Newspapers, Impartiality and Television News

Drawing on a content analysis of television news and newspapers during the 2015 UK General Election along with semi-structured interviews with the heads and/or senior editors of news or politics from each broadcaster examined, we explore the intermedia agenda-setting influence of the national press during the campaign. Overall, we found similar policy-orientated agendas, with more stories emanating from right-wing newspapers and moments when front-page splashes dominated television news coverage. Many broadcasters were editorially comfortable with covering stories originating from newspapers if further context was supplied. Our findings do not point towards any deliberate political bias among broadcasters. We suggest instead that a range of structural constraints and professional routines encouraged broadcasters to feed off stories that were more likely to be supplied by right-leaning newspapers. Since news values are not politically neutral, we argue that if journalists or editors routinely rely on newspapers to help shape the political agenda it compromises their ability to make impartial judgements about news selection. Combining quantitative and qualitative analysis, we conclude, could help to better understand the editorial processes behind the selection of news and to more carefully interpret intermedia agenda-setting than large N studies can supply.


Introduction
The role and perceived power of newspapers in the twenty-first century can often appear contradictory. On the one hand, debates about the future of newspapers regularly focus on their limited lifespan, with dramatic headlines about declining levels of circulation, falls in advertising revenue and predictions about the likely end of the printed press (Carr 2014). The imminent death of newspapers is usually viewed in light of figures showing an enhanced use of online news or greater engagement with new content and social media platforms, such as BuzzFeed and Twitter. From this perspective, the editorial reach and power of newspapers appears to be fast diminishing, as people have greater choice and freedom over what news they consume. On the other hand, newspapers or print journalists remain a frequent reference point for rival news media, and a routine source for understanding how an event or issue is interpreted. Many broadcast news programmes, including heavyweight political shows, not only review the day's papers but ask journalists themselves to interpret the significance of particular stories or to comment upon "the mood" of the press in the wake of the latest political drama or breaking news story. Far from the power of newspapers being diminished, from this perspective it appears newspapers continue to play an important agenda-setting role in raising debate about the stories they select and editorially frame.
On the face of it, discussing the editorial mix of news and opinions in the day's newspapers might appear a fair and transparent way of reflecting the national "debate" or "conversation". After all, most Western democracies have what is commonly known as a "free press"-a term used to denote a newspaper's freedom of expression and editorial independence. But, as research has long shown, many newspapers are far from editorially neutral or even strive to be balanced (Bayram 2013). Indeed, many can be explicitly partisan, favouring a particular political party or policy without counter-balancing alternative views or perspectives. For broadcasters conveying the day's newspapers, this might not represent the kind of diverse mix of news and opinions a "free press" might imply. Wider influences, such as ownership and political affiliations, could play a role in shaping a newspaper's agenda (Hallin and Mancini 2004).
For broadcasters in the United Kingdom, which are legally required to report impartially, this raises important questions about how far they should be informed by newspaper coverage and journalistic commentary. The BBC's impartiality, for example, is regularly under attack from right-and left-wing perspectives, but debates are often informed by anecdotal evidence rather than a systematic review of editorial output. By comparing television news and newspaper coverage during the 2015 UK General Election campaign, the aim of this study is to explore empirically the intermedia agenda-setting role of the press. In doing so, we consider whether broadcasters' editorial decision making was influenced by newspaper agendas.
Impartiality is an important concept in this study. While generally it is a term used to convey whether news is balanced or objective, in practice it can be difficult to measure or interpret empirically (Cushion 2014). We refer to it in a relatively broad way to consider whether broadcasters covered stories pursued by left-or right-wing newspapers during coverage of the United Kingdom's 2015 General Election campaign. Or, more generally, whether editorial decision making was influenced by the news values-criteria journalists use to select one story over another (Harcup and O'Neill 2016)-of particular newspapers. This is important in a UK context because the majority of best-selling newspapers support the Conservative Party and champion right-wing policies, such as tougher austerity measures, harsher immigration laws or lowering taxes. Only a few newspapers, with a far lower combined readership than right-wing newspapers, openly back the Labour Party and support contrasting leftwing policies.
Drawing on a content analysis of television news and newspapers during the 2015 UK General Election along with semi-structured interviews with the heads and/or senior editors of news or politics from each broadcaster examined, we explore the intermedia agendasetting influence of the press during the campaign. Whereas many intermedia agendasetting studies rely largely on content analysis, we also wanted to question and interrogate the editorial judgements made by some of the United Kingdom's most senior broadcasters. While our study cannot easily determine whether television news followed the agenda of newspapers, where possible we empirically explore whether broadcasters' editorial decisions were influenced by press coverage.

NEWSPAPERS, IMPARTIALITY AND TELEVISION NEWS 163
Although a key goal of the study was to explore the relationship between newspaper agenda-setting and television news, we would acknowledge that online news or social media platforms could also editorially influence television news coverage. Similarly, competing television news outlets may influence their rival's choice of stories, as well as coverage in newspapers. But the scope of the study was limited to rigorously exploring the intermedia agenda-setting influence of newspapers on television news coverage.
Intermedia Agenda-setting: Understanding Editorial Influences Between Media Asking why certain issues gain public attention over others, agenda-setting studies represent one of the most significant areas of research not just in the discipline of journalism studies but the inter-connected fields of communication, political science and media scholarship (McCombs 2004). Evolving in theory and practice since the 1970s, agendasetting studies seek to establish empirically whether the media influenced public opinion over time on issues such as climate change, immigration or during election campaigns.
Although media impact on public opinion remains centre stage in agenda-setting debates, a connected if less-researched line of enquiry includes interpreting agendasetting between media (Rogers, Dearing, and Bregman 1993). This is the object of our study and has become known as intermedia agenda-setting, representing a new phase of agenda-setting research (McCombs 2004). As of 2008, Vliegenthart and Walgrave (2008, 861) suggested just "nine empirical studies focus explicitly on the process of intermedia agenda-setting", although they pointed out research exploring who set the media agenda-political parties, say, journalists or members of the public-might also be relevant. Of these nine studies, six were based on US media, with newspapers, magazines and television news being the main media types analysed. Online news and more alternative newspapers were included in two studies. With the exception of Boyle (2001), studies examining newspapers and television news found the former media type had some degree of intermedia effect on the latter. So, for example, in the reporting of the "war on drugs", Reese and Danielian (1989) identified the front pages of the New York Times as especially influential in framing coverage of cocaine on subsequent stories broadcast on US national newscasts. Golan (2006), similarly, suggested a relationship between the countries featured in New York Times coverage of international affairs and the following day's agendas of CBS, NBC and ABC evening news.
More recently, scholars have focused on intermedia agenda-setting in an increasingly multimedia environment (Vonbun, Kleinen-von Königslöw, and Schoenbach 2015). Of course, since the media landscape has radically expanded in the twenty-first century, it might be expected that online or social media sources have a greater intermedia agendasetting effect than "old" or "elite" media because they can publish news instantly. However, evidence of this is not clear. Heim (2013), for example, examined political blogs during the 2008 Democratic Presidential caucuses in Iowa. Rather than bloggers creating new ways of analysing the race, the study argued they were "piggybacking on the work on mainstream media, consistent with hyperlinking studies that have documented political bloggers' reliance on sources as such as the New York Times and Washington Post" (Heim 2013, 511). Meanwhile Sweetser, Golan, and Wanta (2008) identified evidence of blogs shaping the agenda of newscasts during the 2004 Presidential election. Similarly, Conway, Kenski, and Wang (2015) established that Twitter feeds of major candidates involved in the 2012 US Presidential primary contest helped shape coverage in the leading US newspapers. But both studies also found intermedia effects were not unidirectional but multidirectional, e. g. newspapers influenced Twitter and television news influenced blogging. However, Vliegenthart and Walgrave's (2008) longitudinal study of television news and newspapers between 1993 and 2000 in Belgium compared election and non-election coverage and identified intermedia agenda-setting was weaker during election times.
A more recent and comprehensive analysis of media during the 2013 Austrian election campaign argued to the contrary by drawing attention to specific characteristics that shaped the degree of intermedia agenda-setting (Vonbun, Kleinen-von Königslöw, and Schoenbach 2015). Drawing on a semi-automatic content analysis of 17 daily newspapers, 14 online news sites, seven television programmes and the national wire service, APA, they reached three key conclusions. First, online news was the main agenda-setter. Second, elite or quality media did not regularly set agendas, with issues often shaped by other media taking a lead. Third, two factors influenced intermedia effects, which included the proximity of media (a local outlet reporting, for instance) and whether it is a story not widely observed first-hand by journalists (relying on rival reporters for knowledge). Viewed in this context, macro (different mediums) and micro (locality of issues or exclusive nature of them) factors shaped the intermedia agenda-setting process. However, Vonbun, Kleinen-von Königslöw, and Schoenbach (2015) conceded that a major limitation of their findingsand other strictly quantitative agenda-setting studies-was not interpreting how coverage was followed.
As already acknowledged, our study also cannot clearly determine whether newspapers influenced broadcasters' editorial decision making. But in order to investigate more closely whether television news resembled newspaper coverage, we need to consider more qualitative ways of interpreting intermedia effects. For example, Lim (2011) developed a comparative study of US newspapers' reporting of two breaking news stories, asking whether competitors either ignored, followed or upgraded their own coverage. The dominant response was to ignore their commercial rivals, but as Lim (2011) conceded, the study only considered two fast breaking stories and just looked at newspapers rather than between different media types. Informed by Lim's approach, we intend to make qualitative judgements about some of the major stories reported by both newspapers and television news bulletins during the election.
Overall, the explanatory factors shaping intermedia agenda-setting appear both inconsistent and dependent on a range of contextual issues and characteristics that shape the scope of the study, and the relevance of their conclusions. Our brief review of the literature revealed that many studies were based on US news media, which have unique characteristics including a broadcast media that are not legally required to be impartial in coverage of politics; that recent studies tend to focus principally on online and social media breaking news stories; and that many were largely quantitative in scope comparing several media types, rather than more qualitatively exploring the degree of intermedia influence within a particular media.
In a UK context, while there has been much speculation and commentary about the influence of newspapers on broadcast news (Toynbee 2016), there has been little attempt in media or communication scholarship to explore systematically and establish empirically whether the press have had an intermedia agenda-setting influence on television news generally or during a time period such as an election campaign. Loughborough University NEWSPAPERS, IMPARTIALITY AND TELEVISION NEWS 165 (2015) examined press and television news coverage of the 2015 UK General Election and identified a similar agenda of stories between media, with many newspapers pursuing a highly partisan approach to reporting. The majority of papers clearly supported the Conservative Party, adopting an anti-Labour editorial stance and ferociously attacking its leader, Ed Miliband. We also tracked television news coverage during the UK 2015 General Election campaign, publishing a series of articles in The Guardian. While our content analysis suggested the press influenced television news reporting during the campaign, we did not explore the relationship between media content in detail or depth (Cushion and Sambrook 2015).
This study, by contrast, contributes not just a quantitative angle to intermedia agenda-setting, but a qualitative dimension by interpreting how far broadcasters operating with strict impartiality requirements conformed to or resisted the editorial power of the press. In doing so, we are able to rely not just on textual evidence but on the editorial judgements made by some of the United Kingdom's most senior broadcast journalists to understand the pressures and professional challenges they faced in their selection of news and in safeguarding impartiality.
Of course, all UK public service broadcasters do not operate with the same regulatory oversight. ITV, Channel 4 and Channel 5, for example, are commercial public service broadcasters and have specific licence agreements about their news provision, whereas Sky News is a wholesale commercial broadcaster. Commercial broadcasters are regulated in a "light touch" way by Ofcom. Fully funded by a licence fee, the BBC, by contrast, is a fully blown public service broadcaster, mostly regulated by the BBC Trust and subject to far greater scrutiny about the impartiality of its news coverage.
Our systematic study of television news-including the BBC-and newspaper coverage during the election campaign will assess the degree to which particular broadcasters pursued an agenda similar to or distinctive from the national press. After all, in theory the BBC is more likely to operate independently from its commercial counterparts who might be more tempted to follow sensationalist stories pursued by the press. Our research questions were: RQ1: How similar was the policy agenda of UK television news with national newspapers during the 2015 General Election campaign?
RQ2: Can an intermedia agenda-setting influence be identified between press coverage and television news reporting?
RQ3: According to senior editors and heads of news at different broadcasters, did newspaper coverage influence their editorial decision making during the 2015 UK General Election campaign?

RQ4:
Overall, to what extent did the press help set the agenda of different television news bulletins during the 2015 UK General Election campaign?

Method
In understanding intermedia agenda-setting, most studies use large quantitative content analyses to trace which media covered particular issues first and how far the medium shaped the message (e.g. social media break news fastest). Our approach focused on one medium's influence-the press-and was both quantitative (quantifying the symmetry of agendas between newspapers and television news) but also qualitative (considering some of the major stories in detail and interviewing either the heads of television news or senior editors about the editorial power of newspapers).
We conducted two content analyses to compare television news and newspaper coverage during the 2015 UK General Election campaign. Our content analysis of television news bulletins included BBC, ITV and Sky News at 10 pm, Channel 4 at 7 pm and Channel 5 at 5 pm, between 30 March and 6 May. This sample represents the United Kingdom's main broadcasters with varying levels of public service obligations. While the BBC has the most public service broadcasting responsibilities in the provision of news, an inspection of the licence agreements of commercial channels reveals ITV and Channel 4 are broadly similar, with Channel 5 holding the least. Sky, by contrast, has no public service broadcasting obligations.
In total, 2177 television news items were examined over the campaign period, of which 843-38.7 per cent of all news-were election related. We coded every item as either a policy issue or process story together with the story subject (health, economy, etc.). Using Krippendorf's alpha, we achieved a high level to credible agreement for each variable: election relevance was 0.93 with level of agreement 0.97, policy/process was 0.82, with level of agreement 0.92, whilst story subject was 0.74, with level of agreement 0.82.
Our content analysis of UK national press coverage was more focused than television news, as we identified policy stories published in different newspapers the day before they were broadcast on evening bulletins. It was beyond the scope of the study to examine every election item published in UK newspapers and match them with television coverage, since this would have involved reading tens of thousands of articles. 1 In order to interpret the editorial agenda-setting power of newspapers, we primarily focused on television news stories that could have been covered the previous day by the press. So, for example, since many process stories relate to that day's news (a campaign event), newspapers would not have covered them.
However, a key process-related theme to emerge in post-election debates about media coverage that we could not overlook was the attention paid to a possible Labour/ SNP coalition. The Conservative Party spent a good deal of campaign resources drawing attention to this potential electoral outcome. The Labour Party, however, tried desperately to avoid discussing this issue and complained to the head of BBC news about the prominence paid to a Labour/SNP post-election deal in its election coverage. Post-election debates also centred on the role of Conservative-supporting newspapers, with dramatic front-page headlines about the implications of a SNP-minority government days before election day (Cushion and Sambrook 2015).
The first part of our intermedia agenda-setting study thus explored the extent to which both newspapers and television news speculated on a Labour/SNP coalition, including isolating relevant press reporting the day before television news coverage. This involved re-examining all 505 process-related items in television news coverage. Our research team watched every item to quantify each meaningful reference to (or subtle inference about) a Labour/SNP coalition deal. We then used Nexis to search for terms such as "Labour", "SNP" and "coalition" to assess whether newspapers mentioned a Labour/SNP coalition. 2 As already acknowledged, simply because newspapers featured stories about a Labour/SNP NEWSPAPERS, IMPARTIALITY AND TELEVISION NEWS 167 coalition the day before a broadcaster does not mean broadcasters were influenced by-or even followed-the press. The second part of our intermedia agenda-setting study attempted to more closely connect newspaper stories with subsequent television news coverage. Since many process stories related to events happening on that day, it is unlikely they would have been covered by newspapers the day before. Moreover, process stories, such as speculating about a possible coalition deal, would be difficult to trace back to coverage in a specific newspaper. After all, at any point in the campaign a journalist may have considered it relevant to speculate about a Labour/SNP coalition. Since a key aim of the study was to assess whether specific newspapers set the broadcast news agenda, we therefore excluded all process stories and created a sub-sample of policy stories, such as a new housing announcement or health-care proposal. This subset amounted to 140 different policy stories and generated 321 news items between the five broadcasters. By story we refer to a specific issue, such as the launch of a party's manifesto, whereas news items refer to particular conventions of television news (an anchor or reporter package, say, or live two-way). This meant, on some occasions, different perspectives of the major evening stories were covered by several news items. Once again, we entered relevant search terms into Nexis which related to every policy item covered by broadcasters. Our research team spent a considerable amount of time establishing which newspapers covered the same policy items as television news and to help understand the nature of the major stories both media covered.
To move beyond a solely quantitative analysis of intermedia agenda-setting, we conducted semi-structured interviews with the heads of television news or senior editors from the BBC, ITV, Sky, Channel 4 and Channel 5, lasting between approximately 30 minutes and one hour. This included Paul Royale, editor of BBC News at Six and Ten, Katy Searle, BBC Westminster editor, Sue Inglish, the (now former) BBC head of political programming, Geoff Hill, editor of ITV News at Ten, Michael Jeremy, head of ITV news, Ben De Pear, head of Channel 4 news, Esme Wren, head of politics at Sky News and Christine Squires, head of Channel 5 news. 3 All interviews were conducted by the lead author of this article within approximately six months of the election finishing. Of course, simply asking interviewees whether they are influenced by newspapers generally or were during the election was never likely to elicit a clear admittance (although, somewhat surprisingly, some openly acknowledged the importance of monitoring the press' agenda). Instead, we referred to key moments in the campaign when high-profile newspaper stories appeared in coverage-such as a front-page Daily Telegraph splash about a letter from over 100 businesses supporting the Conservative Party-to interrogate editorial judgements and to probe their decision making. Although this provided some insight into the selection of particular stories, not all interviewees could recall in detail some of their editorial choices or how coverage was shaped differently (or not) from newspapers. When they did, we have included relevant responses in the findings.

Findings
Of the 505 election process-related items-from horse-race stories to more general coverage about campaigning-we found close to one-third mentioned the possibility of a Labour/SNP coalition deal (between 29 March and 6 May). Sometimes references to a coalition deal were explicitly put by journalists, such as after a television debate when a Sky News reporter said: "And then there was the extraordinary moment of a kind of Nicola Sturgeon love letter to Ed Miliband that was kind of unrequited, post-election negotiation occurring already". But the possibility of a Labour/SNP coalition deal was often more subtle but implicit in coverage such as "Today Nicola Sturgeon told Sky News that Ed Miliband will need the SNP's help if he wants to be Prime Minister" or when a politician said on BBC news, "You can just imagine the chaos, the absolute chaos, if Nicola Sturgeon's party holds the balance of power". The degree of speculation about a Labour/SNP deal during the campaign was different between broadcasters, with the proportion of items mentioning some kind of coalition deal on Channel 5 (24.2 per cent), BBC (27.7 per cent) and Channel 4 (29.7 per cent) less than ITV (39.4 per cent) and Sky News (40.4 per cent).
We also found that newspaper coverage during the campaign was informed to a large extent by the possibility of a Labour/SNP coalition deal. We identified 254 articles mentioning a potential coalition. Of course, the overall amount of daily articles produced by tabloid and broadcast newspapers is markedly different. But if we isolate the percentage of times one story featured a reference to a Labour/SNP coalition every day of the campaign (38 in total), broadsheet newspapers were likely to make them than tabloids: Times (60.5 per cent), Telegraph (57.9 per cent), Independent (52.6 per cent), Express (26.3 per cent), Guardian (68.4 per cent), Mail (50 per cent), Mirror (21.1 per cent) and Sun (28.9 per cent) (including Sunday equivalents). Although the Mirror did not regularly draw attention to a Labour/ SNP coalition, the Labour-supporting Guardian published the most articles which referenced a possible deal of all UK national newspapers. Of course, this did not take into account tone and prominence. But, in the days before the election, we noted several Conservative-supporting newspapers running alarmist front-page stories about the consequences of the SNP gaining power such as the Telegraph headline "Nightmare on Downing Street", juxtaposed with a picture of the SNP leader, Nicola Sturgeon.
However, interpreting whether the press-in particular Conservative-supporting newspapers-had an intermedia agenda-setting effect in their coverage of a Labour/SNP coalition is difficult to establish. After all, the issue was pervasive throughout the campaign, making it difficult to trace back to a particular newspaper story. Nevertheless, when we assessed whether any of the five broadcasters mentioned a Labour/SNP coalition the day after a newspaper had over the course of the campaign (29 March to 5 May) we found a high proportion of stories involving a reference to some kind of post-election deal: Times (57.9 per cent), Telegraph (50 per cent), Independent (36.8 per cent), Express (23.7 per cent), Guardian (57.9 per cent), Mail (42.1 per cent), Mirror (18.4 per cent) and Sun (23.7 per cent) (including Sunday equivalents).
When quizzed about the prominence granted to the SNP over the campaign or the wider fascination with a coalition deal, many of the interviewees were comfortable with the amount of coverage their news team supplied. It was seen as a legitimate story because the polling evidence suggested a Labour/SNP coalition was the most likely electoral outcome. Some interviewees indicated a potential Labour/SNP deal had significant news value because the Conservative Party regularly drew attention to this possibility, while the Labour Party struggled to rebut it. As Sky News's Esme Wren and the BBC's Katy Searle made clear: the SNP was a great story and I think what was interesting about it is that it was an issue for both the Conservatives and the Labour camps, not only because when Miliband kept on saying, "we won't work with you, we won't work with you", and that was obviously in view of a coalition but the way obviously the Tories made that one of their key themes … So, of NEWSPAPERS, IMPARTIALITY AND TELEVISION NEWS 169 course, yes, we did cover it. We did cover it in some quantity but I don't think we went over on it. (Esme Wren, Sky News) Labour didn't help themselves by the way they framed their own argument against the SNP and that they consistently didn't give a final answer to it, which just wouldn't shut it down. (Katy Searle, BBC) Again, it would be difficult to isolate any newspaper influence here, except to say that the news value of the coalition story was arguably legitimised or reinforced by the media's wider attention to it. The responses here also hint at wider influences at play, such as the Labour Party's campaign strategy, which we consider in our final discussion.
Our analysis of every election policy story in television news allowed us to trace more carefully whether broadcasters reported news published by newspapers the day before. Of the 140 different policy stories covered by television news during the 2015 General Election campaign, 44-31.4 per cent-appeared in newspapers prior to being aired. As a proportion of time spent by bulletins on policy coverage, 61.1 per cent of airtime was dedicated to covering stories previously published by newspapers. This reveals that UK television news pursued a similar policy agenda to newspapers during the election campaign.
In response to critics suggesting broadcasters were influenced by newspaper agendas during the election campaign, most interviewees were quick to reject this accusation. However, several interviewees accepted that newspapers played a role in shaping television news agendas during the election campaign or in political coverage more generally. Katy Searle gave the strongest acknowledgement that newspapers can, if a story is perceived to contain news value, have agenda-setting power: do I accept that we follow a newspaper agenda? Sometimes. I think it's true today as it would have been during the campaign. There is, as you'll know, the feeding off each other kind of mentality a little bit because if someone's going to get a story, if it stands up, you're going to look at it. But I hope that that is the same as the other way round … my job is to make sure that we lead the news as much as possible but of course do we pick up stories from the papers? Of course we do. But did we slavishly follow their agenda? No, absolutely not. (Katy Searle, BBC) Perhaps unsurprisingly, all interviewees rejected the suggestion they were influenced by particular newspapers-namely the right-wing press-and suggested many journalists read papers from all political sides. So, for example, the head of Channel Four, stated that: "I think broadcast journalists probably read the Guardian as much as they read the Telegraph so I don't. I think the more partisan the papers were, the less we necessarily followed them". And yet, when broken down by which newspaper UK national television news was most similar to, Table 1 reveals that a majority-62.4 per cent-of stories emanated from right-wing newspapers combined, in particular the Telegraph and Times (or Sunday equivalents).
Since the right-wing press outnumber left-leaning newspapers, on one level it is perhaps not surprising that broadcasters reported more of the stories they published. But it does confirm that the United Kingdom's evening bulletins pursued a policy agenda that was more similar to right-wing newspapers-rather than the left-wing press -during the 2015 General Election campaign.
Of course, television news does not have a collective agenda, with broadcasters pursing different angles and perspectives to election coverage. Indeed, this was reflected in the 321 policy items we isolated during the campaign: the BBC made up more than quarter of them-28.7 per cent-compared to Sky News (15.3 per cent) and Channel 5 (16.5 per cent). ITV and Channel 4, meanwhile, covered policy to a similar degree (making up 19.9 and 19.6 per cent, respectively). But while the BBC had a more policydriven agenda, compared to rival broadcasters its 10 pm news bulletin was-as Table 2 shows-proportionally the least likely to report policy stories addressed by the UK national press. The BBC, for example, included many citizen-led stories at the end of its bulletins which often involved policy-related stories not covered by newspapers. Just over half of the BBC's policy items were published by at least one newspaper prior to being broadcast.
In contrast, the least policy-driven broadcaster, Sky News, was the most similar to the press' agenda. Nearly two-thirds of Sky News' policy items-63.3 per cent-had previously

NEWSPAPERS, IMPARTIALITY AND TELEVISION NEWS 171
featured in at least one newspaper before being aired, with ITV not far behind (60.9 per cent). Of the commercial broadcasters, Channel 4 were the most distinctive from the policy agenda of newspapers (52.4 per cent). As already established, the Telegraph and Times were the newspapers most similar to the policy agenda of broadcasters. Table 3 further reveals that, with the exception of Channel 4, every other broadcaster ran the most policy-related stories previously reported by either the Telegraph or Times. But perhaps more striking was how broadly similar broadcasters were in the extent to which they reported policy stories from the same newspapers.
Nonetheless, many interviewees explained they were comfortable with covering stories pursued by newspapers if they had clear news value. So, for example, although Esme Wren insisted Sky News does not follow the editorial line of any newspaper, she did reveal that if a story was out in the public domain, which they are if they've been printed in a newspaper, and they're significant enough and we think that they're capturing an audience or the public, of course we feel that we would want to report on that. But then obviously adding our own journalism to it to say the Telegraph or the Sun have got this, here's our correspondent who's been speaking to sources or put it to the Prime Minister and they've come back with this response because we can't ignore a story if it's in the public domain.
This suggests that while broadcast editors were acutely aware of the partisan tactics of newspapers and the political goals behind their scoops, if a story was deemed "significant enough" they would pursue it from their own journalist perspective. In making sense of the policy agenda pursued by newspapers and television news during the election campaign, we need to look more closely at the type of stories originating from the press. After all, if particular newspapers-namely the Telegraph and Timesbroke major "scoops" during the campaign, it makes sense why broadcasters followed their agenda. Figure 1 illustrates the spread of policy-related news across the campaign and draws attention to the major stories first reported by newspapers and then broadcast by television news.
Over the course of the campaign, Figure 1 reveals that a steady stream of election items were reported first by newspapers and covered later by evening bulletins. Towards the latter part of the campaign the number of policy stories reduced because the campaign agenda focused more on the process of the elections. But Figure 1 also shows notable peaks and troughs when broadcasters and the press covered the same policy news.
Some of the peaks of Figure 1-when television and press were most in sync-can be analysed further by examining 12 major stories (interpreted by time granted to it) over the campaign, along with their prominence in the evening agenda and total airtime spent on it in that night's bulletin (see Table 4).
The policy story previously published in newspapers that generated most airtime on television news over the campaign was about how the two main parties would fund the National Health Service (NHS). From a newspaper perspective, it was a story based on a Conservative announcement-an additional £8 billion in funding-that would be made the following day. Of the 12 major stories, 9 were based on a policy announcements about subjects relating to the economy, taxation, migration and the European Union. These stories do not represent clear instances of intermedia agenda-setting, but warrant closer scrutiny than a content analysis allows. It could be-as several interviewees' claimed-that broadcasters supplied greater context and balance than newspaper coverage in these and other stories, but further close textual analysis is needed to support or challenge their perspectives.
Subsequent post-election analysis has suggested some right-wing newspapers coordinated with the Conservative Party on issues such as the right-to-buy policy and in letters from leading businesses (Moore and Ramsay 2015). This was evident in two major   For the Labour and Conservative NHS policy plans story (11 April), the coverage was over two days. ITV and Sky News covered the story in the evening broadcasts on 10 April (a result of covering the next day's newspaper headlines) whereas other broadcasters covered this story on 11 April. stories that were published by right-wing newspapers and had more than a hint of Conservative bias. All evening bulletins reported towards the top of their agendas a front-page Daily Telegraph letter from over 100 business leaders supporting the Conservative Party and a Times interview with a Conservative defence secretary, Michael Fallon, who personally attacked the Labour leader and his defence policy. In both cases broadcasters did not replicate the partisan way the story was carried by much of the press. ITV news, in fact, drew attention to the partisanship of the letter towards the end of a report in a piece to camera: When David Cameron and George Osbourne then turned up at Marsden's brewery in Wolverhampton, whose chief executive had put his name to the letter, the extent of the coordination become Interviewees were asked about why they prominently covered the letter from business leaders, since it originated from an openly partisan source (e.g. Telegraph) and clearly favoured the Conservative Party. In each case, they defended the validity of the story because of the wider context they provided for viewers. So, for example, Sue Inglish argued: it was a perfectly legitimate story to follow and I think we reported it perfectly fairly. We were also very clear about where it came from and we were completely clear about the fact that there'd been Conservative Party involvement in it. (BBC head of political programming) Similarly, Geoff Hill, said "I remember in the Telegraph it was quite a decent story because, it was potentially clear where it came from, but there were a number of quite significant businesses behind there and we looked at that". Christine Squires agreed: "business leaders were writing letters … that is part of what's going on, that's part of the campaign and, therefore, I feel no worry about reporting that at all".
Another Daily Telegraph front-page letter was published on 27 April from 5000 small businesses supporting the Conservative Party, but this was more sceptically received by broadcasters. Esme Wren, for instance, said the Sky team were speaking to those businesses, who [were] asked … to sign this letter and obviously traced that back to Central Office and made that the story to say, of course we know how these letters come about. Central Office calls 20 businesses and says can you sign this letter? So the letter writing is quite an old tactic and we find that quite dated and exposed it for being so.
Regardless of the party political collusion, for Paul Royale the business leaders' letter had strong news value: The business letter, I know there was a bit of argument because it was driven from Conservative Central Office but then if 5000 small and medium-sized business leaders still signed it, as long as you're clear about that, as long as you say the origin of this thing, it's still got 5000 or however many it was to sign this thing. So I think there was a version of the story where it may have appeared that they all spontaneously got this thing together … they still managed to get these people. (Paul Royale, editor of the 6 pm and 10 pm BBC bulletins) When asked about why so much airtime and prominence was granted to an interview Michael Fallon gave to the Times which attacked the Labour leader so personally, again interviewees defended its news value but from different perspectives. According to Geoff Hill: the Fallon comments, I mean, it's difficult to ignore that kind of story when you've got a really high-profile member of party A making a really scathing personal attack about the leader of the opposition party and it's hard to ignore that. (Geoff Hill, editor of ITV 10 pm bulletin) Sky News justified the story by following up on it, rather than just accepting what was written in the Times. The BBC, by contrast, ran an item about the attack, but also-according to Paul Royale-focused on the implications for defence in light of the minister's attack on Labour.

Intermedia Agenda-setting Effects: Why Newspapers Can Distort the News Values of Broadcasters and Compromise Impartial Judgements
Our content analysis delivered some clear-cut conclusions about how closely in sync the agenda of UK television news was compared to national newspapers during the 2015 General Election campaign. First, almost a third of television news items appeared in newspapers prior to being broadcast by television news. This proportion more than doubled when interpreted by the amount of airtime spent by bulletins covering the same policy stories as newspapers. Second, a clear majority of policy stories reported by broadcasters emanated from right-wing newspapers. Third, broadcasters with the most public servicedriven responsibilities-notably the BBC-were proportionately the least likely to cover the same policy stories as newspapers. Fourth, with the exception of Channel 4, every broadcaster ran the most policy-related stories reported by the Telegraph or Times (or Sunday equivalents). Fifth, while most days of the campaign broadcasters reported stories covered by newspapers, there were clear moments during the campaign when television news coverage followed stories originating in the press. Overall, our finding reinforce many previous intermedia agenda-setting studies that demonstrated newspapers, in particular quality or broadsheet titles, act as opinion-leaders (Vliegenthart and Walgrave 2008). Similarly, quality public service media-notably the BBC-were proportionally the least likely to follow the press' agenda.
Our study about the intermedia agenda-setting role of newspapers in coverage of a possible Labour/SNP coalition was less clear cut. While we found television news coverage was regularly informed by speculation about a possible coalition deal the day after many newspapers also covered this issue, it was hard to trace empirically clear instances where newspapers influenced broadcast coverage. The possibility of a Labour/SNP coalition was widely reported throughout the campaign-particularly on Sky News and ITV-and perceived as a legitimate news story by journalists. But a more systemic intermedia agendasetting effect might have occurred in a more indirect way. Since newspapers were regularly speculating about a Labour/SNP coalition during the campaign, it could have legitimised the news values of broadcasters to also cover the possibility of a post-election deal between these parties.
Nevertheless, in conventional intermedia agenda-setting studies our study of policyrelated coverage could be interpreted as evidence that television news followed the UK NEWSPAPERS, IMPARTIALITY AND TELEVISION NEWS 177 national press agenda during the campaign. After all, put in the language of traditional agenda-setting parlance (McCombs 2004), newspapers appeared to transfer the salience of particular issues-notably about the economy-on to the broadcast media. While our content analysis study cannot assume causality between press influence and television news coverage, the close alignment of agendas between media over the campaign and the acknowledgement by broadcast editors that they would follow up on newspaper stories if they had sufficient news value does suggest an intermedia agenda-setting was apparent during the 2015 UK General Election. But our aim was to qualify further the nature of any transfer of agendas between the press and television news, and to interpret it in more qualitative detail than often large N studies can reasonably supply.
Based on our closer assessment of the major stories reported by the press and television, and interviews with the heads of television news broadcasters or senior editors, our findings do not support the view that UK television news bulletins submissively followed the editorial agenda of newspapers or fell under the spell of right-wing newspapers during the election campaign. But since television news bulletins have to remain impartial in coverage of the election campaign, it perhaps makes them more susceptible to being influenced by the press' agenda rather than the other way round because broadcast editors might be reluctant to lead on politically sensitive stories. And yet, despite repackaging many partisan led-stories and more impartially covering the policy positions of parties than the press, our study empirically demonstrated that television news pursued a similar agenda to UK newspapers during the election campaign and followed their lead on some of the major stories. We do not consider that this evidence amounts to any deliberate political bias amongst broadcasters or fuel that can fan the flames of left-or right-wing conspiracy theorists. In our view, it is structural constraints and professional routines that encourage broadcasters to feed off stories that are more likely to be supplied by right-leaning newspapers (because they outnumber left-leaning competitors), and pursue issues and concerns more likely to favour a Conservative Party agenda.
The structural constraints of broadcasters, of course, are obvious when compared to the press. Newspapers typically lead the morning agenda, having been published late into the night or updated early in the morning, and they have more physical space to cover stories than many competitors such as a television news bulletin. Given, as the BBC's Sue Inglish pointed out, the ever-quickening pace of the news cycle, it might seem unlikely today that evening bulletins would be influenced by stories reported either the night before or much earlier that morning. But our study revealed that the salience of many election stories reported by newspapers remained strong throughout the day and were reported either on the early or late evening bulletins. Put another way, despite the pace and fluidity of today's news cycle, and the highly fragmented and hybrid media systems influencing and reshaping stories (Chadwick 2013), we found the morning newspapers continued to share a similar agenda to the evening television news bulletins during the 2015 General Election campaign.
In many cases, the election stories reported first by newspapers and then covered by broadcasters were policy announcements from parties. In this sense, the symmetry between newspapers and television news might appear a relatively innocent act or represent a benign intermedia agenda-setting effect. But our study found moments when broadcasters explicitly followed stories originating in newspapers that were more advantageous to-or even published in collusion with-a particular political party (Moore and Ramsay 2015). We should not overlook other factors, of course, that shaped the selection of news stories. While the Conservative Party ran a well-funded and highly sophisticated election strategy, the Labour Party's campaign has been criticised for not effectively communicating its message or staging events deemed newsworthy (Scammell 2015). Unlike the Conservative Party, Labour struggled to appeal to broadcasters' news values-a point raised by several of our interviewees.
At the same time, the reliance on news values does represent a departure from a public service duty of reflecting party messages (Semetko et al. 1991) to selecting them based on professional judgements. It is this, above all, that helps explain the intermedia agenda-setting role of the press during the election. For many interviewees were comfortable with reporting stories originated in newspapers if they had perceived news value. To paraphrase Esme Wren, if news is in the public domain-from newspapers or elsewherewhy shouldn't broadcasters follow up on it? While this might appear a reasonable professional judgement, news values are far from politically neutral. We found, for example, a clear majority of stories broadcast on television news emanating from right-wing newspapers, rather than being balanced out by left-leaning titles.
If broadcasters increasingly rely on the news values of partisan newspapers to inform their impartial judgements, it could lead to ideologically narrowing the agenda of issues citizens encounter in media coverage. So, for example, we identified two election stories reported in more centre to left-wing newspapers-a Guardian letter written by more than 140 health professionals attacking the government's record on the NHS and a poll of leading economists questioning the government's austerity measures reported in The Independent-that did not receive the same amount of airtime or prominence as similar stories published by right-wing newspapers. When discussed with some of the interviewees, the response was that both stories did not carry sufficient news value. What remains an open question is whether their news judgement would have been different if either story had been more substantially covered by the wider (right-wing) press. This also points to the limitations of our study. Like studying any effect, interpreting intermedia agenda-setting is not empirically easy to disentangle. Whether policy announcements, speculation about coalition deals or letters written to newspapers, intermedia agenda-setting studies have to consider carefully the multiplicity of processes and consequences which shape news agendas. Nevertheless, our study identified clear moments during the campaign when the press set the television news agenda and revealed editors were more than comfortable relying on press stories if they were deemed newsworthy. Since the UK press reflect a more right-wing view of the world, we concluded that relying on their news values to inform editorial judgements risks undermining the impartiality of broadcasters. The effect of our intermedia agenda-setting study, in other words, has clear ideological implications.
Our wider contribution to intermedia agenda-setting debates aimed to move beyond relying solely on a large-scale content analysis by considering some of the major stories when different media agendas converged and by interviewing some of the most senior editorial figures in UK broadcasting about the editorial power of the press. Future studies should consider combining quantitative and qualitative analysis to understand better the editorial processes behind the selection of news and to more carefully interpret intermedia agenda-setting.

DISCLOSURE STATEMENT
No potential conflict of interest was reported by the authors.