair, 1982). Here we introduce the following and final hypothesis:
H9: The use of strategic game frame increases in British newspapers as election day nears.
Methodology
A quantitative content analysis was conducted of the most popular British
newspapers, with an equal number of broadsheets and tabloids as well as left and right wing political leaning. This method was chosen based on its ability to interpret phenomena across different media outlets and translate it numerically (Fico, Lacy & Riffe, 1998). In this study, four generic frames were measured in the election coverage of the three biggest political parties- Labour, Conservatives and Liberal Democrats. It is often paralleled with other methods of observing behaviour similar such as asking individuals to respond to scales (Fico, Lacy & Riffe, 1998). Frames are measured through ‘latent’ variables made up of several items within a codebook which, once computed to create a new variable, measure the overall frame (Fico, Lacy & Riffe, 1998). A codebook was formulated using mainly items from previous framing studies, with relatively few items being created exclusively for this study (Semetko & Valkenburg, 2000; Bartholomé, Lecheler & de Vreese, 2017).
Sample.
The constructed week sample was created from newspaper articles between the time period of April 19th and June 8th, 2017. Seven weekdays were chosen within this period, all representing a day of the week. This sampling method included one Monday, one Tuesday and so on for all days of the week to account for variation in volume and topics on different days and during different times of a month (Hester & Dougall, 2007). This reduced the likelihood of oversampling the weekends for example or events such as the two terrorist incidents that took place at the end of May and early June in London and Manchester respectively (Fico, Lacy & Riffe, 1998). It is also a more cost-efficient and unbiased approach to sample daily newspapers for analysis (Luke, Caburnay & Cohen, 2011).
The sample included 7 British newspapers, chosen based on their audience reach both online and offline as well as for their political leaning and newspaper format. The top
newspaper titles with the biggest audience reach both through their website and print editions - Daily Mail, The Sun, The Guardian, The Daily Telegraph, The Daily Mirror and The Times - are included within our sample (Monthly reach of.., 2017). Within this, three right-wing newspapers were included along with three left-wing newspapers as well as one centrist title. From this a constructed week sampling was conducted and articles were found through the online academic data base Lexis Nexis. The decision to code newspapers and not television programs lies in previous findings that suggest despite an overall decline in newspaper readership over the decades they still play a pertinent role in setting online and television agendas (Scammel & Semetko, 2008; Schmuck et al, 2017). Whilst print newspapers are a ‘dying breed’ in the digital era, it is still the third main source of news for the British public and still hold considerable power (Nielsen, 2017; Scammel & Semetko, 2008).
Data collection.
The constructed week sample was created from newspaper articles between the time period of April 19th and June 8th 2017; the day after the election was called by Theresa May and the day of the election. The articles were collected using the newspaper database Lexis Nexis. Search terms which helped narrow down the sample size include Labour election OR Conservative! election OR Liberal Democrat! election OR Labour polic! OR Conservative! polic! OR Liberal Democrat! polic! OR Labour manifesto! OR Conservative! manifesto! OR Liberal Democrat! polic! Once all articles were collected, each article was read at least once and coders were asked to code in an excel file using the codebook (see appendix). A second coder was used to calculate Inter-coder reliability.
All Four frames were measured using binary variables (0=absent/1=present) from previous framing studies in national media (Valkenburg & Semetko, 2000; Bartholomé, Lecheler & de Vreese, 2017). Other items (15a-15h) were created to look at the frequency of political parties quoted saying negative or critical statements towards another party in
newspaper articles. However, the inter-coder reliability was not calculated and not used in analysis. The decision to use already tested items was taken to ensure that the items
successfully measured the chosen frames. The strategic game frame was measured using five items that all measure different dimensions of strategic game frame that are outlined in Aalberg, Strömbäck and de Vreese’s review (2011). These include the use of opinion polls, discussing the notion of ‘winners’ and ‘losers’ as well as analysing a politician’s performance and or strategies (Aalberg, Strömbäck & de Vreese, 2011; Semetko & Valkenburg, 2000). Human interest frame was measured using four items, all dichotomous, taken from Semetko and Valkenburg’s 2000 content analysis of Dutch television and newspapers. Economic Consequences frame and Attribution of Responsibility frame were both measured using three items with dichotomous ansering categories (Semetko & Valkenburg, 2000). The entirety of the codebook can be found in the appendix, with instructions, definitions of the frames and all the items coded (see appendix). The unit of analysis in this study is the entire newspaper article. Items with descriptive text entries were initially included after three of the frames in the codebook, in an attempt to link the use of certain frames to specific political parties, however given a lack of literature they were not used in later analyses.
Four new variables were computed from items that measured each frame. The human interest frame items 17a-17d were computed to create the scale variable ‘human interest frame’. The economic consequences frame variable that was computed was made up of items 12a-12c that can be seen in the codebook (see appendix).The attribution of responsibility frame was measured using three items 11a-11c with all dichotomous items The final frame,
strategic game frame was made up of five items which include items 13a-13b and 14a, 14c and 14f in the codebook (see appendix). Lastly, A new variable was computed to include all other political parties that are not Labour, Conservatives or Liberal Democrats. This includes items 10d-10g in the codebook, which were computed into the new variable ‘other political parties’. This was done because this study is not concerned with minority political parties.
Coding.
Inter-coder reliability was calculated for all content variables, with roughly twenty percent subsample of articles being coded by a second coder (N=29). The average
krippendorff’s alpha for all variables was calculated(a=.80) and is line with Krippendorff’s .80 minimum that is considered acceptable (Krippendorff, 2011). Inter-coder reliability was calculated for all variables with a range varying from .72 to 1(Appendix table.1). Some variables which achieved an alpha of 1 were easily recognisable such as the mention of a topic or presence of a political party. Less clear variables such as ‘Is the performance of the party/leader/politician/party discussed?’ had lower alpha(a=.71) due to the nature of the question of being more subjective. Not all items coded were used in analysis, with only those included in table 1 being used(see appendix).
Results
The first hypothesis tested the hypothesis that Labour and the Conservatives were the most mentioned political parties in British newspapers. This was tested by running
frequencies on variables measuring the presence or absence of Labour, Conservatives and the Liberal democrats. A new variable was computed which included all other political parties coded and was named ‘Other political parties’. Within the total sample, Conservatives were the most commonly mentioned party with it being mentioned in 82% of all newspaper articles. Labour are not far behind with 78%, however there was very little mention of the
Liberal Democrats who lagged behind on only 19%.. There was also very little mention of other political parties with only 14% of newspaper articles. This variable included all other parties in running such as the Greens, UKIP, Scottish National Party and Plaid Cymru. This shows that the Conservatives and Labour are the most mentioned political parties within the sample and we can therefore accept our first hypothesis that the Conservative and Labour party were the most discussed political parties in the 2017 General Election.
The second hypothesis tested whether or not strategic game frame was more frequently implemented in online articles or print articles. A one-way ANOVA was
conducted to test this. The Levene’s statistic was not significant and therefore homogeneity of variances can be assumed (F= 2.774, p = .65). The ANOVA was not significant, showing that there is not statistically significant difference between the use of attribution of
responsibility frame in print versus online articles F(2,205)= 1.896,p = .153. This hypothesis was based on several studies which found that online formats used significantly more
strategic game frame because they lacked time and resources to do further research into the issue and this frame both appealed to a greater audience and glossed over the issue (Semetko & Valkenburg, 2000; Shmuck et al, 2017).
The third hypothesis regarding human interest frame being used significantly more in tabloid newspapers than broadsheet newspapers was tested by conducting a One-way ANOVA. The newly computed variable ‘human interest’ frame was created using the four dichotomous items in the codebook taken from Semetko and Valkenburg and can be seen in the appendix under codebook (2000). The Levene’s test was not significant (F=1.283, p=.259) and therefore homogeneity of variances can be assumed. Overall, the one-way ANOVA was not statistically significant, we must reject our hypothesis
F(1,205)=.686,p=.409.This shows that overall there was not a statistically significant difference between the use of human interest frame in tabloids and broadsheets.
The fourth hypothesis looked at whether the use of human interest frame was more common in online articles compared to print articles. A one-way ANOVA was conducted. The dependent variable was ‘human interest frame’ and independent variable as ‘print or online’. The Levene’s statistic was not significant and therefore homogeneity of variances can be assumed (F=1.766, p=.174). The ANOVA was statistically significant, meaning there is a statistically significant difference between the three categories and their use of ‘human interest’ frame F(2,204)=7.948, p<.001. The effect size was calculated (h2= .072) and found that there was a small to medium effect size of print versus online on the use of human interest frame. A Bonferroni post-hoc test was used to uncover where the differences lie and there was a statistically significant difference between ‘Online’(M= 2.53, SD= .085) and ‘Print’ (M= 2.16, SD= .93, p=.009), ‘Print’ and ‘Not Specified’ (M= 4, SD= 0, p=.013) but not between ‘Online’ and ‘Not Specified’.
The fifth hypothesis tested whether the economic consequences frame is more common in broadsheet newspapers than tabloids. The newly computed variable ‘economic consequences’ frame was created from the three items 12a-12c that can be found in the codebook (see appendix). A one-way ANOVA was conducted. the Levene’s test was not significant and therefore homogeneity of variances can be assumed (F= 2.886, p=.091). There was not a significant difference in the means of tabloids and broadsheet, meaning that the economic consequences frame was not used significantly more common in broadsheets than tabloid newspapers. F(1,206)=.007,p =.934.
The sixth hypothesis based on previous literature that suggests quality or more serious news outlets are more likely to use attribution of responsibility than tabloid newspapers. A one-way ANOVA was conducted to test this hypothesis; attribution of responsibility frame was the dependent variable and ‘newspaper type’ the independent variable. The Levene’s test was not significant and therefore homogeneity of Variances can be assumed (F= 2.326, p=
.129). The one-way ANOVA was not statistically significant, which shows that there is not a statistically significant difference between the use of attribution of responsibility frame in tabloids and broadsheets F(1,206) = 2.321, p = .129.
Previous studies suggested that strategic game frame is the most commonly used frame in election coverage (Aalberg, Strömback & de Vreese, 2011; Scammel & Semetko, 2008); Azrout, Boomgaarden, Elenbaas, Schuck, van Spanje, Vliegenthart & de Vreese, 2013). The seventh hypothesis tested to see whether this is the case in the 2017 British general election. All newly computed variables that were created to measure strategic game frame, human interest frame, economic consequences and attribution of responsibility were recoded to dummy variables (0=absent/1=present). Frequencies were run on the newly computed dummy variables that measured the presence or absence of the frames. Table 1 shows that the human interest frame was the most frequently implemented frame in election coverage, followed by strategic game frame. Our hypothesis is not supported and must therefore be rejected; strategic game frame is not the most commonly found frame in election coverage.
Table 1. Frequencies of four frames coded as present or absent
Strategic game frame human interest frame economic consequences frame attribution of responsibility frame (0) Absent 21.3% 3.9% 46.2% 31% (1) Present 77.3% 96.1% 53.8% 69%
The eighth hypothesis was based on previous research suggesting that strategic game frame is more likely to be used by tabloids due to the nature of its content and lack of
‘serious’ journalism comparative to broadsheet newspapers (Strömbäck & van Aelst, 2010). From these findings, the eighth hypothesis was created suggesting that the strategic game frame will be more commonly implemented in tabloids than broadsheets. A one-way ANOVA was conducted on the dependent variable ‘strategic game1’ and ‘Newspaper type’ as the independent variable to test this. The Levene’s statistic was not statistically significant and therefore homogeneity of variance can be assumed (F= 1.577, p =.211). Overall, the ANOVA was not statistically significant; the hypothesis must be rejected F(1,206)=1.094, p=.297. The strategic game frame was not used significantly more in tabloids than
broadsheets newspapers.
The ninth and final hypothesis looks at the relationship between the proximity of election day and the use of strategic game frame, stemming from research that found the closer election day gets the more strategic game frame is used. Date was recoded; 1 was coded as the date furthest away from election day (19/04/2017) and 7 as the closest to election day (5/06/2017). Following on from this, a one-way ANOVA was used for analysis. The Levene’s test was not statistically significant and therefore Homogeneity of variances can be assumed (F= .939, p=.468). The ANOVA showed that there were statistically significant differences between the seven categories, meaning that the use of strategic game frame did vary between the dates F(6,201) = 7.194, p < .001. Our hypothesis can be
therefore partially accepted. The results suggest that the use of strategic game frame decreased as election day nears. The effect size was calculated for one-way ANOVA (h2= .18) and only 18% of variance was caused by the date, which is a medium effect size. A bonferroni post-hoc test indicated that there was a statistically significant difference between day 1 (M= 3.8, SD= 1.15) and day 3 (M= 2.58, SD= 1.28, p= .021), day 1 and day
4 (M= 1.77, SD= 1.35, p<.001), day 1 and day 5(M= 2.35, SD= 1.39, p=.017), day 1 and day 7 (M= 2, SD= 1.23, p<.001) and day 4 and day 6 (M= 3, SD= 1.48, p= .001). Whilst the ANOVA is statistically significant, the means show that as election day nears, strategic game frame decreases. Therefore, while our hypothesis is partially accepted because the ANOVA is significant, the use of the frame decreases as election day nears meaning we must partially reject our hypothesis. This result could be partially explained by the sample which was skewed on several variables including broadsheet versus tabloids, print versus online and political ideology. It is also possible that as the election period progressed, the coverage focused more on substance or policies rather than campaign tactics and opinion polls.
Discussion
This study set out to analyse the use of four generic frames in the case of the 2017 British general election. The initial hypothesis regarding the most discussed political parties in British newspapers supports previous findings that suggest Britain’s First Past the Post system, which favours the biggest political parties, spills over into election news coverage resulting in very little coverage being given to minority parties (Scammell & Semetko, 2008). This is in stark contrast to the 2015 British election in which minor parties gained
significantly more coverage than in 2017 (Deacon et al, 2017). In this sample, a clear bias for the Labour and the Conservatives can be found in the coverage, with the Liberal Democrats gaining much less coverage than previous years (Deacon et al, 2017). A new variable was computed to account for all other political parties- Greens, UKIP, SNP, Plaid Cymru - which were rarely mentioned.
The seventh hypothesis stemmed from previous findings that suggested strategic game frame was the most commonly used frame during election time (Strömbäck & van Aelst, 2010; Scammel & Semetko, 2008). This study did not find support for this and instead
found that human interest frame was the most commonly used frame. This could have been the result of recoding the four frames into dummy variables. The recoded variables took any value above 1and recoded it into 1, meaning that if one of the items was present the frame would be coded as present. For the ‘human interest frame’, there was an item ‘Does the news article quote a political actor/party/leader/individual?’ which was coded present a large number of times (N=180) and this could have skewed the presence of the frame overall in the sample. Recoding the four variables that measured each frame differently is an option,
however when this was done and frequencies were run again human interest frame was still the most commonly found frame (88.9%). Other possible answers can be found in Strömbäck and van Aelst’s study on framing during election periods in the Belgian and Swedish media (2010). Their research suggested that the strategic game frame and human interest frame are one of several ways the press attempt to push political and electoral news on less-eager audiences (Strömbäck & van Aelst, 2010). The United Kingdom in recent years has had a political apathy crisis, specifically young voters, and the use of these two common frames could be a way by the media to make them more aware and politically engaged with electoral news. The fifth hypothesis concerned the use of economic consequences frame in tabloids and broadsheets and did not find significant difference between the two groups, contradictory to theory. The lack of a significant difference between the two groups might be explained through a lack of interest in the economy during this election. ‘It’s the economy wot won it’ might not apply here both in the media and the public. In previous research looking at election coverage in the same election, the economy was the sixth most prominent issues, behind issues such as health, defence and the electoral process (Media coverage of the 2017 General Election.., 2017).
The second, seventh and eighth hypotheses were all rejected, with the analysis showing that not only was strategic game frame not the most commonly used frame during
this election period, but that its use in tabloid and broadsheet as well as online versus offline was not statistically significant. This could be due to the sampling method resulting in a significantly larger number of broadsheets being coded compared to tabloids. It could also be linked to previous literature claiming that the more competitive an election, the more likely it is to result in the use of strategic game frame (Schmuck et al, 2017; Strömbäck & Shehata, 2007).Regarding a lack of significant findings concerning the more frequent use of strategic game frame in low-quality or tabloid newspapers, research findings were mixed with some that found support for this and others finding that both strategy and ‘personal attacks’ are significantly more prevalent in quality or broadsheet newspapers (Strömbäck & van Aelst, 2010; Bartholomé, Lecheler and de Vreese 2017). It is also possible that the lack of
significant results could be linked to a more general finding within the press this election took place in a very diverse political and social context. Other factors could have worked. The election itself was a snap election and therefore unexpected which might have affected how the media approach it. Furthermore, it was called in a ‘post-Brexit’ Britain in which the media have been increasingly criticized for its coverage of politics. Finally, our hypothesis suggesting that human interest frame is more commonly used in online articles than print was supported. This means that online articles did use the human interest frame significantly more than print articles. Whilst previous literature did not look at this, it is supported by findings that suggest human interest frame is used significantly more by tabloids (Schmuck et al, 2017, Strömback & van Aelst, 2010). Online outlets often have less time, are more competitive and less economic means to research a story; the use of human interest frame could be a way to approach some of these problems (Schmuck et al, 2017).
This content analysis study contributed to literature on the use of framing during election periods in several ways. A framing analysis on the British general election
election periods(Jackson, 2011). The findings in this study did not find support for previous studies that strategic game frame is firstly more common in tabloid newspapers. Secondly, it is more common in competitive elections or proportional representative government systems and finally that it is increasingly common as election day nears (Schmuck et al, 2017;
Strömbäck & Shehata, 2007). This study did not find support for any of these notions. The low-quality newspapers included within this sample were not found to use more strategic game frame compared to high-quality newspapers, which Is similar to findings by Schmuck et al (2017). who also failed to find support for this hypothesis in the cases of Switzerland and Austria The lack of significant findings could be the result of a biased sample, with a significantly higher number of Broadsheet newspaper articles being coded (N=157) than Tabloids (N=51). A greater sample size with a more balanced number of articles coded from either type could yield greater results.
Limitations.
One of the main limitations of this study is the use of a constructed week sample. The sampling method, chosen based on its ability to produce a balanced sample that removes the possibility of overrepresentation of certain days of the week and certain events was not representative across all the variables (Luke, Caberney & Cohen, 2011). The resulting sample had an overrepresentation of broadsheet newspapers (N=157) compared to tabloids (N=51) as well as a high number of articles from a ‘centre’ wing newspaper(N=51) compared to left (N=96) and right-wing(N=61). In fact, this created a skewed sample on variables such as ‘newspaper type’ which are of importance. It proved problematic given that these variables were vital in the analysis. This could explain the lack of significant findings in terms of the use of frames and the newspaper format or type supported by previous literature. A different sampling method may yield more promising results. Such a sampling method is intended to avoid overrepresentation of certain days of the week and - newspaper type, print/online and
political position - which could explain the lack of significant findings which align with previous studies. Furthermore, a greater sample size could increase the likelihood of significant results (N= 208).
The analysis used to falsify the fourth hypothesis, which suggested that the strategic game frame is the most commonly used frame In electoral coverage, recoded all four frame variables into new dummy variables (0= absent/1= present). This was done in order to be able to see which frame was the most used regardless of the number of items coded. Whilst it was the simplest option to find the most common frame, it could explain the finding that the human interest frame is the most commonly used frame in election coverage. The item ‘Does the news article quote a political actor/party/leader/individual? Which is used to compute the new variable ‘human interest’ frame was coded as present a lot of the time meaning this could have increased the ‘presence’ of the frame overall. This could explain why it overtook strategic game frame as the most commonly used frame in election coverage.
Conclusion and Future Research.
Possible future research in this field could result in different findings more in line with previous literature which suggests strategic game frame is the most common frame in electoral coverage. Another study conducted with a different sampling method may yield more promising results. Furthermore, a comparative angle could show a difference in election coverage of British newspapers over time or across countries. The United States for example is often and usually inevitably compared to the United Kingdom; in the case of election coverage this could be helpful in looking at already identified trends of the latter following the lead of the former in reporting styles and framing (Scammell & Semetko, 2008; Nielson, 2017). Despite some studies comparing the United Kingdom and European countries, it is limited research and more could be done in this area. Given the lack of significant findings, a future study looking at the strength of the same frames in the election coverage of different
British newspapers during the same election, using the same independent variables of
newspaper format and online versus offline could give some insight into the strongest frames. This could identify how often certain items that measure a frame come up in an article and not just the presence of it overall. As an alternative to this framing analysis of British newspapers, a content analysis on the television debates during the campaign period could prove integral to see if similar results are found concerning the same frames. This research, whilst it did not yield significant findings for all hypotheses, it identified the prevalence of certain frames in the 2017 election coverage in British newspapers and can be a stepping stone in future research in the case of British election coverage.
Appendix Table 1.. Intercoder reliability for Content variables
Variable name Krippendorff's alpha (a)
V5(newspaper type) 0.9116
V6(print/online) 0.8183
V10a(party: labour) 0.8417
V10b(party: cons) .8403
V10c(party: Liberal Democrats) 0.8403
V10d(party:Green) 1 V10e(party:UKIP) 1 V10f(party:SNP) 1 V10g(party:Plaid Cymru) 1 V11a(attr. Resp.) 0.8304 V11b(attr. Resp) 0.7206 V11c(attr. Resp.) 0.7697 V12a(eco. Cons) 0.7967 V12b(eco. Cons) 0.7957 V12c(eco. Cons.) 0.7799 V13a(strat.) 0.7121 V13b(strat) .7286 V14a(confl.) 0.7547 V14b(confl.) 0.7927 V14c(confl.) 0.779 V14f(confl.) 0.7849
V17a (human interest) 0.8403
V17b (human interest) 0.8630
V17c (human interest) 0.7957
Codebook
Codebook for Content Analysis conducted on British newspapers articles:
Framing in the 2017 British General election in British newspapers
Department of Communication Science Thesis Supervisor: Carlos Brenes Peralta
Introduction
Thank you for taking part in this research. This study is done in partner with the University of Amsterdam. This content analysis intends to analyse the frames most commonly used in the newspaper coverage of political parties in the 2017 British General election between the 18th of April and election day, the 8th of June. These dates were chosen because Theresa May, the current Prime Minister, called the election on the 18th of April and election day was the 8th of June and defines the campaigning period. The three biggest political parties from previous elections in 2010 and 2015 were chosen. This is because the United Kingdom is a dual-party system, favouring the dominating parties. These political parties include; the Labour party, Conservatives and the Liberal Democrats. These three were taken because they are the three parties with the biggest vote share. The British electoral system is historically a two-party system but in recent years has evolved, with the Liberal democrats gaining votes and partnering up with the Conservatives in 2010 for the first
Coalition in almost fifty years. News articles will be taken from The Sun/The Sun on Sunday, The Daily Mail/The Mail on Sunday, The Guardian/ The Observer, The Independent, The Times, The Daily Mirror/Mirror on Sunday. These titles were chosen because they first three have the biggest circulation numbers and the Independent is the fifth most visited newspaper site (the second more left-leaning and progressive newspaper after The Guardian). The newspaper titles were chosen based on their audience reach both online and print, with the Daily Mail being the most visited newspaper website in 2016-2017 and the Sun having the biggest circulation numbers in 2017 (Statista, 2017). All other titles were also in the top 10 for either circulation number or most visited website, save the Independent which no longer runs print editions. The items used to code the chosen frames will be coded as (1) Present, (0) Absent. Given the unit of analysis is the whole news article, multiple frames and political
actors can be identified in each article. If this is found to be the case, code all frames and actors that are identified.
Any further questions on this research may be directed to: ceri.taylor@student.uva.nl
Central Concepts: British newspapers titles
The British press are partisan, with a slightly right-leaning press. In the most recent elections in 2017 most newspapers supported Theresa May. The seven newspapers chosen for this study are amongst the most popular with the biggest daily reach both online and offline. The title with the biggest daily reach is the Daily Mail, a right-wing newspaper. Two other right-wing newspapers were chosen, The Daily Telegraph and the Sun, along with three left-wing newspapers who have much lower circulation numbers. These include the independent, which recently went through a digital transitional period to an online only newspaper, The Guardian, which holds a digital-first policy but still maintains print editions, and the daily Mirror. One of the few centre newspapers is also included: The Times. In the United Kingdom, tabloids are the most read newspaper formats. This format is aimed for the mass public, reporting in a sensationalist style and favouring simplified reporting of political issues with a focus on infotainment (Schmuck et al, 2017). The two names stem from the style they were printed, with tabloid typically being narrower than broadsheets, meaning their stories tended to be shorter too. Meanwhile, broadsheets otherwise labelled upmarket, encouraged more traditional approach to news gathering and more in-depth analysis. They also tend to have a more affluent and educated readership. For Clarity, the given Newspaper titles will also be categorised into either tabloid or broadsheets.
Political Parties
This refers to the three biggest political parties in the United Kingdom; The Labour party, Conservatives and Liberal Democrats. This focus on only three of the parties is linked to the UK being originally a two-party governmental system until recently with a majority being either in the hands of Labour or Conservatives. The 2010 Conservative-Liberal Democrat coalition was the first time in almost four decades. Other parties are also included but are not of interest in our findings. For clarification below is a list of well-known politicians from either party.
Conservative party:
Theresa May (Prime Minister) Boris Johnson (Foreign Secretary) Michael Gove
David Cameron (Previous Prime Minister, 2010-2016) Philip Hammond (Chancellor of the Exchequer)
Jeremy Hunt (Healthy Secretary) Damian Green
David Davis (Secretary of State for exiting the European Union) Amber Rudd Labour party: Jeremy Corbyn Diane Abbott John McDonnel Andy Burnham Yvette Cooper Chuka Umunna
Ed Miliband (previous leader of Labour until 2015) Tony Blair (previous PM 1997-2007)
Gordon Brown (previous PM 2007-2010) Sadiq Kahn (Mayor of London)
Liberal Democrats:
Tim Farron (current leader of the Liberal Democrats)
Nick Clegg (previous leader of the Liberal Democrats 2007-2015) Vince Cable
Political Spectrum
A political spectrum is a system of classifying different political positions upon one or more geometric axes that symbolise independent political dimensions. Within this left and right wing are either extreme of the spectrum, communism and socialism being traditionally on the left and conservatism and capitalism on the right. The newspaper's outlined above are equally distributed across this spectrum. If the political leaning of a newspaper cannot be identified, please refer to the definition given above of British newspapers titles for clarification before coding.
Other definitions: Political actors
Political actors refers to not only a political figure or leader but can include political parties, governments or organizations.
Further definitions of concepts are briefly given below, next to the items that measure them in order to speed up the procedure. However, if there any other misconceptions please refer back to this page for clarification.
Framing
This refers to the process through which journalists or media outlets select or
emphasise certain aspects or perspective of an issue or event to promote a certain problem definition or solution to the issue at hand (Entman, 1993). Examples of this include emphasising the winning and losing aspect of elections, focusing on a specific individual in an issue or event. In this study, four
frames will be analysed. These include human interest, economic consequences, attribution of responsibility, moral and strategic game frame. Four of these frames and the items that measure them are taken from Semetko and Valkenburg's 2000 study analysing how the 1997 Amsterdam meeting of European heads of state was framed. The frames will be briefly defined below for clarity.
Human interest frame
This frame attempts to personalise or emotionalise the story through putting a face to the story thereby decreasing the distance between the audience and the story. Examples of this include a story about Jeremy Corbyn's current leadership of the Labour party which discusses his past actions concerning political protests and his previous ties to Hezbollah.
Economic Consequences frame
This frame covers a story or issue based on the possible consequences it will have economically on an individual, region or country. The general impact of an event is always an important news value and the economic impact is always considerable (Semetko & Valkenburg, 2000).
Attribution of Responsibility frame
"This frame attempts to cover a problem or issue in such a way as to attribute responsibility for its cause or solution to either the government or to an individual or group." (Semetko & Valkenburg, 2000; 96). This frame is largely seen in television news, where "news stories are about specific events or particular cases" but given the rise of personalisation and critical journalism this may have increased (Iyengar, 1991; 2 in Semetko & Valkenburg, 2000; 96).
This frame is broader than those previously mentioned, with subframes included in this generic frame. These include horse-race frame, conflict frame and of course strategy and game frame. "This frame portrays elections as an ongoing contest or horse-race between candidates and parties and is often compared to typical sports coverage. The strategic game frame focuses on the winners and losers and highlights is often compared to typical sports coverage"(Dimitrova & Kostadinova, 2013;77). The frames explained below are encompassed within the broad frame of strategic game frame but are treated as independent frames in previous research (Valkenburg & Semetko, 2000)
Horse-race frame (synonymous with Game frame):
This frame is based on the idea of a horse race, meaning that coverage of elections for example focus purely on the winning and losing aspect of the election. This means that a lot of the coverage is based on how candidates are doing in the election, along with public opinion data surrounding their performance.
Strategy frame:
This frame is often linked with game frame and put together to create the strategic game frame as identified above. This specific aspect focuses on the candidate's or political parties’ motives for political actions, tactics to reach political goals, campaigning and how news media cover the political process (Schmuck et al, 2017).
Conflict frame
This frame emphasises the conflict between individuals/parties/governments/countries as a way to gain the audience's attention (Semetko & Valkenburg, 2000). It is amongst the most commonly used frames, especially in political news but has been blamed for
oversimplifying complex political issues.
Procedure
Please answer the following questions to the best of your ability for every news article coded. The Unit of analysis is the entire article and all items must be coded. The 3rd option (2) not clear is available but should be avoided if possible. If certain news articles are not clear, code them to the best of your ability and identify it as a difficult case in the accompanying excel file. Read the news articles carefully and only begin coding once the article has been read at least once. If any questions or answers are unclear read over the definitions of the concepts. Any further issues contact the email given in the introductory page.
General Information V1. Coder ID:
Coder ID: (1)= Ceri Taylor (2)= Second Coder ___
V2. Text ID: ___
This is to identify every news article that is coded and can be found in the accompanying excel file and word document containing the sample of news articles. The first article is given text ID 101 and continues onwards in numerical order, i.e. 101, 102, 103, 104, 105…
Contextual Variables V3. Newspaper title
Table categorizing titles into Tabloid/broadsheet and identifying political position (C) = Centre (L) = Left (R) =Right
What is the Title of the Newspaper you are coding? (1) The Independent
(2) The Daily Mirror (3) The Sun
(4) Daily Mail (5) The Guardian (6) The Daily Telegraph (7) The Times
V4. Political position
What is the Newspaper source's political leaning? This refers to the political spectrum of being left or right wing and the newspapers are labelled as such in the table under Newspaper type (Tabloid versus Broadsheet)
(1) Left (2) Centre (3) Right
V5. Is the Newspaper title considered a tabloid or broadsheet?
Broadsheet Tabloid
The Times (C) The Sun(R)
The Daily Telegraph(R) Daily Mail(R)
The Guardian(L) The Daily Mirror(L)
Tabloid (1) Broadsheet (2)
V6. Format of the Newspaper edition.
This refers to where the news article appeared in the daily print edition of the title or the online edition. Please note one of the titles (The Independent) is a digital only newspaper and must therefore always be coded as online.
(1) Online (2) Print
(3) Not Specified V7. Newspaper section
What Newspaper ‘section’ does the news articles appear in?
Please write in the box the section the news article was in. This is clearly stated in the Word document attached.at the beginning of each news article under ‘SECTION’.
___
V8. Newspaper Topic
What is the topic of the news article? If more than one topic is mentioned, code all that are mentioned. More than one answer can be coded. Electoral process here includes
campaigning, policies and manifestos, however if for example policies concerning the National Health Insurance are discussed code as electoral process (1) present and health/health care provision (1) present with all other topics coded as (0) Absent. V8a. Election/Electoral Process
(1) Present (0) Absent
(1) Present (0) Absent
V8c.Defence/Military/Security (1) Present
(0) Absent
V8d. Health/Health care provision (1) Present (0) Absent V8e. Taxation (1) Present (0) Absent V8f. Economy/Business/Trade/Budget (1) Present (0) Absent V8g. Social security (1) Present (0) Absent V8h. Immigration (1) Present (0) Absent V8i. Personal life
(1) Present (0) Absent V8j. Climate Change
(1) Present (0) Absent V8k. Other :
If the topic of the news article is not listed here, please indicate the subject by typing it in the appropriate box.
____
V9a.Date (Day) of news item/Newspaper edition _
V9b.Date (Month) of news item/newspaper edition? _
V10.Political Party
What party is mentioned in the news article? Code each variable as (1) present or (0) absent depending on whether the political party is mentioned in the news article. V10a. Labour : (1) Present (0) Absent V10b. Conservatives: (1) Present (0) Absent V10c. Liberal Democrats: (1) Present (0) Absent V10d. Green: (1) Present (0) Absent
V10e. UKIP: (1) Present (0) Absent
V10f. Scottish National Party: (1) Present
(0) Absent
V10g. Plaid Cymru : (1) Present (0) Absent
V11.Attribution of Responsibility Frame
If one or more items are present, the attribution of responsibility frame has been evoked.
V11a. Is responsibility attributed to a specific political actor/individual/group/party in the news article?
(1) Present (0) Absent
V11a.1 If (1) is coded in the previous question, please indicate the political actor that is attributed responsibility in the news article by naming them. _____
V11b. Does the news article suggest solution(s) to the problem/issue? (1) Present
(0) Absent
V11c. Does the news article suggest that the problem requires urgent action? (1) Present
(0) Absent
V12.Economic Consequences Frame
If one of these items are present, the economic consequences frame has been evoked. V12a. Is there a mention of financial gains or losses now or in the near future? (1) Present
(0) Absent
V12b. Is there a reference to economic consequences of pursuing or not pursuing a certain course of action?
(1) Present (0) Absent
V12c. Is there a mention of the costs/degree of expense involved? (1) Present
(0) Absent
V13. Strategic Game Frame
If one of these items is present, the strategy game frame has been evoked. V13a. Is the performance of the party/leader/politician/party discussed? (1) Present
