How to Reconcile Water and Fire: Social Media Logic and Journalistic Independence

Mass Media Logic and Its Factors

The logic of mass media is conceptualized as a form and process of communication in which the media, when offering a message, decide what content to present and how to present it, including framing and constructing the interpretation of the message. These decisions are often globally standardized, meaning that the formats of the broadcast are more or less the same across different media organizations (Altheide, 2013; Hodson & Petersen, 2019). The logic of mass media is largely driven by entertainment, which helps to satisfy a wide audience. In addition, in this approach the media are deemed as having the ability to make the recipients adjust their daily activities to the media schedule and content.

The interest in media logic(s) research has led to the emergence of studies that isolate certain (sub)logics that explain how media organizations and media content change depending on the development of external forces. For example, we can list such logics as audience, commercial, organizational, professional, managerial, and technological ones (Blassing & Esser, 2022; Fincham, 2023). The conclusions drawn from these studies mainly highlight the continued importance of professional logic on one hand (Enli & Simonsen, 2018; Fincham, 2023), and the coexistence of commercial and audience logics on the other (Blassing & Esser, 2022).

The deepened conceptualization of media logic focused on news production offered by Esser and Strömbäck (2014) introduces three principles that explain how this logic works in practice, namely professionalization, commercialization, and media technology (see also Esser, 2013). These principles have a significant impact on how news is produced and distributed, in particular, how technology shapes content and its accessibility, and how commercialization dictates the selection a presentation of messages. Professionalization determines the standards of journalists’ work, as well as the news values that should be upheld to serve the public interest and maintain professional autonomy in deciding what is news. The above principles reflect a type of rationality that is cultivated in mass media institutions (Altheide & Snow, 1979; Kalsnes et al., 2017), but they are particularly applicable for news media organizations. These principles define the essence of the “professional journalistic logic,” where the term “traditional journalistic logic” is also used in a similar sense (Fincham, 2023; Hermida & Mellado, 2020; Sirén-Heikel et al., 2023).

The implications of mass media logic include that audiences are free what to choose and learn about current events, but this choice is limited by the gatekeeping role of professional journalists. Gatekeeping is understood as the process of making decisions about the events and issues that will be covered (Napoli, 2019). In this process, news values (factors) and newsworthiness belong to the most important components of mass media logic applied by journalists (Esser & Strömbäck, 2014).

News as a genre results from the process of information gathering and editorial decisions, and is conditioned by professional norms and factors (Nossek, 2015). News is also defined as a story that meets the conditions determined by news factors (Harcup, 2014). The literature devoted to traditional mass media news factors (MMNF) is rich, and their catalogue, although not uniform, can be identified quite easily by following the most popular categories offered by communication scholars interested in news journalism. We mainly find the factors of importance, elites, frequency and continuity, personalization, unexpectedness, success, negativism, conflict, and simplification (see among others: Allan, 2010; Eilders, 2006; Galtung & Ruge, 1965; Harcup & O’Neill, 2017; Kepplinger, 2015; Schulz, 2003; see the detailed description of MMNF: Nowak-Teter & Łódzki, 2023). At the same time, research on social media news factors has been scarce and focused on shareability, that is the “sharing” activity of users. In other words, social media news factors are determined by the ability of social media news message to be shared (Al-Rawi, 2017; Dwyer & Martin, 2017; Harcup & O’Neill, 2017; Hermida, 2016; Hurcombe et al., 2018; Trilling et al., 2017). The review and comparison of MMNF and SMNF was provided by Ewa Nowak-Teter and Bartłomiej Łódzki (2023). They not only worked out the concise catalogue of these two types of factors, but also juxtaposed the relevant categories. On the basis of this categorization we will develop our analytical framework by extending the analysis to the further categories of social media news factors that go beyond the conditions of shareability.

Social Media Logic and Its Factors

We can encounter opinions that there is no general media logic, but many media logics, depending on the type of media (Hepp, 2020). Other scholars claim that the rationality of social media called “social media logic” is generally expected to replace (theoretically and practically) the logic of mass media. Transforming media logic(s) are seen as concepts that explain the process of mediatization, which also applies to the news media sphere. Thus, social media logic is often defined in contrast to mass media logic in terms of content production, distribution, and media use (also network media logic; Klinger & Svensson, 2014; Dimitrova & Strömbäck, 2010). However, we should also keep in mind that different logics can enter into the reciprocal relationships.

Social media logic consists of the set of rules that determine how social media work, including connectivity, popularity, algorithmization, datafication, and programmability. Therefore, social media logic can be defined as a certain rationality built into social networks. In particular, connectivity promotes interpersonal and intergroup contacts. It describes how connections between users can be initiated and maintained (Bossetta, 2018; van Dijck & Poell, 2013). Friendly connections on social media are monetized, and the boundaries between editorial and marketing activities are blurred (Haim et al., 2021; Hardy, 2021). Connectivity takes the form of “shares” and “comments” on Facebook, enabling redistribution of content and encouraging interaction and discussion between users. By observing patterns of users’ engagement in connections, we can learn how the logic of social media works in practice.

Popularity is about observing how users decide what is popular and how social platforms use popularity data to increase their own popularity. The use of knowledge about popularity metrics, meaning what kind of posts are popular among the audience, is one of the most important features of social media news journalism (Haim et al., 2021; Hodson & Petersen, 2019). The popularity principle has become essential in social media journalism due to the use of metric-driven practices, that is, engagement measured by clicks, reactions, shares, comments, and time spent or return visits (Meijer & Kormelink, 2019). However, the simplest way to measure popularity is to observe the number of reactions.

Algorithmization determines how developers of social media applications prioritize the selection, sequencing, and visibility of posts (Bossetta, 2018). Facebook’s algorithms are constantly changing its digital infrastructure by observing the users’ behavior and testing their responses to the changes in the platform at many different levels (Mackenzie, 2019). These changes impose certain requirements on journalists (Willig, 2021) and are reflected in the editorial logic specified by professional rules. This logic can be replaced or mixed with algorithmic variables to achieve the expected popularity scores for social media news.

Datafication can be seen as a by-product of connectivity, as data are collected by observing how users make connections. Datafication also stands for the quantification of the users’ activities, where their behavior is translated into the language of big data and further analyzed to determine the regularities that guide these behaviors (Bossetta, 2018; Haim et al., 2021; Mayer-Schönberger & Cukier, 2013). Datafication is based on the collection of large amounts of data about the users’ behavior. They are also able to “program” their own spaces on social media in an intuitive way and without any programming skills. The principle of programmability thus involves a certain kind of affordances of the platforms that reflects the technical ability to organize and manipulate the users’ activities in such a way that they make a creative and communicative contribution to the system of the platform. In other words, the technical side of social media allows the users to decide what content to publish and how to publish it (e.g., in terms of visual and textual form, as well as time and place). In addition, the platform registers the activity of users (datafication) and adapts the flow of content and communication to the needs of the targeted user groups. Programmability is linked to popularity, as it facilitates the creation of popular content (Hutchinson, 2019) through machine learning focused on statistical predictions (Mackenzie, 2019). It reflects the dynamics of the power relationship between social media owners, their users and commercial organizations, where users provide data about their activities and platforms give access to these data for the needs of commercial organizations.

In the literature on social media logic, the aforementioned principles are mainly offered as a conceptual framework (Enli & Simonsen, 2018; Gulyas, 2013; Hermida & Mellado, 2020; Walters, 2022; Welbers & Opgenhaffen, 2019). What we would like to offer in this study is the empirical analysis of the application of this logic by journalists, namely, social media editors (SMEs). In particular, we decided to focus on the question of the practical side of the application of both mass media logic and social media logic in news content. We are interested in how journalists combine or mix these two logics to balance newsworthiness with popularity factors (Christin, 2020; Tsuriel et al., 2019) and to reach different audiences (Casero-Ripollés et al., 2016). Consequently, we ask the following research question (RQ1): How are mass media logic and social media logic applied in social media news journalism and what is the relationship between them? To answer this question, we will conceptualize and operationalize social media news and its factors.

Social media are normalized and adapted to the function of journalism (Broersma & Eldridge, 2019). This normalization leads to a possible reorientation of the power dynamics in the relationship between social media and news journalism, accompanied by a clash between the logic of journalism and the logic of social media. News organizations engage and invest significantly in social media, even though they perceive the platforms as “frenemies” (Sehl et al., 2021). This clash includes the news media’s desire to produce content that participates in a chase to maximize user engagement by collecting reactions, shares, and comments, at the expense of quality based on traditional principles. As Mats Ekström and Oscar Westlund (2019) also claim, we are witnessing a redistribution of power from news media to social media companies, and this redistribution concerns, among other things, the transformation of principles, contexts, and professional backgrounds against which the news is presented. The transformation of journalistic principles is characterized by “boundary blurring” and “boundary crossing” strategies, where old and new media logics interact, compete, and co-evolve to respond effectively to the threat of new media logics (Chadwick, 2017, p. 166). As social media logics are strongly commercially driven (Walters, 2022), we are witnessing the pressures that force elite media organizations to compete across all platforms and provide value for money. The strategies adopted by journalistic brands include the aforementioned “blurring” approaches and the fusion of genres and norms (e.g., major breaking stories and a tabloid style of writing; Magin et al., 2021; Chadwick, 2017).

The concept of “social media news” has been distinguished in the literature by analyzing differences between “ordinary news” and “social news” where ordinary news is produced more or less according to mass media news factors (MMNF), and social news is produced using social media news factors (SMNF; Al-Rawi, 2017; Dwyer & Martin, 2017; Hurcombe et al., 2018; Kümpel et al., 2015; Lischka, 2018). Currently, MMNF tend to be replaced or mixed with factors of shareworthiness (also shareability; Klinger & Svensson, 2014; van Dijck & Poell, 2013). SMNF are defined in terms of their ability to elicit user responses, which mainly take the form of “sharing” (Martin & Dwyer, 2019). Consequently, we can define social media news as news stories published on social networks that fulfill the conditions set by the social media news factors. The set of factors forming the ability of the social media news to be published mainly includes visualization, video, emotionalization, celebrity, entertainment, weird stories, and transparent positionality (for an explanation of these categories, see Nowak-Teter & Łódzki, 2023).

However, the conceptualization of SMNF based on shareability may not be sufficient, as it does not include the influence of measures other than shareability, for example, reactions as such (e.g., likes, loves etc.), and comments. For this reason we complement our analytical framework with the measures of “reactions as such” (e.g., likes, reflecting connectivity) and comments (popularity). We will observe the application of the connectivity principle through the measure of both shares and comments, while the popularity principle will be investigated through the measure of reactions as such. In this way, we seek to find out what role these principles play in journalistic decisions. Using these categorizations, we aim to deepen the research devoted to the social media news factors by qualitatively observing their application in journalistic practice on social platforms and assessing the consequences for professional independence.

Social Media Logic and Journalistic Independence

Adherence to professional journalistic norms, including news factors, is one of the key indicators of journalistic independence. Although this independence is sometimes seen as a “utopian” value, it is also seen as something that professional journalism should strive for (Asp, 2014; Bennet, 2014). However, what this independence undoubtedly includes is the freedom to define, shape, and control journalistic work and to take responsibility for independent decisions, both in the individual and organizational dimension (Lauk & Harro-Loi, 2017; Őrnebring, 2016). Professional independence, in the sense of being free and independent in one’s work (Deuze, 2005), is followed and strengthened by the imperative of public service value—to be watchdogs and active collectors and disseminators of information; to be objective—to provide impartial, neutral, and fair message; and to follow the latest news (Deuze, 2005).

Journalistic independence is complex and context-dependent, and can be analyzed at different levels (institutional, organizational, and individual) to determine “from whom” and “from what” journalism is or should be independent, especially considering economic and political forces (Deuze, 2019; Deuze & Witschge, 2018; Poell & van Dijk, 2014; Schudson, 2005). Recent analyses, examining the legal and technical conditions of journalistic independence, have shown how increasing automation challenges editorial decision-making (van Drunen & Fechner, 2023). In particular the new dependencies are observed in relation to third-party technology companies, data brokers, engineers, and IT product owners. What helps to maintain journalistic independence in these conditions is the preservation of journalistic values embedded in the very concept of independence and reflected in the set of traditional news factors. However, we are witnessing a transformation of news values and news factors in line with the evolution stemming from the dominance of traditional media logic to the predominance, or at least a large share of, social media logic. The research conducted among U.S. journalists showed that they still use a “journalistic gut feeling” and strive to preserve traditional journalistic values. At the same time, they are aware that they need to take the audience metrics into account when deciding what to publish on platforms (Walters, 2022). As a result, there is a notable shift toward algorithmic logic, where news value has become a quantifiable concept.

As we can see, the interplay between traditional media logic and social media logic that takes place in news journalism is conditioned by the challenges posed mainly by automation and algorithmization, and has implications for the independent performance of this profession. Social media logic admittedly organizes the professional work of journalists, at least to some extent (Enli & Simonsen, 2018), but this logic is moderated by the media logic of mainstream mass media. The problem of journalists’ adaptation to the social media world mainly concerns the interaction and tension between the structures, principles, and affordances of social media and the established norms and practices of journalism, such as objectivity, impartiality, verification, and gatekeeping (Gulyas, 2013; Hermida & Mellado, 2020; Vu, 2013).

An approach to gatekeeping tasks can be considered as one of the most important conditions that make the difference between the two analyzed logics. In the case of traditional mass media logic, journalists take on the role of professionals who “keep the gate” for news and apply the traditional news factors to select them, by following newsroom routines as well as editorial and organizational constraints (Shoemaker & Reese, 1996). In the case of social media logic, social media editors rely heavily on the techno-commercial and algorithmic mechanisms of platforms (Poell & van Dijk, 2014) in their news decisions. Although in both approaches the commercialization of news according to audience needs plays an important role, in the case of social media logic this commercialization is supported by algorithmic guidelines and economic pressure on news journalism which effectively accelerates the commodification of news (Walters, 2022). This means that the privileged forms of media content are embedded in the specific affordances of social platforms, guided by algorithms that take advantage of big data on consumer behavior (Hermida & Mellado, 2020).

As we can observe, professional journalists are more or less inclined to adapt or “normalize” their journalistic practice to social media (Hedman & Djerf-Pierre, 2013; Hermida, 2013) making this practice more or less subordinated to this logic (Thurman et al., 2019). Given the expansion of social media platforms in journalism, we can assume that social media logic has consequences for journalistic independence. This independence, in the case of social media journalism, may be limited due to the pressures imposed on the editorial boards of news organizations that follow algorithmic measures related to connectivity and popularity goals, but the extent, conditions, and consequences of these limitations have not yet been identified and analyzed in detail.

What is interesting in this respect is how editorial decisions are based on algorithmic indications and whether there is any room for journalistic autonomy and independent decision-making under these conditions. The potential conflict between editorial and click-based approaches to journalistic work arises from the presented differences between editorial, professional logic based on original reporting and offering news as information, and social media logic based on communicating with the audience and producing news as stories (Christin, 2020; Napoli, 2019). In this second approach, journalists may feel constrained by the quantification of their practices. Taking into account these concerns, we formulate the following research question (RQ2): Whether and how does social media logic influence journalistic independence?

Materials and Methods

Given that the characteristics of journalistic practices are best recognized and understood by observing the natural context described by media professionals, we decided to take advantage of in-depth expert interviews (Bogner et al., 2009; Gillham, 2005), as a method of collecting data about their newswork. Expert interviews are a form of qualitative in-depth interview with experts who have access to specialized knowledge; in this case we took advantage of the knowledge of Polish news journalists. This method ensures valuable and nuanced insights, gained through a purposeful conversation that is flexible and interactive (Legard et al., 2003). In particular, we were interested in the views of media practitioners on the role and application of different types of news factors in their daily practice. When collecting and analyzing the qualitative data, we focused on deepening the understanding of participants’ experiences, attitudes, beliefs, and feelings about media logics. We employed a miner’s model, assuming that the interviewee possesses specific information and that the interviewer’s role is to extract it (Babbie, 2021). We were looking for data that reflected both the overt and covert meanings of the knowledge offered by SMEs. At the same time, considering the professional role of the participants, we were oriented toward responsive interviewing, that is, not only asking questions, but also responding to the idea shared by the participant and then asking further questions to understand the views of SMEs (Rubin & Rubin, 2012).

Since the study of the literature on the topic allowed us to identify two sets of news factors, we applied the concept-driven categories representing mass media logics and social media logics (MMNS and SMNF), as well as data-driven categories reflecting bottom-up ideas offered by SMEs (Lindlof & Taylor, 2011; Schreier, 2013). The data-driven categories that we tracked in the analyzed content of the interviews allowed us to discover MMNF and SMNF raised by SMEs, which had not previously been included in the analyzed literature. In addition, the data-driven categories were useful in the process of identifying the explanations, reasons, and causes why SMEs perceive newswork for Facebook as limiting (or not) their journalistic autonomy. Thus, we sought to identify the reasons why they feel (in)dependent in their journalistic practice.

We conducted 24 semi-structured expert interviews (Bogner et al., 2009; Galletta, 2013; Gillham, 2005). The experts represented 12 news media organizations based in Poland. Semi-structured interviews can be used as qualitative (or mixed-methods interviews) that combine open-ended and closed-ended questions. The value of the semi-structured interviews lies in their flexibility, as they are sufficiently structured to address specific problems included in the theory-driven research questions and to leave a space for the meaning of the problem revealed by the participants of the interview. In our case, this type of qualitative interview applied to the group of experts was able to provide unique and insightful knowledge about the essential facets of the research problem.

We considered different types of media organizations to obtain a reflection of both quality press and tabloid press (daily newspapers; four social media editors interviewed), news portals (three), radio stations (three), and TV news channels (two). Each media organization was represented by journalists who identified themselves as playing the role of SMEs in news media organizations. The media organizations are owned by American, German, Polish, or international media groups (only private media were considered), and the SMEs declared that they use internationally known tools and applications in their journalistic work (e.g., Crowd Tangle). As the number of people working with social media in these organizations was not large, we decided to anonymize the interviews by limiting the identification to a number, without revealing the name of the organization or other demographic data.

Recruitment (Hennink & Kaiser, 2021) was initiated via LinkedIn and Twitter, using the keywords search: “social AND media AND editor,” accompanied by the name of the target media organization. A total of 110 requests were sent out and 24 Polish journalists agreed to participate in the interview. According to the interviewees, it was due to the specific working conditions of SMEs (e.g., time pressure and immediacy of their tasks). Telephone and online interviews were conducted in July and August 2021. A typical interview lasted approximately 60 to 90 minutes. The interview protocol was approved by human subjects, and the author obtained informed consent from each interviewee, assuring them of voluntary participation, anonymity, and confidentiality. All interviews were conducted exclusively and personally by the author using Polish as the native language of the interviewees.

The interview questionnaire consisted of three parts. In the first one we asked an open-ended question about news factors, and the interviewees were asked to indicate the news factors they considered important for news on Facebook. The second question in this part has a hybrid nature, as we first presented the SMEs with a list of MMMF and asked them to choose the most important and most often used by them, and then asked them to justify this choice. The same procedure was used for the SMNF. This set of questions was designed to answer the first research question (RQ1) concerning the use of different media logics in journalistic practice. In this part of the interview we also inquired about additional news factors that do not belong to MMNF and SMF and that were not listed in the first question of the interview.

To deepen the investigation devoted to media logics operationalized by two sets of news factors in the second part of the interview, we prepared the questions on the relationship between news factors and social media news factors as well as between the popularity and connectivity measures (RQ1). The third part of the interview was devoted to the conditions of journalistic independence and the limitations that the social media journalist faces when creating news for Facebook. Here, we asked an open-ended question to find out whether they felt restricted in their independence as journalists when creating material for Facebook, and if so, why. This question was designed to answer the second research question (RQ2): Whether and how does social media logic influence journalistic independence?

Findings

The empirical data come from Poland, where the media system is characterized as a hybrid politicized system (Dobek-Ostrowska, 2019), as it shares certain common features with Hallin and Mancini’s (2012) polarized pluralism media system model. Thus, our research data may reflect the characteristics of an important group of European countries. Poland is located at the medium level of politicization among CEE countries (Dobek-Ostrowska, 2019), but political intervention is exercised mainly in public broadcasting media, while private media is an independent sector. Although the Polish media system has some peculiarities characteristic of CEE countries (as the politicization of media can be deeper than in Western countries) the general characteristics place this country in the area of the mentioned polarized pluralism with regard to European media systems. Poland as a research case is characterized by a high level of internet and social media use among the general population, which means that the journalistic field can be highly saturated with new media technologies. In 2021 (at the time of conducting the interviews) in the European Union on average 90.21% of households had a broadband internet connection, while in Poland the same connection was available to 91.72 of households (Eurostat, 2021b). Similarly, at the time of the survey, 57% of Europeans and 55% of Poles participated in social media (Eurostat, 2021a).

To answer the first research question (RQ1)—How are mass media logic and social media logic applied in social media news journalism and what is the relationship between them?—we asked the SMEs for general thoughts about the news factors they find useful and important in the process of news production. In answering this open question, the SMEs mentioned, among others, the following characteristics of news materials:

It must relate to the current situation, and such a text should be in the trends of the day, i.e. what people want to read, what they are looking for. (SME23)

A topic that is being discussed, one that most people have an opinion about, and one that evokes emotion. (SME24)

Therefore, we can observe both the factors that belong to traditional news factors, when the interviewees mentioned the current situation, and a topic that is discussed (frequency and continuity), as well as the factors that belong to the category of social media news factors in case of a topic that evokes emotions (emotionality). However, SMEs also stated that they consider whether a text is in line with “the trends of the day” and whether the material contains the content that people want to read and are looking for, and therefore will react, share or comment on it. This means that social media workers know this from their practice and have an intuition about these trends. They also explained the use of automated tools (e.g., Crowd Tangle) to find out what kind of content should be offered on the Facebook profile. Looking for a more precise explanation of what kind of news is suitable to be published on social media, we received a response in which SME (3) described the news production strategy as follows:

To a certain extent, social media people are able to select and prepare material that will interest readers and make them click on the link. However, it is a lottery every time, because the viral potential of a material does not only depend on the subject matter or the “display” (a title specifically prepared for a social network). The click-through rate of a post also depends on the time of posting, the initial interactions, and the page or group on which the post was posted. For this reason, the rule of thumb for posting messages and texts on social media is simple: post as much as you can, because you never know what you might get (SME3).

Here, we can see that the journalistic strategies go beyond simply fitting the message to certain news factors. Instead, SMEs also have to follow the non-transparent algorithmic rules of popularity, which makes their work difficult and unpredictable (a lottery). To sum up the results of the first open question of the interview, we can say that SMEs do not only mix mass-media and social-media logics and find useful both traditional and social-media news factors, but they also look more broadly at the relevant timing and the “surplus strategy” (insert as many posts as possible).

We then offered the interviewees the examples of news factors according to the categories developed on the basis of the literature on the subject. Using the second interview question dedicated to traditional news factors, we found out that the news factor of importance belonged to the most essential category that was claimed to be the formative factor for the news-making activity:

In my experience, it follows that such news are the most “clickable” and the most engaging—they trigger comments, discussions, and shares, they make people think. They also fulfill the informative function. (SME3)

We want to be a media organization that explains the world to people; we try to present the important issues as simply as possible and make users aware of these issues. (SME8)

I choose the important issues that are of interest to our target audience. (SME13)

The interviewees explained the use of the importance factor by its impact on large or influential social groups. They also claimed that the strategy of combining the societal importance with the attractiveness and controversy of the topic is able to evoke strong negative emotions and is most effective in terms of click-through rates. However, we must keep in mind that the SMEs faced with the questions concerning journalistic values and confronted with social media logic might make some attempts to rhetorically frame social media logic as compatible with the journalistic logic. In this way, they might try to justify their social media practices.

The SMEs also shared their “experience-informed” opinion that covering controversial topics, including scandals, can produce similar results:

Scandals are the best, e.g., vaccination scandals or the death of famous people or celebrities, even the death of a terminally ill boy, especially when they are followed by events that evoke affection or indignation. (SME12)

Extreme emotions undoubtedly increase reach on Facebook. (SME21)

We are expected to write texts for Google algorithms and SEO (search engine optimization). (SME24)

In the next part of the interview, we asked about the usability of the SMNF in preparing social media news. Among the SMNF, we found the factor of emotions and strong negative emotions as the categories that were chosen as most effective in journalistic work, when Facebook users are the target of the news:

Emotions, both positive and negative, because they engage users. (SME3)

Events with a high emotional charge and political statements with extreme opinions. (SME4)

Unfortunately, I have to say that this is a battle, death, suffering . . . (SME7)

Among the reasons for the use of these factors, the SMEs interviewed explained that there is an expectation from the heads of the editorial departments that the editors will offer sad or tragic news, often motivated by the goal of ensuring high traffic to the Facebook page. However, we also encountered opinions that the work of SMEs is not only determined by Facebook ratings, but also by a certain professional intuition. This intuition often tells them which story is worth posting. There were also SMEs who were aware of their social roles and responsibilities:

The news media play an important social role; we are gatekeepers, and we set the agenda. (SME15)

However, other attitudes reflect the audience-centered approach that comes from the belief that they are serving the needs and interests of their audience:

On social networks, we often provide news about public figures and celebrities because people are interested in their lives. (SME14)

We also noticed that the interviewees tended to mix certain mass media and social media news factors, especially when they mentioned controversies as important features of their materials. At the same time, they also pointed out the strong negative emotions as highly effective news factors. In particular, when they listed the controversial topics of corruption and immoral behavior (SME12), controversies concerning politicians (SME10), and political scandals (SME8; SME9), they also identified fear or threat to ordinary people (SME8; SME10; SME12) as important news factors.

In addition, in the interviews the visualization factor tended to be mixed with frequency and continuity (e.g., follow-up video materials; SME12) or personalization:

When something bad, unfair happens we like to present a human story. (SME8)

Among the visual materials the SMEs consider live streaming to be the most engaging. When asked to justify their use of visualization, they openly admit that their decisions are algorithmically driven, and that they take into account not only Facebook’s algorithm (SME15), but also Google’s preferences:

We use visualization because it is the most engaging, taking into account the algorithms. (SME22)

In the second part of the interview we strived to deepen the understanding of the relationship between MML and SML (RQ1). To this end we combined the concept of news factors (MMF and SMNF) with the principles of connectivity (shares and comments) and popularity (reactions) to observe the strategies used by the SMEs to achieve satisfactory scores for their news messages. For example, when asked about news factors that can ensure connectivity (shares and comments), SMEs identified controversial messages as the most promising. In particular, anti-vaccination statements and supporters of conspiracy theories proved to be powerful in generating comments (SME21). The latest COVID-19 news and other trending topics had the potential to generate a large number of reactions. SMEs were well informed about the typical topics and forms (visualization) of messages of the most connective stories:

Video is the most shared, e.g., winning a gold medal at the Olympics, which was shared for patriotic reasons, but also the so-called “social video,” e.g., when bears fight. (SME17)

Controversies related to politicians and “non-obvious” issues that people tend to disagree on are also identified as a general category of news that lends itself to Facebook in terms of generating shares and comments. Among the most popular stories, the SMEs cited breaking news dealing with dramatic stories (e.g., planting a bomb; SME15; death notices; SME12), pandemic restrictions, COVID-19 statistics, and also some “games” with numbers and statistics. In doing so, they noticed the need to maintain some balance between bad news and good news:

It’s not like that on Facebook there should only be good, positive news, but also not too much bad news either. (SME18)

To achieve popularity goals the SMEs explained that it is important to offer the human interest stories:

The hero must always be a human being with whom users can easily identify and sympathize. They also like to judge someone’s behavior. (SME 16)

These human stories can be positive, successful, and break down barriers, or they can encourage a certain amount of struggle. (SME14)

They also emphasized the importance of the formal and organizational side of news production. Popularity goals are achieved through the use of algorithmic tips, for example:

We watch Facebook traffic sources, we use the option of Facebook suggested post options, we use Google Trends, etc. (SME8)

At the same time, interviewees noted that there is often a high degree of uncertainty in producing news for Facebook. Although they recognize certain algorithmic rules that can be exploited in news production, there is always a certain degree of risk and surprise, when some stories “grow” much faster than similar stories with the same form or topic. For this reason, the sensible strategy is to publish as many posts as possible, while maintaining the suitable quality of the message. Here, “high quality content” is understood both as having traditional news value and as being appropriate in terms of algorithmic tips (SME19).

We also elicited their views regarding the issue of professional independence involved in Facebook news production (RQ2). In this regard, we found that the awareness of the potential conflict between the Facebook algorithms’ recommendations about the news and their impact on independent journalistic decisions is quite limited. When asked about this conflict (do social media limit your journalistic independence?), SME answers included both denial and justification:

It is a bit as if these messages had to be selected in order to predict people’s reactions. (SME6)

No, publishing on Facebook just has its own rules [. . .]. You can write big, high-quality text and then “twist” it to make it clickable. (SME2)

No, because writing news for social media is very important [. . .] most of the time I choose topics that I know people want to read about. That topic can fit in with trends, and the users will react accordingly, and that way the media brand grows stronger. (SME20)

Some SMEs openly admitted that they try to mix different objectives and news factors, not wanting to reject traditional values, but admitting that they have to follow popularity measures:

No, Facebook does not limit my journalistic independence. Why should it? We use algorithms to prepare news and we look for the topic that will “heat up” the audience, while at the same time we have a mission to explain the world to the audience. Well, I think we have to “reconcile water and fire.” (SME8)

Yes, because I would reject certain subjects if I had more freedom; now I have to write about something everyone knows and likes. (SME15)

Yes, because in big media companies, SMEs select news for social media based on algorithms, and professional journalists would often not agree to publish this news, for example, because of its cruelty. (SME 14)

In other cases, the restrictions of working for Facebook are seen as an unwelcome but necessary part of professional journalism:

Yes, Facebook limits me. What I post on Facebook is not always interesting, but I know it will be controversial. (SME7)

Overall, social media editors are often unwilling to openly admit certain limitations, as this would probably force them to face and resolve the internal conflict between professional responsibility and the need to practice news journalism in the real conditions that they experience in their daily work. It can also be the case that SMEs assess their autonomy in terms of the third-person effect, in the sense that they notice some limitations caused by the social media logic in journalism in general, but they are convinced that they personally do their best to offer quality news. Consequently, what they do is to offer the social media news that contains a combination or a mixture of news values and algorithmic principles.

Discussion and Conclusion

One of the main insights we can offer in the first step is the conclusion that social media logic is applied in social media news journalism, but the logic of mass media is still “in the game.” The picture of the relationship between different media logics and their factors, as well as the consequences for journalistic independence, based on the opinions expressed by the investigated journalists, is presented in Figure 1. This relationship can be characterized as both “friendly” (when MML and SML are mixed) and tense, since the reasons for applying SML may be different from those for applying MML (e.g., driven by effective communication and popularity goals versus professional and editorial ambitions). This “frenemy” relationship leads to palpable tensions regarding the independence of journalistic practices (RQ2).

OPEN IN VIEWER

Figure 1.

Proposed Relationship Between Different Media Logics and Their Factors, as Well as Their Meaning in Journalistic Independence.

In particular, the mentioned mixture of logics influences media practice and journalistic professional independence by encouraging SMEs to use news factors that can evoke the highest audience engagement (see Figure 1). Importance and strong negative emotions emerged as the concept-driven categories reflecting the most important mass media and social media news factors (MMNF and SMNF). These are the features of breaking news that build community activation, which is in line with previous findings (Carlsson & Nilsson, 2015; Hermida, 2016). The data-driven categories of news factors that we discovered in our data show that the SMEs feel determined to offer the news that follow the trends of the day, and that they apply a “surplus strategy.” They feel obliged to prepare and publish as many social media news stories as possible while striving to maintain good quality.

Looking for an explanation of the structure of the social media news messages in terms of different media logics, we found that the interviewed SMEs generally perceived the MMNF of importance as one of the most significant news factors, based on the assumed impact of the message on numerous or influential social groups. However, they also tended to mix MMNFs with SMNFs. In particular, they considered mixing the MMNF of controversy with the SMNF of strong negative emotions to be particularly effective in terms of gaining online traffic. We can therefore conclude that the algorithmic principles of social media, to some extent, dictate the way in which SMEs practice their news work.

The reasons for the mixing the two logics include editorial ambitions and aspirations for original reporting on the one hand, and pressure to communicate effectively with audiences and provide them with metrics-driven, engaging stories on the other (Christin, 2020; Napoli, 2019; Poell & van Dijk, 2014). At the same time, however, we found that the gatekeeping principles of media logic still form a basis for news journalism, and that the social importance of the issue can determine the message.

SMEs tend to put pressure on generating lively discussion, probably also following the algorithmic requirements in this respect. As we learned from the SME interviews, publishing important news is based on the perception that it is essential for people, and posting emotional news is related to editorial goals of engaging the users. These two functions of social media news, that is, satisfying information and entertainment needs, are aptly combined in the hybrid social media content. The view that many different media logics are separate and valid at the same time (Hepp, 2020) may be less well-founded than the idea of the conjunction of media logics. Rather than being replaced, mass media logic is merged with social media logic in different configurations. This combination of different logics can be explained by “an imposed partnership” (the “frenemies”) between news journalism and social media platforms (Sehl et al., 2021) caused by the essential growth of the social and economic power of the latter. However, we also noticed that there are certain news factors of social media logic (e.g., emotionality and strong negative emotions), that can be considered as those factors that help the SMEs to present important and controversial news (mass media logic news factors), in the sense that these factors reinforce each other.

Our analysis also followed the relationship between the MMNF and SMNF and the principles of connectivity and popularity. The participants shared that to ensure connectivity (shares and comments) they perceived controversial messages and trending topics as the most promising. This means that SMEs have learned, to a certain degree, how to initiate and maintain connections between users and between the users and the media organization (Bossetta, 2018; Hardy, 2021; van Dijck & Poell, 2013). Popularity goals are achieved through the publication of breaking news and dramatic stories, as well as through the use of algorithmic tips. Consequently, we can add further evidence that metric-driven journalistic practices, based on the knowledge of popularity measures, are the hallmark of social media news journalism (Haim et al., 2021; Hodson & Petersen, 2019). At the same time, we found that the stated strategy of publishing as many posts as possible and maintaining the relevant quality of the message proved to be most preferable. This is related to the issue of autonomy, which can be considered more essential for SML than for MML, since journalists do not exactly know the details of the constantly changing algorithms. Therefore, it is more difficult for them (compared to the case of mass media) to defend their independence in these uncertain conditions created by the logic of social media. This uncertainty is rooted in the constantly changing algorithmic requirements for news-making (Walters, 2022). At the same time, it is not the case that social media logic exclusively promotes the (extreme) commercialization of news production. Instead, we claim that the processes of algorithmization and datafication form favorable conditions to strengthen and accelerate the commercialization of news. What is more, it takes place in highly uncertain conditions, which makes it much more difficult to maintain journalistic independence than in the era of mass media logic.

In line with our initial assumption, we were able to observe the conflict between professional independence and the need to practice news journalism within the constraints imposed by Facebook. However, the SMEs were sometimes reluctant to admit that this conflict exists. When faced with the situation of having to choose between traditional values or popularity measures, which may limit their journalistic independence, they tended to formulate certain justifications (publishing on Facebook simply has its own rules), and to offer a mix of news values and algorithmic factors. The need to write popular news is thus linked to certain organizational tensions (e.g., they would refuse to cover certain topics if they had more freedom). Consequently, their strategy is to reconcile water with fire and mix different logics in their journalistic practice.