What is News?

The deceptively simple question “What is news?” remains pertinent even as we ponder the future of journalism in the digital age. This article examines news values within mainstream journalism and considers the extent to which news values may be changing since earlier landmark studies were undertaken. Its starting point is Harcup and O'Neill's widely cited 2001 updating of Galtung and Ruge's influential 1965 taxonomy of news values. Just as that study put Galtung and Ruge's criteria to the test with an empirical content analysis of published news, this new study explores the extent to which Harcup and O'Neill's revised list of news values remains relevant given the challenges (and opportunities) faced by journalism today, including the emergence of social media. A review of recent literature contextualises the findings of a fresh content analysis of news values within a range of UK media 15 years on from the last study. The article concludes by suggesting a revised and updated set of contemporary news values, whilst acknowledging that no taxonomy can ever explain everything.


Introduction
Asked how they define news, journalists sometimes reply: "I know it when I see it." Pressed on why something has been deemed newsworthy, a typical response is: "Because it just is!" (Brighton and Foy 2007, 147). Definitions relying on such "gut feeling" (Schultz 2007) arguably obscure as much as they reveal about news selection, prompting academics to offer their own explanations, which can involve devising taxonomies of news values. One of the most influential and frequently cited of these is a paper by Johan Galtung and Mari Ruge (1965), "The Structure of Foreign News", published more than 50 years ago.
Identifying and recording the news values found within published pieces of journalism cannot provide a complete explanation of the journalistic process but that does not mean such study is without value. News values are worth studying because they inform the mediated world that is presented to news audiences, providing a shared shorthand operational understanding of what working journalists are required to produce to deadlines. It is the way news values work in practice that results in them being articulated and conveyed to new journalism trainees and journalism students, and they are also used by public relations professionals and others aiming to obtain maximum news coverage of events (or pseudo-events).
It was to test the continuing relevance of Galtung and Ruge's landmark piece of scholarship that we conducted our own study of the news values that can be identified within published outputs, "What is news? Galtung and Ruge revisited" (Harcup and O'Neill 2001). As with its more illustrious forerunner, this struck something of a chord, becoming one of the most widely read and cited articles in the history of the journal Journalism Studies. Our study concluded, "although there are exceptions to every rule, we have found that news stories must generally satisfy one or more of the following requirements" if they are to be selected:

1.
The power elite: Stories concerning powerful individuals, organisations or institutions.

2.
Celebrity: Stories concerning people who are already famous.

3.
Entertainment: Stories concerning sex, showbusiness, human interest, animals, an unfolding drama, or offering opportunities for humorous treatment, entertaining photographs or witty headlines.

4.
Surprise: Stories that have an element of surprise and/or contrast.

5.
Bad news: Stories with particularly negative overtones, such as conflict or tragedy. 6.
Good news: Stories with particularly positive overtones, such as rescues and cures. 7.
Magnitude: Stories that are perceived as sufficiently significant either in the numbers of people involved or in potential impact.

8.
Relevance: Stories about issues, groups and nations perceived to be relevant to the audience. 9.
Follow-up: Stories about subjects already in the news. 10.
This was offered, not as the last word on news values, merely as a contribution towards "rendering news selection a more transparent and better-understood process" (Harcup and O'Neill 2001, 279). As was stressed in the above study, any exploration of news values can only provide a partial explanation of what lies behind journalistic news decisions. Examining news outputs may explain as much about news treatment as news selection (Staab 1990;Harcup and O'Neill 2001) and Staab (1990) argued for a functional model that takes into account the intentions of journalists. For Donsbach (2004) news values necessarily involved subjective judgements and can never be truly objective criteria, while Hall (1973) suggested that news values themselves are part of an ideologically constructed way of perceiving the world that favours and "naturalises" the perspectives of powerful elites, a view echoed by Herman andChomsky (1988) in their propaganda model andMcChesney (2000). Although this ideological role of news selection (and treatments) is not the subject of either our 2001 study or the new one, the identification of prevalent news values can be used to inform future research into potential ideological impacts. Therefore, despite limitations, we argue that it remains valuable to unpick the criteria involved in the selection of news since this "is one of the most important areas of journalism studies[. It] goes to the heart of what is included, what is excluded, and why" (O'Neill and Harcup 2009, 162). Further, the journalistic selection process has been described as "probably as important or perhaps sometimes more important than what 'really happens'", when it comes to determining whether or not something becomes news (Westerhahl and Johansson 1994, 71).
Thus, our list of news values was offered simply as one tool to aid analysis, discussion and, we hoped, understanding of this significant part of the journalistic process. In this fresh WHAT IS NEWS? 1471 study we also set out to examine the extent to which any taxonomy of news values devised in the age before Twitter, Facebook and other interactive platforms, can be taken as read today. That is one of the key questions to be addressed in our revisiting news values and attempting to update our list of selection criteria.
News Values: A Literature Review Brighton and Foy (2007, 194) draw attention to the difficulties faced by journalists when asked to explain news: "Reiterating what has been said to both authors on countless occasions during the research process of this book, the response to the question 'why is this news?' may well remain: 'It just is!'" However, this does not mean that journalists are completely unable to articulate why they selected one story over another: Indeed, they [new values] pepper the daily exchanges between journalists in collaborative production procedures … [T]hey are the terse shorthand references to shared understandings about the nature and purpose of news which can be used to ease the rapid and difficult manufacture of bulletins and news programmes. (Golding and Elliott 1979, 114) For their part, scholars have taken different approaches to the theorisation and study of news values, sometimes focusing on the apparent newsworthiness of an event or news actors to uncover why a story has been selected, but also considering the organisational, cultural and economic factors that may also influence news selection. Galtung and Ruge's (1965) work was based on hypothesising selection criteria and formed part of a wider study on how foreign events were reported in the Norwegian press. Some scholars use their own experience and/or analysis of news outputs to conceptualise news values (Golding and Elliott 1979;Harcup and O'Neill 2001). Others take an overview of previous scholarship to produce a summary of key news values (Harrison 2006;O'Neill and Harcup 2009;Caple and Bednarek 2013;Harcup 2015), while others use ethnographic observations and interviews (Schultz 2007;Dick 2014). Such studies are aimed at helping us understand the judgements that are made when journalists select news, described as "cognitive" news values by Caple and Bednarek (2015, 3).
Approaching the issue of news values from observations of Danish TV journalists, Schultz (2007) argues that six news values dominate: timeliness, relevance, identification, conflict, sensation and exclusivity. Being first with a story-exclusivity-adds value for producers who must attract audiences, and can override other news values, such as timeliness. For Schultz, a distinction can be drawn between three types of news value: undisputed, taken-for-granted and rarely articulated news values, which she calls doxic news values; and explicit news values: either dominant, undisputed orthodox values, or debatable, dominated (heterodox) news values (195)(196). O'Neill (2012) suggests a hierarchy of news values may exist, with celebrity dominating, not just in the popular press but also in "quality" UK newspapers.
No theory of news values can explain everything, not least because arbitrary factors including luck, convenience and serendipity can come into play; as when a planned story falls through at the last minute, for example, and a previously discarded one is selected to take its place. News selection is also subject to the influence of journalistic routines such as issues of access and meeting deadlines (Schultz 2007;Phillips 2015); competition for exclusives (Allern 2002;Schultz 2007;Phillips 2015); the influence of proprietors and advertisers (Caple and Bednarek 2015); external influences including the role of public relations professionals and "spin doctors" (Brighton and Foy 2007); the belief systems of journalists as a result of their social environment (Donsbach 2004;Phillips 2015); and the influence of peers within the workplace or "habitus" (Bourdieu and Wacquaint 1992, 133). Shoemaker and Cohen (2006) found that the same newsworthiness model can be applied to news from many countries, while Lavie and Lehman-Wilzig (2003) found little difference between genders in the factors determining journalists' selection of news.
News values can be seen less as a reflection of what type of information citizens want or need, and more as a reflection of organisational, sociological and cultural norms combined with economic factors (Weaver et al. 2007). This leads Allern (2002) to propose a supplementary set of commercial news values, whereby sensationalist stories are most likely to be pursued; stories that are costly to pursue are less likely to make it into the news; and news subsidies, such as well-prepared press releases and photo opportunities, are more likely to be taken up by resource-starved and hard-pressed journalists, and translated into news items. This news subsidy by the public relations industry, described as pre-fabrication by Bell (1991), has been observed in "quality" media as well as less well-resourced news organisations (Lewis, Williams, and Franklin 2008). Commercial pressures can also result in the selection of news stories on the basis of their perceived appeal to target audiences rather than by any inherent qualities of newsworthiness or importance (Niblock and Machin 2007, 191;Stromback, Karlsson, and Hopmann 2012, 726). This is particularly pertinent for online news. For example, an analysis of news in the five largest newspapers of the Netherlands established that audience clicks affected news selection (Welbers et al. 2015), a conclusion supported in a study of over 300 editorial gatekeepers in the United States (Tien Vu 2014).
Furthermore, who is applying news values can be as important as what news values are being applied. Using Bourdieu's sociological approach, Schultz (2007, 197) argues that the degree of autonomy afforded to a journalist making choices about news will depend on the type of news organisation they work for, the type of journalism they produce, and the level at which they operate. For Bourdieu, journalists can position themselves only within the confines of the social or workplace space that surrounds them (Bourdieu and Wacquaint 1992), and, as such, they hold different levels of editorial capital in the journalistic field (Schultz 2007, 194). Put simply, some journalists are in a more powerful position than others and their decisions about what makes a story newsworthy will win out over other journalists' decisions. In addition, the newsworthiness of some types of events or issues is open to discussion or interpretation. For example, "hard news" tends to comprise generally undisputed orthodox news values, while the news values associated with "soft news" are less dominant and may be more open to contestation (196).
In their critique of Galtung and Ruge's work on news values, Brighton and Foy (2007) argue that, while "there is still a place for a matrix which sets out the variables" (3), both the times and the media have changed since the 1960s. Their consideration of the impact of broadcast journalism, rolling news and digital media since the original Galtung and Ruge study led them to devise their own set of factors to be taken into consideration when selecting (and analysing) news stories in the twenty-first century: Relevance; Topicality; Composition; Expectation; Unusualness; Worth; and External influences. They add that news values will vary from medium to medium (29).
Discussing the Harcup and O'Neill (2001) study, Phillips (2015, 18) argues that conflict should be part of any taxonomy of news values, rather than being subsumed within the WHAT IS NEWS? 1473 category of bad news. The Harcup and O'Neill (2001) study acknowledged that content analysis of news outputs fails to make a clear distinction between news selection and news treatment, and this is explored further by Caple and Bednarek (2015, 11) in research that offers a complementary discursive approach to how news values are constructed through various semiotic resources and practices: "With the digitisation of the production and dissemination of news, news discourse has become a visually enriched product. The lack of research concerning news values and imagery has been noted by only a handful of researchers." They highlight how photography and design emphasise news values such as superlativeness, negativity, personalisation and aesthetic appeal (8)(9)(10)(11)). Dick's recent study of news values that inform interactive infographics found that, as one interviewee put it, "some visualisations of information can work to challenge conventional news values" (Dick 2014, 499).

"What Works Best on Facebook?"
In the digital and social media environment, Phillips (2015, 6) notes that while hard news may be shared on Twitter, it is Facebook that drives more traffic (Phillips 2012). This importance of Facebook over Twitter when it comes to driving traffic has been found in numerous studies, including those by Anderson and Caumont (2014), Newman (2011) and Olmstead, Mitchell, and Rosenstiel (2011). The hugely successful Mail Online "creates loyalty by giving readers what it already knows that they want", found Angela Phillips (2012, 675), and they know it by observing which stories are the most likely to be shared by Facebook. "Most of the stories that attract large numbers of hits are also the jokey stories that people pass on", adds Phillips.
Online news therefore presents journalists with instant feedback about whether their news selection decisions (and methods of presentation) marry with those of significant numbers of readers. A study of news factors operating in the German media by Wendelin, Engelmann, and Neubarth (2015, 12-15) found internet and social media audiences to be less keen than journalists on politics and other stories of "social significance". In a study of local online news, Schaudt and Carpenter (2009) note that research about readers' preference for soft or hard news, at least in the United States, is contradictory. Examining the types of articles favoured by readers, they found that readers chose to view soft news about 63 per cent of the time and, unsurprisingly for a local site, the most popular news value was proximity (76 per cent of stories) followed by conflict (31 per cent of stories). The rise of digital and online media also demonstrate the role that technology and audiences play in decisions about what makes the news, with user-generated content becoming more important in news production (Domingo et al. 2008). However, Jonsson and Ornebring (2011) note that participatory journalism is more prominent in online tabloid news, and content has relatively little to do with hard news, there being a greater emphasis on popular culture-orientated content or the personal, building brand loyalty. This knowledge influences both selection (Tien Vu 2014; Welbers et al. 2015) and presentation (Thurman and Myllylahti 2009).
While it must be acknowledged that user-generated content has affected the news we receive (Domingo et al. 2008), somewhat utopian predictions about open-Web democracy revolutionising news production (Beckett 2008;Gillmor 2004;Rusbridger 2012) have not come to pass. Instead of radically changing the production process, the main role performed by the bulk of the online audience seems to be sharing and disseminating news on social media platforms (Phillips 2012, 675). As Emily Bell (2015) puts it: "The key question for news organisations, tied to the goal of big traffic, is now 'what works best on Facebook?'" As is evident from this brief literature review, much has been written on news values since 2001. However, largely absent from the literature are empirical attempts to subject taxonomies of news values to the type of study that Harcup and O'Neill set out to conduct on the original Galtung and Ruge (1965) criteria. That is what we aim to do with the study discussed in this article.

Methodology
Analysis of published outputs cannot tell us everything about journalism, of course, but it can tell us something. Most obviously, it can tell us what has been selected for publication as news. This can be useful because it can help take our knowledge of what is news beyond the anecdotal and the realm of what we may think we know into more systematic analysis of what has actually been selected. This in turn allows for comparisons (between different news organisations or platforms, for example, or over time) and can complement other forms of research such as ethnographic studies, in-depth interviewing or critical discourse analysis. Study of published outputs can also be used in an attempt to apply-and thereby test-scholarly explanations of news values such as that put forward by Galtung and Ruge. Results of such a study can in turn be used to as a basis for further discussion and exploration about what does (and does not) become news.
Our 2001 study, which was conducted in the hope of doing just that, was based on content analysis of page-lead news stories published in the United Kingdom's three market-leading daily national newspapers (The Sun, Daily Mail and Daily Telegraph) during a sample month in 1999. Despite predictions of its imminent demise, the printed press remains worthy of study in part because of the continuing reach of its journalism: the paid-for titles selected for this study recorded a combined average daily sale of more than six million copies during the sample month (Ponsford 2014), and more than two million copies of the two free titles are distributed every day (Feeney and Beattie 2013;Press Gazette 2014). Circulation on such a scale remains socially significant. In addition, Phillips (2012) points out that online it is still major news organisations that obtain the largest proportion of audience share: while the BBC dominates, it is newspapers, notably the Daily Mail, The Guardian, The Telegraph and The Sun, which also feature in charts of the top news originators. Furthermore, she demonstrates that that "Twitter did not make 'dead tree editions' obsolete and it certainly did not render journalists obsolete" (670). Newspapers also continue to break major news stories and retain the capacity to influence other media, as highlighted by research suggesting that broadcasters took cues from newspapers' agendasetting during coverage of the 2015 UK general election campaign (Cushion and Sambrook 2015). Crucially, the news values of newspapers remain worth studying because what appears in such publications is by definition the result of journalistic selection in a highly competitive market.
For the new study, rather than simply conduct an exact replica of our earlier one 15 years on, we have expanded it to encompass a wider range of titles: The Sun, Daily Mail, Daily Telegraph, Daily Mirror, Daily Express, The Times, The Guardian, The Independent, plus free-distribution dailies Metro and London's Evening Standard. A sample week from November 2014 has been selected for each of the 10 newspapers, with titles spread over four weeks to cover the entire month and avoid results being skewed by one story dominating WHAT IS NEWS? 1475 coverage for days. As with the earlier content analysis, this study is of lead stories published on news pages, excluding sport, finance, features and comment pages. This time only the front and right-hand-side news pages have been considered, because such pages are regarded within the industry as being the prime location for news. Exceptions were allowed when right-hand pages were given over to full-page advertisements (most frequently in the Telegraph), when the left-hand page lead may be chosen for analysis.
This resulted in a data-set of 711 stories, each selected by the newspaper as a news page lead. Each story was then analysed in an effort to identify which, if any, of the Harcup and O'Neill (2001) news values could be detected in the new sample; how news values interact with each other; and whether any amendments, additions or deletions to the list of news values might be considered appropriate. Our methodology was deceptively simple if time-consuming: we read each of the 711 stories and used our 2001 set of news values as a coding sheet, identifying which if any of the various factors we could identify in each story. Deciding which news values were present in any particular story involved a close reading of the text, consideration of the content and an evaluation of the context (was it following up a story that had originally been published elsewhere, for example?); the fact that both of us have worked as journalists in addition to studying and teaching journalism perhaps helped in this process. To ensure consistency each story was read and coded by both researchers, with disparities in coding discussed before arriving at consensus. Clearly there is subjectivity involved in such a process-just as there is subjectivity involved in the journalistic process itself-and we would never claim anything different. But we nevertheless believe the sample of stories is large enough to allow for some tentative conclusions to be drawn.
As well as expanding the range of newspapers studied, a further addition has been to examine a phenomenon little talked about at the time of our original research, let alone when Galtung and Ruge first published: the role of the audience. In the digital age audiences are also selecting and disseminating stories and this in turn is affecting journalists' own news selection decisions (Tien Vu 2014; Welbers et al. 2015). Stories are increasingly being disseminated on social media platforms through audience recommendations and "shares" (Olmstead, Mitchell, and Rosenstiel 2011;Hermida et al. 2012;Phillips 2012) in what Phillips terms "sociability". To explore this area alongside our content analysis of newspapers we also examine which, if any, news values can be identified in the case of the stories from UK news providers that were shared most frequently on Facebook and Twitter. Consideration of the most popular news items shared on social media during 2014, the same year in which data were collected for newspaper output, will inform discussion of the ways in which phenomena such as "sociability", or what we call "shareability", may now be taken into consideration by journalists when selecting (or preparing) material for publication.

Findings: Newspapers
In considering 711 page lead stories across 10 newspaper titles (see Tables 1 and 2), our findings show that bad news is the big winner; but good news is still significant, and some stories fall into both categories. One example of a story containing both good and bad news appeared in the Telegraph on 20 November 2014: "Britain Bars US 'Pick-up artist' Who Gives £1000 Lessons in Misogyny". The man is in the news because of his sexist attitudes (bad news), but this article is about banning him from entering the country (welcomed as good news by many people opposed to sexism). Other issues/events that could be interpreted as either good or bad news dependent on the reader's viewpoint are written and presented emphatically as either positive or negative (seemingly in line with the newspaper's own agenda), such as: "EU Must Change or We Quit", the frontpage lead in the Telegraph on 20 November 2014. While all the newspapers favoured bad news, certain titles leaned most heavily towards it, notably the Times (the most bad news, at 71 stories), Mail, Mirror and Guardian, all with around six times as much bad news as good. This finding that a story might simultaneously be good news and bad news illustrates that any taxonomy of news values ought not to be thought of as a list of alternatives. It is clear that one story might tick several boxes, and how certain news values operate in different combinations might usefully be an issue for future research.
Perhaps unsurprisingly, surprise is a significant factor, ranking second to bad news overall. Entertainment was the third biggest news value recorded, but was most prominent in the Mail, followed by Metro, Sun, Express and Times. The Sun had the biggest total of stories recorded as meeting the celebrity criterion. Entertainment and celebrity were both more significant in the popular red-top and mid-market titles, although they were by no means insignificant at the quality end of the market, as has been noted previously (O'Neill 2012). The power elite seemed to be of more significance as a news value in the broadsheet/quality press, possibly because these papers report "serious" news more prominently. The newspaper's own agenda appeared to come into play more in the mid-market titles than elsewhere, with the Mail and Express between them accounting for more instances (28) than the other eight titles combined (23). Of course, there may also be more subtle agendas at work that might require ethnographic study and interviews with journalists to discern.
Overall, we found many similarities between titles in the snapshot sample as well as some differences, with freesheets mostly mirroring a mixture of the results from the midmarkets and red-tops. However, in revisiting our 2001 set of news values, we found that some categories were too broadly defined when subjected to empirical testing; some categories need to be unpicked and some new categories seem to warrant inclusion, and these are put forward in the Conclusion. Entertainment encompasses both a dark and light side, which might be better conveyed in two distinct categories. Some of the darker but "entertaining" human-interest stories might perhaps be more accurately labelled as drama. News stories about sport could be added to the entertainment mix alongside show business, while the surprise category ought to contain "unusualness" within its definition. We also noticed that there were many stories involving large corporations and these might be included within the power elite category.
There is a case for teasing out the broad definition of bad news to add conflict as a separate category, as has been suggested by Phillips (2015). We also found ourselves agreeing with Caple and Bednarek (2015) that strong visuals can be a reason for selecting a story, and that in semiotic terms, a visual can be seen as conveying and emphasising a number of news factors, such as bad news, drama or surprise. The importance of visuals and, to a lesser extent, audio became particularly noticeable when considering stories shared on social media (see below).
On investigation it became apparent that the term "magnitude" can mean more than merely involving large numbers; it ought to include extreme behaviours or occurrences (a man dying in extreme temperatures in an oven, for example, or a particularly unpleasant crime). Finally, there is a strong case to be made for exclusivity to be considered a news value in its own right, given the frequency with which we found newspapers drawing attention to the fact, as in: "a poll by the Times indicates … " (24 November 2014) or "It follows an investigation by the Independent … " (29 November 2014).

Findings: Social Media
Any study of contemporary news values must now also consider the impact of social media on the traditional division between the roles of journalists as active (producers, selectors, gatekeepers) and of audiences as passive (receivers, consumers). To this end, having considered news selected by journalists, we then examined those stories that were most frequently shared on social media by UK news audiences in the same year. Although the two data-sets have different specifications, they nonetheless allow for some preliminary comparisons of notions of newsworthiness as decided by journalists and audiences, respectively.
Facebook was the most common way of sharing, according to research by journalism. co.uk, which produced a chart of the "top 10" most popular news stories from mainstream media outlets, as measured by Facebook shares, likes and comments (see Appendix A). These ranged from a Mail Online story featuring pictures of icicles, whose 586,250 Facebook interactions made it the most popular, to a Mirror article about giant spiders, in 10th place with 315,613 Facebook shares, likes and comments (Albeanu 2015a).
The "top 10" stories from the same mainstream news outlets as measured by shares via Twitter suggest that the most shared news items on Facebook were shared many more times than the most shared news stories on Twitter. The most tweeted story was a Mail Online piece about a YouTube video teaching dog-owners how to cook treats for their pets. Also on the list was a Telegraph infographic about Palestinian children killed by Israeli forces (Albeanu 2015b). Both items also featured in a different list of the most tweeted stories of 2014 originating only from national newspaper websites (Searchmetrics WHAT IS NEWS? 1479. By combining these two Twitter lists and removing any repetition it is possible to identify the "top 15" most tweeted items from news sources in 2014, from the Mail's "paw-fect recipes" at 76,752 tweets to the Sun's photospread of a scantily clad "supermodel", which was tweeted 17,852 times (see Appendix B).
Although a crude measure, we took the top 10 Facebook stories and the top 15 most tweeted stories as being a snapshot of which news stories most interested those members of the audience using social media in 2014. To that end, we have attempted to identify which of the Harcup and O'Neill (2001) news values can be said to apply to each of these 25 stories, and to rank factors in order of frequency (see Table 3).
Between the newspaper and social media data-sets, there were some similarities as well as differences; bad news is almost as popular a factor online as in print. However, the most popular stories online are rarely concerned with the power elite. There may be little doubt that digital media can help challenge mainstream news agendas, but the most popular stories do not reflect this democratic ideal. Rather, the most common news value is entertainment; such stories seem to be shared by online readers because they are fun, and sharing them can brighten the day. This suggests a possible new news value as highlighted by previous studies (Olmstead, Mitchell, and Rosenstiel 2011;Hermida et al. 2012;Phillips 2012;Tien Vu 2014;Welbers et al. 2015): shareability. Precisely what qualities give one story more shareability than another are hard to pin down, but Janine Gibson (a former Guardian website editor who in 2015 became editor-in-chief of BuzzFeed UK) has noted that the most shared stories tend to be "stuff that makes you laugh and stuff that makes you angry" (quoted in Newman 2011, 24). The quest for shareable stories is perhaps likely to become an increasingly important consideration within newsrooms (Bell 2015). As is noted in the Reuters Institute 2015 Digital News Report, based on a survey of more than 20,000 people in 12 countries:

TABLE 3
The totals, and ranking, of Harcup and O'Neill's (2001)  The news that is most read, shared, and discussed in social media is produced by professional news organisations … Twitter tends to work better for serious news brands like the BBC and New York Times. Facebook tends to favour more accessible content and brands that focus more on entertainment and lifestyle content such as PlayBuzz, the Huffington Post, and BuzzFeed. (Newman 2015, 81) So, if news selection-and presentation-becomes dominated by pre-emptive thoughts about what works best on Facebook, does that inevitably mean that the news agenda will become increasingly dominated by "soft" entertaining items? The evidence of our admittedly limited study is mixed, suggesting that the public's news preferences may be more complicated than that. Celebrity does not dominate social media shares as much as might perhaps have been predicted, ranking equal seventh in our sample, with the Mail's celeb-driven "sidebar of shame" notable by its absence. Other parts of the Mail's output were represented, however, and it provided the most shared story on both Facebook and on Twitter: both were strong on visuals, with pictures of a frozen Michigan lighthouse capturing many people's attention on Facebook, while a video on cooking for your pet intriguing Twitter users. It seems fair to conclude that such visual-led items seem to score highly for "shareability" on both Facebook and Twitter. While such "soft" stories top both social media lists, there is also some evidence to back up Phillips's view that Twitter can also drive traffic to harder news. For example, one of the most shared stories on Twitter was about the number of Palestinian children killed by Israeli forces, a story that was presented as an infographic demonstrating the deaths over a period of time. The visual nature of this supports Dick's (2014) finding that something that might not be considered a news story in its own right-because of the lack of a conventional new news angle or "peg"-can emerge as a popular item when data over a period of time is given the infographic treatment. The list of most tweeted stories also included hard news items on the referendum on Scottish independence and about student protests in Venezuela, neither of which achieved similar prominence on Facebook (compare Appendix A and Appendix B).
This preliminary examination of how news values operate on social media is necessarily tentative, and it is worth pointing out that many of the people sharing links may not normally be "readers" of the news site in question and may only have come across it via a link in the first place. Users of social media may even be sharing links while expressing disagreement with-or disapproval of-a particular item. Such is today's journalism landscape.

Conclusion: Towards a Contemporary Set of News Values
Much remains the same, yet much has changed in journalism since Harcup and O'Neill (2001) was published. The importance of visuals, highlighted by Caple and Bednarek (2015) and Dick (2014), is supported by our analysis of news items in both newspapers and on the most shared lists on social media, leading us to conclude that arresting audio-visuals are certainly worth listing as a news value in their own right. There is also evidence to support the arguments of Phillips (2015) and Schultz (2007), among others, that conflict and exclusivity are both worthy of consideration in their own right. Last and probably not least, the findings point towards the importance of a concept we call "shareability". Shareability may be hard to define, although Gibson's "stuff that makes you laugh and stuff that makes you angry" perhaps comes close (Newman 2011, 24), but it appears to WHAT IS NEWS? 1481 be an increasingly important consideration in the selection process (Olmstead, Mitchell, and Rosenstiel 2011;Hermida et al. 2012;Phillips 2012;Tien Vu 2014;Welbers et al. 2015).
Future research could usefully explore the extent to which these factors can be identified in the output of news media across a range of platforms and in different political-economic contexts. In particular, there is a need for empirical research into whether or not news organisations' desire to have their output widely shared on social media may be impacting on selection decisions; and, if so, with what consequences? We offer this updating of Harcup and O'Neill (2001), which was in turn an attempt to update Galtung and Ruge (1965), in the hope that it will prompt more research into news rather than less.
The findings of our new study lead us to propose for discussion an updated set of contemporary news values that, in various combinations, seem to be identifiable within published news stories. Although there will be exceptions, we have found that potential news stories must generally satisfy one and preferably more of the following requirements to be selected: . Exclusivity: Stories generated by, or available first to, the news organisation as a result of interviews, letters, investigations, surveys, polls, and so on.
. Bad news: Stories with particularly negative overtones such as death, injury, defeat and loss (of a job, for example). .

Conflict:
Stories concerning conflict such as controversies, arguments, splits, strikes, fights, insurrections and warfare.
. Surprise: Stories that have an element of surprise, contrast and/or the unusual about them.
. Audio-visuals: Stories that have arresting photographs, video, audio and/or which can be illustrated with infographics.
. Shareability: Stories that are thought likely to generate sharing and comments via Facebook, Twitter and other forms of social media.
. Entertainment: Soft stories concerning sex, showbusiness, sport, lighter human interest, animals, or offering opportunities for humorous treatment, witty headlines or lists.
. Drama: Stories concerning an unfolding drama such as escapes, accidents, searches, sieges, rescues, battles or court cases. . Relevance: Stories about groups or nations perceived to be influential with, or culturally or historically familiar to, the audience.
. Magnitude: Stories perceived as sufficiently significant in the large numbers of people involved or in potential impact, or involving a degree of extreme behaviour or extreme occurrence.
. Celebrity: Stories concerning people who are already famous.
. Good news: Stories with particularly positive overtones such as recoveries, breakthroughs, cures, wins and celebrations.
. News organisation's agenda: Stories that set or fit the news organisation's own agenda, whether ideological, commercial or as part of a specific campaign.
It is important to reiterate that the above criteria can be contested since they are also governed by practical considerations, such as the availability of resources and time, and subjective, often unconscious, influences, such as a mix of the social, educational, ideological and cultural influences on journalists, as well as the environment in which they work, their position in the workplace hierarchy and the type of audience for whom journalists are producing news. These can cause fluctuations, with certain news values rising up the hierarchy in different situations, which may explain why events with similar intrinsic news values are not always given the same prominence. Also for online news, the pressure to obtain clicks and shares will also influence decisions about what news to select, as well as news treatment (Thurman and Myllylahti 2009;Phillips 2012;Bell 2015). In other words, who is selecting news, for whom, in what medium and by what means (and available resources), may well be as important as whatever news values may or may not be inherent in any potential story.
Thus, we reiterate that the above news value taxonomy should be seen as a tool for analysis and further research-designed to provoke discussion and, indeed, contestationnot something to be churned out as if it is the last word on the subject. And, whenever the question arises of why a particular story is seen as newsworthy, there will always remain an element of truth to the reply: "It just is!"

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