10.4185/RLCS-2017-1200en | ISSN 1138 - 5820 | RLCS, 72-2017 | |
Accountability in social networks. Ever-evolving stylebooks and feedback through Twitter
Abstract translation by José Luis Rojas Torrijos (Professor of Journalism at the University of Seville)
In a changing media landscape, characterised by rapid technological innovations, the acceleration of the news cycle, the sectoral crisis and the increasing competition among media organisations to maximise the audience (Brock, 2013; Eberwein, 2010; Schlesinger and Doyle, 2015), information professionals must keep as one of their essential objectives the publication of high-quality content that meets the principles of journalistic ethics and the high standards of stylistic correctness. Given citizens’ growing concern for information quality and the moral foundation of journalists (Plaisance, 2013), the concept of accountability becomes especially relevant in the present context.
Accountability is understood as the commitment of media organisations to be held accountable by society for their practices (Bardoel and d’Haenens, 2004; Harro-Loit, 2015; Mauri-Ríos and Ramón-Vegas, 2015). According to McQuail (2003: 19), accountable communication occurs when professionals “take responsibility for the quality and consequences of their publication, orient themselves to audiences and others affected, and respond to their expectations and those of the wider society”. To be accountable, media companies have to take into account three essential values: self-regulation, transparency and the promotion of the public’s participation (Ramón-Vegas et al., 2016). Based on these three ingredients, media accountability instruments (Bertrand, 2000) play a key role in helping journalists, users and organisations to monitor, comment and criticise the quality of information (Eberwein et al., 2011).
So far, most academic studies on accountability instruments have focused on analysing traditional instruments, such as ethical codes and stylebooks (Alsius, 1999; Aznar, 1999; Himelboim and Limor, 2011; Roberts, 2012), letters to the director (Pastor, 2010) or ombudsmen (Evers, 2010; Maciá, 2006; Starck, 2010). Recent research studies have highlighted how the digital environment has enabled the adaptation of these consolidated accountability instruments and the emergence of new mechanisms, such as editorial blogs, ‘report the error’ buttons, media observatories and criticism on social networks (Fengler et al., 2014; Mauri-Ríos and Ramon-Vegas, 2015; Rosique-Cedillo and Barranquero Carretero, 2016). In the light of this evolution, this article aims to examine in depth how four international media companies have adapted their key accountability instruments, their stylebooks, to the digital environment, and more specifically to Twitter. To this end, we analysed the contents of four pioneering stylebooks in this social network, which have been developed by The Guardian (@Guardianstyle), The Economist (@Econstyleguide), Associated Press (@APStylebook) and BuzzFeed (@Styleguide).
1.1. Stylebooks as professional self-regulation instruments
One of the main accountability systems in journalism (Bertrand, 2000; Eberwein et al., 2011) are stylebooks, which constitute a valuable instrument of self-regulation that helps professionals to fulfil their social responsibility of reporting with rigour and correctness and improve the communication channels with the audience. Stylebooks are regarded as the accountability instrument with the greatest impact on journalists’ work and professional behaviour, according to the results of Media Accountability and Transparency in Europe (Fengler et al., 2014).
Journalistic stylebooks have their origin in the manuals of style of editorial companies, which contained rules for the use of the printing press in English-speaking universities such as Chicago and Oxford at the end of the 19th century (Rojas Torrijos, 2011). These manuals raised awareness within the field of journalism about the need for instruments like these, designed to unify criteria on the use of language and non-linguistic communication procedures (design, colour or photography) in publications; all this in order to enhance the quality of the resulting information product.
Therefore, media stylebooks have a long history, which is initially linked, on the one hand, to the birth of the first schools of journalism in the United States (the Deskbook of The School of Journalism of the University of Missouri in 1910 is considered to be one of the pioneering journalistic manuals); and, on the other hand, to the growth of the leading daily newspapers in this country, which already before the 1920s had developed a manual of this type to standardise the use of language. Outstanding examples of these newspapers are The New York Times, which in 1923 produced its first book of style, which is the precedent of the Manual of Style and Usage. In Europe, meanwhile, the pioneer in journalistic stylebooks was the Style Book of the Manchester Guardian (1928), today The Guardian.
However, the definitive consecration of the stylebooks as contemporary journalistic phenomenon was due to the expansion of the major international news agencies. Given the increasing geographical dispersion of their delegations around the world, these big companies needed to produce texts that clearly collected editorial standards and common writing criteria for all their correspondents and contributors. First, United Press (later United Press International, UPI) published its book of style in 1942 and The Associated Press did the same in 1953.
The arrival of stylebooks in Spain occurred later, mostly thanks to the efforts of the Efe agency, whose first work of this kind dates back to 1975 (Manual de estilo de Efe-Exterior, which is the first antecedent of the current Manual de Español Urgente) (Gómez Font, 2000); and the work of major newspapers such as El País, which produced its first stylebook (Libro de Estilo) in 1977, La Vanguardia (1986) and El Periódico de Catalunya (1986). These media opened up the way and laid out the foundations for the subsequent manuals of style we know today in Spanish journalism, which are much more common in printed than in audiovisual media. Among the latter, there are outstanding works such as the manuals of Radiotelevisión Española (RTVE), Corporació Catalana de Mitjans Audiovisuals (CCMA) and Canal Sur Television.
However, the notion of journalistic style has evolved with the passage of time as a result of a change in the use, content and objectives of this type of manuals by media companies. Thus, the book of style went from being conceived as an instrument of internal consumption and restricted circulation, and known only within the newsrooms, to being a creation of public domain that would be later sold and marketed in bookstores in successive editions. On the other hand, the book of style ceased to be only a set of language rules and writing guides and expanded its objective to include “other regulatory aspects of journalism: those that ensure the production and dissemination of news are ethically and deontologically correct” (Aznar, 2005: 126).
This dual dimension, which contains a set of guidelines to ensure the correct use of language and collects the professional ethical precepts and news production criteria, is precisely what has enriched stylebooks beyond their initial approach and has turned them into key instruments for the current state of journalism and media accountability. Therefore, we can see that stylebooks do not only constitute a mechanism of professional self-regulation and contribute to the consolidation of journalistic ethics, but also represent the media’s commitment to make quality texts that give them credibility.
The media’s dissemination of stylebooks among the wider public has been more evident in recent years after the emergence and development of the Internet as a universal and accessible information platform. In this way, the media increasingly take advantage of the possibilities of the Internet to reach more people, in a better way and with transparency, and show through the contents of these texts their ways of working and their signs of identity in comparison to the competition.
In addition, it should be noted that journalistic organisations have found on the Internet the solution to the aging faced by much of the contents of their stylebooks, which in their traditional print editions were already obsolete at the time of publication. The Internet enables the online hosting of stylebooks and extends their validity, given that there they become dynamic texts that can be updated immediately and continuously at the same rate that language and journalistic innovations take place. Stylebooks have to be reviewed again and again to include neologisms and exclude expressions and terms that cease to be used as before.
Stylebooks find in the Internet their natural habitat. To such an extent that less and less media publish their manuals on paper and, when they do so, in most cases, it is because they also offer an online version. According to the survey carried out by the blog Periodismo Deportivo de Calidad (Quality sports journalism) , there are more than fifty books of editorial and journalistic styles that can be found on the internet today, in both PDF and html formats. The works that have been made available to the public include the Editorial Guidelines of the BBC, the Handbook of Journalism. Standards and Valuesof the Reuters agency, The Style Guide of The Economist and The Guardian Style Guide. In the field of the Spanish language, the list of outstanding works includes the “Manual of style”(Manual de Estilo) of RTVE, the “Style book” (Libro de Estilo) of La Voz de Galicia, the “Guiding principles” (Principios Rectores) of El Comercio (Peru), and the“Writer’s Manual” (Manual del Redactor) of RCN (Colombia).
The clearest exponent of this latest generation of manuals of style is the Stylebook of the American Associated Press (AP), which combines the paper version, of around 450 pages (typically republish each spring) with the digital version of the work, which users can access after subscription payment and whose reviews and expansions are incorporated few months after the venal Edition. Since 2015, it is also an interactive ebook  that facilitates consultation by the user. Equally, the so-called ‘Bible of journalists’ also has a Spanish-language version, which, in this case, is only available online from December 2013. The Spanish Manual of style of the AP, which is smaller than the English-language version, is adapted to be read on mobile phones and tablets, and stands out for paving the way in the study of language. In fact, it is a work conceived for the news production of an international news agency that has to use a standard language that is valid in all latitudes of the Spanish language, which reaches 400 million speakers worldwide.
1.2. Participatory stylebooks based on interaction through social networking sites
The latest trend of the digital journalistic stylebooks is to feed on the permanent interaction with readers through social networks. These platforms, as indicated by recent research on the subject, are not only a content-dissemination platform, but also a key space for accountability in which audience participation plays an essential role (Fengler et al., 2015; Mauri-Ríos and Ramon-Vegas, 2015).
In fact, Twitter, despite not being the social network with the largest number of users today (317 million active users against the 1,871 million users of Facebook, according to data by Statista for January 2017) , is the preferred platform for many users when it comes to searching sources, sharing news and communicating with journalists and media organisations through their respective accounts. The microblogging platform has a series of characteristics that resemble a traditional news organisation, so that it is primarily used to distribute and share news (Kwak, Lee, Park and Moon, 2010), unlike other social networks with less journalistic orientation. Moreover, although it is a social network is constant evolution, with new services, instruments and even formats created by tweeters on a daily basis (Fitton, Gruen and Poston, 2010), “Twitter is, above all, shared information and ongoing feedback about what is said” (Tascón and Abad, 2011: 12).
The immediacy, accessibility and high degree of interaction enabled by Twitter have been determining factors in the choosing of this social network by various media organisations to present their books of style, through accounts that share fragments of the guidelines of style included in their respective manuals and offer very up-to-date recommendations about the use of terms and phrases that raise doubt or controversy.
This has been done so far by media organisations with pioneering manuals of style at the international level, such as the British Newspaper The Guardian, the great news agency The Associated Press (in English and in Spanish), the weekly newspaper The Economist and, more recently, the born-digital Buzzfeed. These stylebooks base their raison d’être on Twitter because they feed on the virtual interactions and conversations that editors establish on this platform with users, who speak out and offer suggestions on stylistic and ethical issues. In many cases, this audience feedback is what prompts stylebooks to be modified, expanded and fully updated.
Other digital and traditional media have also launched Twitter accounts for their style guides or similar texts (like Yahoo’s news channel back in the day and the British tabloid Daily Mirror) and editorial manuals, like the one developed by the University of Chicago Press. However, this study will focus on the analysis of the contents of the accounts of four stylebooks developed by international, English-language media organisations: The Guardian, The Associated Press, The Economist and Buzzfeed.
On this basis, the main objective of this research study is to analyse the publications of four dynamic stylebooks on Twitter. In the first place, The Guardian Style Guide (@guardianstyle) was the first book of style on Twitter. It has published more than 17,000 tweets since April 2010 and has more than 67 thousand followers. In second place, the AP Stylebook (@APStylebook) is the account of The Associated Press Stylebook and Briefing on Media Law, whose first edition dates back to 1953. It is the Twitter account with the largest number of followers, 261,000, and the most active along with The Guardian, since it has made more than 6,300 posts from March 2009. In third place, we examine The Economist Style Guide (@Econstyleguide), whose last update was published on the weekly newspaper’s website and whose last print edition dates back to 2012. Its Twitter account was opened in March 2013 and since them its timeline has grown in number of followers, going from 600 in the first months to 25,900 today. It has made more than 2,000 posts since its launch. Finally, we examined BuzzFeed Style Guide (@styleguide), which so far is the latest account of a journalistic book on Twitter, where it was launched in November 2014. Since then it has posted more than 1,300 messages and has collected a base of 6,700 followers.
The specific objectives of this research are to examine: the volume and frequency of posts (SO1); the interaction of these accounts with users (SO2); the use of hashtags and links (SO3); the level of engagement  (retweets and favourites) generated (SO4); and the stylistic and ethical nature of the contents published (SO5). This latter objective aims to determine whether the contents published by these Twitter accounts pursue an eminently linguistic function or also address ethical issues of the profession, as it occurs in the commercial and digital editions of many of these manuals.
To ensure the rigour of the study and the comparability between the examined accounts, the observation was carried out over a broad time period, between 1 January 2015 and 31 December 2016 (two full years of content published by the four accounts that make up the sample).
2.2. Research technique and procedure
Quantitative and qualitative content analysis was used to achieve the main and specific objectives of this work (Altheide, 1996; Bryman, 2016; Wimmer and Dominick, 2011). The following section explains the methodological strategy of the work, detailing the instruments and procedures used to gather and examine the data.
Twitonomy, a tool for the monitoring and analysis of social networks (http://twitonomy.com), was used to obtain the raw data needed to carry out the content analysis. This online application, created by Digitonomy, has recently been employed in other research studies on communication that aim to examine content published on Twitter (García-Avilés et al., 2016; Marcos García, Alonso Muñoz and Casero Ripollés, 2017; Moya Sánchez and Herrera Damas, 2016; Barrios Rubio, 2016).
Like other programs used for this purpose, Twitonomy can access data on content published through Twitter API (Application Programming Interface). This interface can only retrieve the last 3,200 tweets published by any open account. To ensure the retrieval of all the contents published between 1 January 2015 and 31 December 2016, data were extracted from 18 January 2017. Taking into consideration the period selected for the sample, the total number of downloaded tweets was n= 5,511. The origin of messages is as follows: @Guardianstyle (n= 2,380), @APStylebook (n=1,468), @Econstyleguide (n=732) and @Styleguide (n=931).
The application provided essential contextual data of each tweet (account and date of publication), the full text of each message and its engagement data (retweets and favourites). These raw data, as well as the general metrics provided by Twitonomy were processed with Microsoft Excel to obtain the results and observations needed to respond to the specific objectives SO1, SO2, SO3 and SO4.
To achieve a greater level of depth and detail of the stylistic and ethical nature of the content published (SO5), content was coded in two stages. In the first stage, the sample of units (n= 5.511) were classified limit the sample to the tweets that made reference to ethical and stylistic issues (n= 2.416), and exclude exclusively promotional tweets, retweets and replies. As we can see in table 1, the weight of the latter type was remarkable in The Guardian (@Guardianstyle) and Buzzfeed (@Styleguide).
In the second phase, the Tweets referring exclusively to ethical and style issues (n=2,416) were coded through a uniform system of reliable, mutually exclusive categories that were adjusted to the specific purposes of the research. To do this, we designed a coding sheet that included 81 items that allowed us to register the issues addressed in each message (see table 2).
Table 1. Typology of tweets published over the period of study.
Source: Authors’ own creation
This coding sheet is based, on the one hand, in the 62 items formulated by Salvador Alsius in his thesaurus on journalistic ethics (Tesauro sobre la Ética Periodística), built from the contributions of Cooper (1989), Day (1991), Lambeth (1992) and Black, Steele and Barney (1995). This thesaurus, which has become consolidated as a solid working tool in competitive national and international projects (Alsius, 2010; Alsius et al., 2011), systematises the four great principles of journalistic ethics (truth, fairness, freedom and responsibility), detailing the specific issues addressed in each principle. On the other hand, the coding sheet was complemented with 19 items referring to style elements (general writing and titling principles; spelling and orthographic rules and recommendations; grammatical recommendations; recommendations on the use of lexicon; and other style issues). For the creation of these items we considered the categories used by many of the aforementioned national and international stylebooks.
Source: Authors’ own creation based on Alsius’s thesaurus (2010) and the categories of other manuals of style.
Before the systematic analysis of the entire sample, we conducted a pre-test with 400 tweets (100 messages from each account) to ensure the reliability of the category system. After making the necessary adjustments, all of the tweets were encoded in Microsoft Excel. Finally, following the indications of Brennen (2013: 24), we extracted tables and quantitative results, made qualitative observations, formulated examples, interpreted the results and drew the conclusions of the study.
Table 3. Volume and frequency of posts from 01/01/2015 to 31/12/2016
Source: Authors’ own creation
The first analysis of the data extracted from @APStylebook, @Guardianstyle, @Econstyleguide and @Styleguide shows clear differences between the platforms studied. Firstly, it is observed that @Guardianstyle was the account that made the largest number of posts during the analysed period (n=2,380 messages, with an average of 3.26 tweets per day), followed by @APStylebook (n=1.468 tweets; 2.01 posts per day), @Styleguide (n=931 tweets; 1.27 posts per day) and, finally, @Econstyleguide (n=732 tweets, with a daily average of 1 post per day) (see table 3). However, as it is specified below, @Guardianstyle and @Styleguide included a high percentage of retweets in their timelines.
Table 4. Volume of posts per day during the analysed period
Source: Authors’ own creation
The analysis of the publication day of the contents (table 4) indicates that most messages from @APStylebook, @Guardianstyle and @Styleguide were posted between Monday and Friday. On the contrary, @Econstyleguide did not prioritise any day of the week to publish its contents.
The examination of the evolution of posts during the analysed period (see Figure 1) reveals, on the one hand, the regular publication of @Econstyleguide; and the constant figures of @Styleguide and @APStylebook (altered by publication peaks in specific months, associated with the celebration of certain events, such as the Conference of the American Copy Editors Society, which BuzzFeed covered intensively with the hashtag #ACES2016, and the development of online chat by the Associated Press, which were identified with the hashtag #APStyleChat). On the other hand, it is important to note the remarkable decline in the volume of publication of @Guardianstyle between 2015 and 2016.
Figure 1. Evolution of publications in the sample of accounts in the 2015-2016 period
Source: Authors’ own creation
3.2. User interactions: mentions, replies and retweets
With regards to the interaction with users, the analysis detected a surprising fact. While @Guardianstyle, @APStylebook and @Styleguide usually mentioned other accounts (in 864, 504 and 646 tweets, respectively), @Econstyleguide did not make any reference to any account, as shown in table 5. The Economist has also been characterised for never responding to other Twitter users, in clear correspondence with its definition in this social network (“The Economist’s official style feed. Our word is final”). On the contrary, replies are an essential ingredient in @Guardianstyle (15.25% or 363 of its tweets were replies), @APStylebook (8.51% or 125 of its tweets) and @Styleguide (21.05% or 196 of its tweets).
Table 5. Data on interaction of accounts with other Twitter users
Source: Authors’ own creation
Another defining element to measure the interactive nature of these accounts is the number of messages they retweet, which also encourages conversation on social networks. In @APStylebook and @Econstyleguide the volume of retweets is below 2%. To be precise, @APStylebook only retweeted 26 messages during the analysed period (1.77% of its contents) while @Econstyleguide only retweeted two messages (0.27% of its content). Thus, @APStylebook and @Econstyleguide arethe two accounts that publish the most content of their own and most of these contents are recommendations included in their respective stylebooks, although this is not always made explicit. Therefore, in these two cases Twitter is used to amplify the dissemination of style information contained in the style manuals, which are accessible for a fee (the print edition of The Economist and the for-sale and web editions in the case of AP) and whose access and recognition is more restricted.
This fact contrasts clearly with the model followed by @Guardianstyle and @Styleguide, which have a high percentage of retweets. To be precise, 495 of the messages published by @Guardianstyle were retweets (20.8%) while 25.99% of the content published by @Styleguide were retweets (242 posts). This signals the trend of these accounts to go beyond advertising their books of style to open up the focus to other interesting publications. Regarding @Guardianstyle, its more retweeted accounts include individual users, mainly journalists, both from The Guardian and other British media, as well as celebrities, like footballer and TV commentator Gary Lineker and actor and director David Schneider, and other accounts that parody the news.
For his part, Buzzfeed retweeted individuals who commented on the errors published by any media as well as other stylebooks and accounts of reference, such as the online edition of the Oxford Dictionary (@OxfordWords), the American English dictionary @MerriamWebster and @Dictionarycom, one of the most followed digital dictionaries in this language. In this respect, it is important to note that the fact that such accounts retweeted each other give relevance to that the content of their stylebooks. For example, in the following message (Figure 2), the retweets of @Styleguide include @Guardianstyle (which, in turn, quotes and replies to a user on the use of the coma in enumeration, the so-called Oxford comma).
Sometimes, users themselves mention in a same Tweet the different stylebooks with presence on Twitter to answer a question, which generates a conversation between these accounts. The fact that there are crossed references between stylebooks on Twitter means that some serve as reference to others, that these works, with the exception of @Econstyleguide (which is self-sufficient and does not follow other account in this social network) feedback from each other and are routed together towards a common goal: educate users about the importance of knowing and using these instruments of professional self-regulation.
Figure 2. Tweet of The Guardian’s stylebook as retweeted by Buzzfeed’s Styleguide
3.3. Use of hashtags, links and multimedia elements
Twitter, throughout its evolution, has developed its own language, enriched with signs and formats that characterise its 140-character-long messages and differentiate it from other social networks. Another key aspect to know the role played by these stylebooks on Twitter is, therefore, analyse the use of hashtags and links in the tweets. On the one hand, the inclusion of hashtags can move a conversation beyond the source account and attract and engage a greater number of users. Furthermore, the inclusion of links in a tweet provides the context that complement the explanation of what is exposed briefly there. In this regard, it is relevant to point out that there are major differences between the four accounts examined.
While @Guardianstyle, @APStylebook and @Styleguide very frequently used links in their tweets, it is surprising to note that @Econstyleguide has not published a single link during the analysed period, not even a link to The Economist’swebsite (see table 6). This fact, combined with the non-existent interaction with users pointed out above, clearly reinforces the insular and unidirectional nature of the account of the stylebook of this British medium. It should also be noted that @Econstyleguide subverts another essential component in Twitter: it did not include any hashtag in the messages it posted during the analysed period. In contrast, @APStylebook was the account that used the widest variety of hashtags (712) and with the highest frequency (0.49 hashtags per tweet), followed by @Styleguide (132 hashtags and an average of 0.14 hashtags per message). The variety of hashtags used by @Guardianstyle is much smaller (33) and their presence is more discreet (only 0.01hashtags per Tweet), which limits the possibilities to open up social discussions through this formula.
It is relevant to comment that, on a regular basis, the AP news agency invites the user to participate in a conversation with the editors of the stylebook through the hashtag #APStylechat. It is an event periodically held on Twitter that allows followers of this account to make questions about the manual of style and know how it resolves doubts that arise daily. AP clearly uses its Twitter account to expand the use and awareness of its famous book of style among citizens. Thus, for example, ait used on several occasions the hashtag #ACES2016, in allusion to the Conference of the American Copy Editors Society, where the updated edition of the AP style book was presented in the spring of 2016.
Table 6. Data on the use of links and hashtags in the examined accounts
Source: Authors’ own creation
Besides the hashtags, the inclusion of multimedia elements (pictures, videos and graphics) in accounts aimed at disseminating knowledge about the correct use of the language and good journalistic practices can be assumed to further reinforce their explanatory and didactic function. In this sense, the stylebook that takes advantage of the reach of Twitter to promote education is @APStylebook. The account of the American news agency creates style information pieces, images with the logo of the medium over the recommended and defined word (Figure 3).
Figure 3. Information piece of AP’s stylebook
While @Econstyleguide is purely textual and does not include links or images in its tweets, the rest of the accounts often use multimedia resources (photographs of newspaper clippings and screenshots that explain and exemplify, and even humorous GIFs) and links to the source (news or article published in a medium, dictionary and even the website of the stylebook). In the case of @Guardianstyle, the inclusion of links to The Guardian’s website’s “Mind Your Language” section is very common (https://www.theguardian.com/media/mind-your-language), which delves into the editorial decisions made in reference to the use and abuse of language (see Figure 4).
Figure 4. Tweet with link with multimedia contents published by @Guardianstyle
3.4. Engagement generated by the examined accounts: retweets and favourites
Data on the engagement generated by the contents of the four studied accounts (retweets and favourites) also provided elements of interest (see table 7). Despite that, as mentioned, the model of @Econstyleguide is characterised by the lack of interaction with other users and does not include links and hashtags, it is striking to see that almost all of its messages -99.18% (726)- were retweeted. In fact, for every 100 followers The Economist gets 47.42 retweets on average, which reveals the loyalty of its followers, trust on its messages, and the credibility of the medium. On the other hand, 92.37% of the contents published by @APStylebook (1356) were also retweeted, while in the case of @Guardianstyle and @Styleguide (Buzzfeed), the volume of retweeted messages was markedly lower (61.47% and 40.71% of their contents, respectively). Now, if we focus on the total number of retweets obtained by the accounts, we can note that @APStylebook led the ranking with 54,921 retweets, followed by @Guardianstyle (23,704), @Econstyleguide (11,568) and @Styleguide (2,612).
Table 7. Data on engagement generated during the analysed period (retweets and favourites)
Source: Authors’ own creation
With regards to the messages that were marked as favourite by users, we can also note the great impact of the contents published by @Econstyleguide (99.73% of its posts -730 tweets- were favourited). 95.1% of the tweets of @APStylebook (1,396 messages) were favourited. In the case of @Guardianstyle and @Styleguide, the percentage of favourited tweets were 67.94% and 57.79% respectively. The total number of favourited tweets obtained by the accounts examined was as follows: @APStylebook (63,480); @Guardianstyle (23,613); @Econstyleguide (13,818); and @Styleguide (4,411).
3.5. Analysis of ethical and style issues
The analysis of the tweets that refer exclusively to ethical and stylistic issues (n=2,416) also provides relevant results that reflect the nature and purpose of these dynamic journalistic stylebooks in the microblogging network. As it can be seen in table 8, the four stylebooks performed an eminently stylistic function and devoted much less space to ethical issues.
Table 8. Function of the tweets of the four accounts under study
Source: Authors’ own creation
3.5.1. Ethical issues
Of the 62 ethical issues considered in the coding sheet, only 27 were reflected in the tweets of one or more of the Twitter accounts under study (see table 9).
Table 9. Ethical issues addressed in the tweets published by the four accounts under study
Source: Authors’ own creation
If these issues are grouped together according to the four ethical principles formulated by Alsius (2010), we can notice that the issues addressed the most in the four stylebooks on Twitter are related to the principles of truthfulness and fairness (see table 10). However, it should be noted that the Associated Press and The Guardian prioritised issues related to the principle of truthfulness, while Buzzfeed and The Economist emphasised elements circumscribed in the principle of fairness. Broadly, the questions related to the principle of responsibility had a lower impact. Contents on the principle of freedom were virtually non-existent, except for a tweet posted by @Styleguide.
Table 10. Grouping of ethical issues in the four fundamental ethical principles (Alsius, 2010)
Source: Authors’ own creation
On the principle of truthfulness, @APStylebook tweeted 17 messages about the selection, quoting and credibility of sources, using the hashtag #APStyleChat, providing specific recommendations on the rigour and transparency in the publication of data from surveys (sample selection, formulation of questions, the agencies that fund and/or carry out the survey, interpretation of results). This same account devoted 7 tweets to the ethical dimension in the treatment of images and captions, providing guidance, as the example shown in Figure 5.
Figure 5. @APStyleguide’s tweet on the treatment of images
@Econstyleguidepublished a series of recommendations about accuracy and precision and the importance of separating news and comments (for example, the tweet “Like “important” and “funny”, “interesting” makes assumptions about the word it describes that may not be shared by the reader”). For its part, @Guardianstyle gave great importance to detection and explanation of errors of other media (56 tweets) and to the rectification of its own mistakes (11 tweets). The exposition of errors is usually performed with a didactic yet relaxed tone (see Figure 6), with the aim of generating more engagement among users.
Figure 6. Example of corrections in a tweet posted in @Guardianstyle
Recommendations on other areas such as speculations and rumours and anonymity and pseudonymity were less frequent among the range of ethical issues linked to the principle of truthfulness.
On the principle of fairness, a great interest has been demonstrated by the accounts under study which offer recommendations aimed at preventing discrimination on the grounds of sex, race, disability or sexual orientation, which is the guiding principle set out in the main transnational ethical codes (UNESCO, Art. 9; International Federation of Journalists, Art. 7; Council of Europe, article 33). On the occasion of the USA 2016 election campaign, three of the four platforms (@APStylebook, @Guardianstyle and @Styleguide) criticised the use of the term “alt-right”. The account of AP also pointed out the difference between “migrants” and “refugees”. On the other hand, @Econstyleguide opted to avoid the adjective “browns” (“Do not call people who are neither pure white nor pure black browns”). @Styleguide warned of the discriminatory use of certain verbs (“We often see the same things described differently: Black people ‘riot’ and ‘loot’; white people ‘protest’ and ‘find’).
In addition to addressing gender issues, the accounts under study provided recommendations to treat gender identity (see, for example, this tweet from AP: “Use the term cross-dresser, not transvestite”) and news about minors or persons suffering from diseases or disabilities. In one of the tweets that generated more engagement (posted on 10 October 2016), @APStylebook reflecting a clear commitment to the inclusive language by promoting the use of the term “wheelchair user” instead of “confined to a wheelchair” or “wheelchair-bound” when referring in a news story to people on wheelchair. Other issues related to the principle of fairness, and the right of reply or the presumption of innocence had less impact, with the exception of a series of tweets published by @APStylebook in September 2016 to provide guidelines to inform about detainees and suspects.
As mentioned, these accounts have not offered recommendations on issues related to the principle of freedom, with the exception of the tweet posted by @Styleguide on the rights of sources (“Engage with the groups and communities you write about and edit – do more listening than talking, but ask q's respectfully”). The accounts under study did not publish, during the period of analysis, tweets about the rejection of political, economic and any kinds of interference.
Finally, with regards to the principle of responsibility, it is worth noting that in 2015 @APStylebook offered a series of recommendations to inform about suicide, advocating for the use of phrases like “killed himself, took her own life or died by suicide” instead of “committed suicide”, an expression that involves an illegal action as it was also noted by BuzzFeed. For its part, @Econstyleguide reflected on terrorism and insurgency, and recommended avoiding the use of the term “terrorist” (“Use the word “terrorist” with care, preferably only to mean someone who uses terror as an organised system of intimidation”). It also encouraged avoiding violence and bad taste, as well as the need to offer a respectful treatment towards different religious beliefs.
3.5.2. Style issues
Regarding questions of style, the analysis shows that the four accounts have mostly focused on providing recommendations on lexicon (see table 11). Moreover, they have also devoted considerable attention to the promotion of spelling and grammar guidelines.
Table 11. Tweets about style issues posted by the four analysed accounts
Source: Authors’ own creation
The following table presents the detailed results for each of these areas results:
Table 12. Style issues included in the tweets posted by the four accounts under study
Source: Authors’ own creation
First of all, regarding recommendations on the use of lexicon, there is a predominance of tweets that aim to expand the knowledge of users about the range of existing words in a multitude of fields of specialisation. For example, in the field of sports, @APStylebook explained the meaning of hat trick (“three goals scored by one player in one game, usually referring to soccer or hockey”) and even of the term “tiki-taka” (“a system of intricate, one-touch and rapid passing artistry developed by Spanish soccer club Barcelona”). Often, the accounts examined also report on the updating of language through the emergence of neologisms, such as “emoji”, “metadata”, “e-commerce” and “spoilerer”. In addition, certain dates are used to delve into the origin of words. For example, @APStylebook took advantage of “Boxing Day” for this informative task (“Today is Boxing Day in British Commonwealth countries. The term came from the practice of giving gift boxes to employees and others”). The accounts usually also point out the lexical inaccuracies to avoid. At this point, the task carried out by @Econstyleguide is particularly active (through precisions such as “The people in trains and buses are passengers, not riders”).
The four accounts devoted attention to proper names, names of places and names of brands, in particular @APStylebook. On the other hand, @Guardianstyle, @Econstyleguide and @Styleguide focus more on issues about foreign words and geographical variations of language, especially those that have to do with the differences between British and American English, as illustrated in Figure 7.
Figure 7. Explanation of geographic variations of language in The Economist
As mentioned, the guidelines on spelling and orthographic issues represent the second favourite aspect of style for the four accounts. In particular, @APStylebook and @Styleguide are the accounts that most frequently tweet guidelines on punctuation marks. The accounts commonly tweet about the use of capital letters (“Capitalize president only as a formal title before names: President Barack Obama, President-elect Donald Trump”), hyphen (“Hyphenate great-grandfather, great-great-grandmother”) and other marks, such as the semicolon (“Semi-colons should be used to mark a pause longer than a comma and shorter than a full stop. Don’t overdo them”).
Recommendations on abbreviations, initials and acronyms (such as “NCAA is acceptable in all references to the National Collegiate Athletic Association”) play a particularly important role in the aim of AP, although they are also addressed by other platforms like @Guardianstyle (“QED stands for quod erat demonstrandum - ‘which was to be shown’”). The treatment of figures and percentages is common in all accounts, with explanations like this on BuzzFeed: “Use figures in NYC street names (6th Street, 23rd Street), and spell out the numbers in avenues (Second Avenue, Tenth Avenue)”. The attention devoted to typography, as shown in table 12, is very limited.
In third place, between grammatical rules and recommendations, the attention devoted by @Econstyleguide and @Styleguide to the formulation of articles, pronouns, adjectives, adverbs, conjunctions, and prepositions is outstanding. Other recommendations are related to gender and grammatical number and the formation of tenses. In relation to this last aspect, The Economist reminds its followers that it is allowed to use regular and irregular participles in some verbs through the tweet: “Burned or burnt? Dreamed or dreamt? Dwelled or dwelt? Leaped or leapt? Learned or learnt? Modern English allows both”.
In a last order of things, it was precisely @Econstyleguide which devoted more space to tweet general writing recommendations (precision, clarity, expressive simplicity and correction). The recommendations contained in its book of style include the following: “Avoid the language of spokesmen, businessmen, lawyers, bureaucrats and academics”, that is, write and speak as a journalist). It advises followers to write shorter rather than longer texts and to use as many points as possible to maintain short sentences and to facilitate readers’ understanding of the text. When it recommends followers to write more concisely, it also reminds people of Voltaire (“The best way to be boring is to leave nothing out”). Interestingly, its most engaging tweets include a quote from George Orwell, who is mentioned in the introduction to the web edition of the stylebook (see Figure 8).
Figure 8. Quote from George Orwell in a tweet of The Economist’sstylebook
Other questions addressed by these accounts include style guidelines for titling elements, body text, journalistic genres and pronunciation and phonetics.
4. Discussion and conclusions
The analysis of the stylebooks of The Guardian (@Guardianstyle), The Economist (@Econstyleguide), Associated Press (@APStylebook) and BuzzFeed (@Styleguide) reveals how these accountability instruments have managed to adapt successfully to the social network Twitter, where they can evolve continuously and feedback through the participation of users. Apart from the differences between accounts with regards to the volume and frequency of tweets and the engagement figures obtained, this study offers a series of conclusions of the first order.
First of all, it is observed that the participatory and multimedia model of @Guardianstyle, @APStylebook and @Styleguide (based on maximising mentions and retweets -including between themselves-, ensuring smooth interaction with other users and maximising the publication of links, tags, and images to carry out dissemination and pedagogy) contrasts very notoriously with the weak capacity of interaction and use of multimedia possibilities that characterises the normative, insular and one-way model of @Econstyleguide. This second model clearly departs from the two key features promoted by the innovative accountability instruments in the digital environment: dynamism and the ability to feedback from the audience, with the aim of making it jointly responsible for the quality and accuracy of the content published by the news media (Mauri-Ríos and Ramon-Vegas, 2015).
Secondly, it is surprising, if we consider the double traditional dimension of stylebooks, that the volume of content addressing ethical issues is so small in comparison to the recommendations about style. As the analysis has shown, the range of ethical issues includes elements of the different ethical principles (especially those of truthfulness, fairness and responsibility). The ethical recommendations -ranging from guidelines on the use of sources to all kinds of guidelines to treat vulnerable groups- can be of great value for news professionals, beyond the four organisations under study. However, this imbalance between volume of ethical issues in comparison to standards of stylistic correctness limits decisively the possibilities of these accounts for the purposes of accountability. If we look at the fifteen most retweeted tweets and the fifteen most liked tweets of the four accounts, we can see that all of them are recommendations or thoughts about the proper use of words and expressions in the everyday journalism. On the contrary, tweets addressing ethical aspects of the profession they are conspicuous by their absence.
Having said that, it is important to point out that these accounts on Twitter, when addressing both ethical and stylistic issues, present an extraordinary advantage over traditional printed style and web-based manuals: their permanent updating. The capacity demonstrated by the stylebooks under study is particularly relevant to promote and legitimise terms from multiple areas and to disseminate neologisms. This dynamism is also important to spread ethical guidelines on specific areas such as the treatment of sexual diversity and mental health in the media, which are areas poorly addressed in the conventional stylebooks and ethical codes. All this is done without neglecting the publication of essential grammar and spelling principles to enable professionals to develop quality information.
Taking into account these results, future research should take into consideration the following points. Firstly, future works should follow carefully the evolution of the stylebooks of these four media organisations and of other national and international news organisations, both on Twitter and other social networks that might emerge. In particular, it will be of special interest to observe whether the Spanish media which have decidedly opted to promote and maintain stylebooks throughout their history, such as El País, RTVE and CCMA adopt Twitter as a complementary platform to their paper and website outlets for the dissemination of their contents. Secondly, in-depth interviews with the senior managers and officials of these organisations must be used to know the motivations, goals and work routines of the professionals involved in the creation of content for these dynamic stylebooks.
Thirdly, the application of data collection techniques such as surveys, interviews and focus groups would also allow to discern the value and usefulness of these mechanisms between users (journalists and citizens) who employ the stylebooks on Twitter and to detect possible improvements. Ultimately, researchers must be aware of the emergence of new accountability instrumentsin the digital environment driven by journalistic organisations, external agents and citizens, to examine their performance and usefulness to ensure the quality of news contents.
Altheide, D.L. (1996): Qualitative media analysis. Thousand Oaks, California: Sage.
Alsius, S. (1999). Codis ètics del periodisme televisiu. Barcelona: Pòrtic.
Alsius, S. (Ed.) (2010). The ethical values of journalists. Field research among media professionals in Catalonia. Barcelona: Generalitat de Catalunya.
Alsius, S.; Mauri, M.; Rodríguez-Martínez, R. (2011). “Spain: a diverse and asymmetric landscape”. En: Eberwein, T.; Fengler, S.; Lauk, E.; Leppik-Bork, T. (Eds). Mapping media accountability – in Europe and beyond. Köln, Germany: Herbert von Halem. pp. 155–167.
Aznar, H. (2005). Comunicación responsable. Barcelona: Ariel.
Bardoel, J.; d’Haenens, L. (2004). “Media responsibility and accountability: New conceptualizations and practices”. Communications, 29(1): 5–25.
Barrios Rubio, A. (2016). “Narrativa periodística en la convergencia de medios”. Estudios sobre el Mensaje Periodístico, 22(1): 163–176.
Bertrand, C.J. (2000). Media ethics and accountability systems. London: Transaction Publishers.
Black, J.; Steele, B.; Barney, R. (1995). Doing ethics in journalism: a handbook with case studies (2ª ed.). Boston, Massachusetts: Allyn and Bacon.
Brennen, S. (2013). Qualitative research methods for media studies. Nueva York, Routledge.
Brock, G. (2013). Out of print. Newspapers, journalism and the business of news in the digital age. London: Kogan Page.
Bryman, A. (2016). Social research methods (5th edition). Oxford, Oxford University Press.
Cooper, T.W. (1989). “Global universals: in search of common ground”. In: Cooper, T.W.; Christians, C.G.; Plude, F.F.; White, R.A. Communication ethics and global change. White Plains, Nueva York: Longman. pp. 3–19.
Day, L.A. (1991). Ethics in media communications: cases and controversies. Belmont, California: Wadsworth.
Eberwein, T. (2010). “From Woodburg to Blogville – and black. Media observation daily press and weblogs”. In: Eberwein, T.; Müller, D.(Eds.). Journalismus und öffentlichkeit. Wiesbaden: Verlag, pp. 143-165.
Eberwein, T.; Fengler, S.; Lauk, E.; Leppik-Bork, T. (eds.) (2011). Mapping media accountability – in Europe and beyond. Colonia: Helbert Von Halem Verlag.
Evers, Hubb (2010). The news ombudsman. Watch-dog or decoy? Diemen: AMB.
Fengler, S.; Eberwein, T.; Mazzoleni, G.; Porlezza, C.; Russ-Mohl, S. (2014). Journalists and media accountability. An international study of news people in the digital age. Nueva York: Peter Lang Publishing.
Fengler, S. et al. (2015). “How effective is media self- regulation? Results from a comparative survey of European journalists”. European Journal of Communication, 30(3): 249–266.
Fernández Peña, E.; Ramajo, N.; Arauz, M. (2014). “Social media in the Olympic Games: actors, management and participation”. In: Billings, A.C.; Hardin, M. (Eds.). Routledge Handbook of Sport and New Media. London: Routledge. pp. 153–164.
Fitton, L.; Gruen, M.; Poston, L. (2010). Twitter for dummies. Indianápolis: Wiley Publishing.
García-Avilés, J.A.; Carvajal-Prieto, M.; De Lara-González, A.; Arias-Robles (2016). “Developing an Index of Media Innovation in a National Market. The case of Spain”. Journalism Studies. Published online: 24 March 2016.
Gómez Font, A. (2000). “Los libros de estilo de los medios de comunicación en español: necesidad de un acuerdo”. Espéculo, Revista de Estudios Literarios, 16. https://pendientedemigracion.ucm.es/info/especulo/ele/g_font.html
Harro-Loit, H. (2015). “Journalists’ views about accountability to different societal groups”. Journal of Media Ethics: Exploring Questions of Media Morality, 30(1): 31–43.
Himelboim, I.; Limor, Y. (2011). “Media Institutions, News Organizations, and the Journalistic Social Role Worldwide: A Cross-National and Cross-Organizational Study of Codes of Ethics”. Mass Communication & Society, 14(1): 71–92.
Kwak, H; Lee, C.; Park, H.; & Moon, S. (2010). "What is Twitter, a Social Network or a News Media?” Proceedings of the 19th International Conference on World Wide Web. 26-30 April, Raleigh (EE.UU.) http://www.ambuehler.ethz.ch/CDstore/www2010/www/p591.pdf
Lambeth, E. (1992).Committed journalism. An ethic for the profession. Bloomington: Indiana University Press.
Maciá, C. (2006). La figura del defensor del lector, del oyente y del telespectador. Madrid: Universitas.
Marcos García, S.; Alonso Muñoz, L.; Casero Ripollés, A. (2017). “Usos ciudadanos de Twitter en eventos políticos relevantes. La #SesiónDeInvestidura de Pedro Sánchez”. Comunicación y Hombre, 13: 25–49.
Mauri-Ríos, M.; Ramon-Vegas, X. (2015). “Nuevos sistemas de rendición de cuentas de la información periodística. Una exploración del escenario online español”. El profesional de la información, 24(4): 380–389.
McQuail, D. (2003). Media Accountability and Freedom of Publication. Nueva York: Oxford University Press.
Moya Sánchez, M.; Herrera Damas, S. (2016). “Cómo medir el potencial persuasivo en Twitter: propuesta metodológica”. Palabra Clave, 19(3): 838-867.
Pastor, L. (2010). Teoría de las cartas al director. La gestión periodística del público. Barcelona: UOC.
Plaisance, P.L. (2013). Media ethics. Key principles for responsible practice (segunda edición). Thousand Oaks, California: Sage.
Roberts, C. (2012). “Identifying and defining values in media codes of ethics”. Journal of Mass Media Ethics: Exploring Questions of Media Morality, 27(2): 115–129.
Rojas Torrijos, J.L. (2013). “Libros de estilo en continuo movimiento a través de Twitter”. En Periodismo deportivo de calidad http://periodismodeportivodecalidad.blogspot.com.es/2013/04/libros-de-estilo-en-continuo-movimiento.html
Rojas Torrijos, J.L. (2011). Libros de estilo y periodismo global en español. Valencia: Tirant Humanidades.
Rosique Cedillo, G.; Barranquero Carretero, A. (2016). “Analysis of media observatories in Spain. A tool for civil society in media reform processes”. Revista Latina de Comunicación Social, 71: 443–469.
Schlesinger, P.; Doyle, G. (2015). “From organizational crisis to multi-platform salvation? Creative destruction and the recomposition of news media”. Journalism, 16(3): 305–323.
Starck, K. (2010). “The news ombudsman: viable or vanishing?”. En: Eberwein, T.; Müller, D. (Eds.). Journalismus und öffentlichkeit. Wiesbaden: Verlag, pp. 109-118.
Tascón, M.; Abad, M. (2011). Twittegrafía. Madrid: Catarata.
Wimmer, R.D.; Dominick, J.R. (2011). Mass media research: an introduction (novena edición). Canadá: Wadsworth, Cengage Learning.
How to cite this article in bibliographies / References
JL Rojas Torrijos, X Ramón Vegas (2017): “Accountability in social networks. Ever-evolving stylebooks and feedback through Twitter”. Revista Latina de Comunicación Social, 72, pp. 915 to 940.
Article received on 22 June 2017. Accepted on 30 July.