How Loyalty Works: Why Do People Continue Their Relationship with Journalism?

ABSTRACT While loyalty has become increasingly relevant for journalism, it has rarely been examined as a single phenomenon and even less from an audience perspective. The empirical knowledge surrounding loyalty has centred on its role in trust, news choice, and the sustainability of the profession. Nonetheless, we still do not know how a relationship with journalism unfolds and becomes meaningful for news users. This research aims to advance an in-depth understanding of the dynamics of loyalty with journalism. To this end, we conducted 35 semi-structured in-depth interviews with regular news users (18–65 years old). By exploring the language, feelings, emotions, and stories that people use to describe their relationship with journalism, this article offers a holistic and comprehensive examination of what loyalty means as part of audiences’ everyday news consumption. In doing so, we aim to expand journalism’s vocabulary about loyalty, adding width and depth to our understanding of the potential benefits and implications that loyalty entails for the profession.


Introduction
Developing and maintaining a loyal audience appears to be essential both today and for journalism to have a future (Wenzel and Nelson 2020). Within this landscape of abundant news offerings, loyal audiences provide sustained attention that translates into highly profitable measures of engagement, such as clicks, time spent and shares (Krebs and Lischka 2019;Hansen and Goligoski 2018). Likewise, loyal news users appear to be more likely to provide financial support to news organizations either through direct donations or membership programmes (Price 2020;Ferrucci and Nelson 2019;Batsell 2015). Furthermore, audience members who are loyal to a news brand appear to rely more on the news it provides and therefore make use of it as a means of navigating life and participating in society (Broersma 2019;Nelson and Kim 2021). However, despite the growing relevance of loyalty in journalistic practice and journalism studies, it has rarely been examined as a single phenomenon and even less from an audience perspective.
Empirical knowledge around loyalty has tended to focus on media trust and news brand attachment. While such research has provided relevant data on the role of loyalty in both trust and news selection, studies took a production-orientation perspective, overlooking how loyalty works from the point of view of news users: how does a relationship with journalism unfolds and becomes meaningful for them. Moreover, studies relating loyalty to trust have often hypothesized that audience loyalty to journalism depends highly on news credibility. Yet, the relationship between trust and news preferences has proven not to be as straightforward as investigations have often suggested (cf. Tsfati and Cappella 2003), and although trust and credibility are important antecedents of news preferences, it seems that they are not the only forces behind users' loyalty to a given news media.
Studies linking loyalty to brand attachment, on the other hand, have tended to focus on quantification and profitability of user attention (Napoli 2011;Webster 2014). While research has implicitly shown that emotions, interaction, and participation are relevant aspects of loyalty, this perspective has tended to emphasize the profitable aspects of loyalty and ignored those elements that may not be economically rewarding but may still bring about implications for journalism. Also, a common approach in these studies has been to operationalize loyalty as a unit of measurement or to examine it through self-reported surveys with predetermined categories based on production concerns. Such narrow operationalization of loyalty has led to useful, but partial understanding of the kinds of practices, feelings, and sensations that this construct conveys among news users.
The present research aims to advance our understanding of loyalty by developing an in-depth understanding of how loyalty with journalism works from the perspective of its users. To this end, we conducted 35 semi-structured in-depth interviews with regular news users (18-65) in Chile. They therefore provided "an expert insight" (Lindlof 1995) into journalism use. By exploring the language, feelings, emotions, and stories that people use to describe their relationship with journalism, this article sets out to advance our vocabulary of loyalty based on the everyday relationship audiences maintain with journalism, bringing more breadth and depth to our knowledge of the potential benefits and implications of audience loyalty for the profession.
In the next section of this article, we describe the different approaches in which loyalty has been studied over the years and outline those areas that have been insufficiently explored. Then, we explain the methodology employed to conduct our research and, in section four, we present the main results. Finally, in the conclusions, we analyse the implications that the findings of our research have for journalism studies and practice.

Loyalty Conceptualised in Relation to Trust
The relationship between audience loyalty and trust in news has been an area of growing interest in recent years (e.g., Nelson and Kim 2021;Goyanes 2020;Lischka and Messerli 2016). This body of research has focused on examining the ways in which trust functions as an antecedent of loyalty. Most of these studies have confirmed earlier findings showing that users' satisfaction and trust determine their loyalty to the news (Horppu et al. 2008;Laroche, Habibi and Richard 2013;Lischka and Messerli 2016). However, scholarly conceptualizations of loyalty in relation to trust place particular emphasis on the latter and may overlook how loyalty can be shaped by other influences. For example, Nelson and Kim (2021) studied how news audiences' trust in one news platform leads to loyalty to other news platforms. But, while studies corroborated the existence of a clear and positive link between trust and audience loyalty, they failed to provide insights into how loyalty works as a single phenomenon.

Loyalty Conceptualised in Relation to Brand Attachment
Another relevant strand of research on loyalty is the scholarship linking the notion to brand attachment. From a business perspective, loyalty has been defined as "a deeply held commitment to rebuy or repatronize a preferred product/service consistently in the future, thereby causing repetitive same-brand or same-brand-set purchasing" (Oliver 1999, 34). By its focus on describing the strength of consumer-brand relationship, journalism scholarship has traditionally studied brand loyalty with news outlets to gain a better understanding of how news organizations can "achieve long-term business success" and "stay competitive" (Chan-Olmsted and Kim 2001, 85). In other words, how to make casual visitors become more regular consumers. In this sense, brand attachment studies have provided significant knowledge about what makes users pay attention to the news. For example, recent research showed that higher brand loyalty is influenced by users' emotional engagement with news content (Sashi 2012), that the more users share, rate and like news articles online, the stronger they perceive brand loyalty of the news outlet (Krebs and Lischka 2019), and that users' interaction with news sites increases their brand loyalty with particular news organisations (Krebs and Lischka 2019;Lim et al. 2015;Lin et al. 2018).
However, the limitation of understanding loyalty from a brand perspective is that it often focuses on the news industry solely in terms of quantifying and selling users' attention to advertisers (Napoli 2011;Webster 2014). This business perspective seems to overlook those aspects of loyalty that may not be (directly) economically profitable, as it fails to consider those "problems" and "solutions" related to the loyal use of news that go beyond its economic value for journalism as an industry (cf. Costera Meijer and Groot Kormelink 2021). This constraint yields, therefore, a limited understanding of the kinds of attitudes or emotions that constitute loyalty over and above measures and indicators of profitable news attention.

Loyalty Operationalization Through Quantified Measures and Self-reported Scales
Generally, scholarship on loyalty has relied on quantified measures of news attention or self-report scales of brand attachment. The quantified operationalization of loyalty has a long history in journalism studies and practice. Research on television viewing behaviour in the 1980s and 1990s operationalized audience loyalty as viewing time (Barnes 1990;Barwise and Ehrenberg 1984) or repeated viewing (Ehrenberg and Wakshlag 1987;Webster and Wang 1992). Currently, this approach to loyalty has morphed into metrics of "regularity" i.e., the process of measuring people who return again and again to perform some action or use an online journalistic product. For instance, Neilson and Gibson's (2022) study on the relationship social media editors maintains with marketing and editorial departments illustrates the persistence of a quantified operationalization of loyalty in journalism practice. The research found that newsrooms metrics such as number of page views per user, time spent viewing content and return visits are treated as signs of loyalty since they are "believed to be associated with a higher likelihood that a casual reader will become a paying subscriber" (567). Such operationalization is consistent with Napoli's definition of loyalty as a "dimension of audience behaviour that can be captured by exposure-based data". According to Napoli (2011), "audiences loyal to a given television or radio program are usually defined in terms of the number of times they watch or listen to that program: the frequency of their exposure" (90).
However, a growing body of research has shown that sustained attention to journalism involves emotional and material dimensions, which metrics can only partially capture Groot Kormelink 2021, 2015;Groot Kormelink and Costera Meijer 2020). Thus, operationalizations of loyalty as unit of measurement risks oversimplifying how loyalty can be measured and what it actually means for news users.
In addition to the above, even when a more audience-oriented approach has been used, the dominant method is the closed-ended survey. For example, Goyanes (2020) study on loyalty as an antecedent of trust, measured loyalty by asking participants about their adherence to statements such as "I tend to go to the same news source(s) when I get my news" or "I consider myself loyal to the news source(s) I get my news from". While this approach offered insights into a somewhat wider range of loyalty motivations than those inferred through the quantification of clicks, it failed to capture those ways of experiencing loyalty that individuals themselves may not recognise on a conscious level. Research on news usage has shown that individuals pay attention to news in ways of which they are not always aware (Costera Meijer and Groot Kormelink 2018. Therefore, a more in-depth exploration of loyalty through the perspective of news users can help to enable the identification of the kinds of practices, feelings, or sensations that loyalty entails, even the ones that may be unrecognisable by news organizations.

Loyalty Approached Through the Concept of Parasocial Interaction
A fourth approach to loyalty has been taken in terms of parasocial interaction. This body of research proposed a relationship that users have with on-screen characters to which viewers can respond in the same way as in a typical social relationship; what Horton and Richard Wohl (1956) call "intimacy at a distance". Research showed that, during the television viewing process, audience members gain knowledge about the news character, feel good about the way broadcasters "help" them meet their needs, and relate their credibility to -among other things -their perceived sociability (Perse 1990). Similar to these findings are those linking parasocial interaction with attitudinal similarity, attraction and identification with a TV personality, as elements that could lead to loyalty (Eyal and Rubin 2003). Likewise, studies on parasocial interaction in online communities have pointed out that the valence of a website's content (Hoerner 1999) and/or the design and presentation of textual and graphical materials (Kim and Kim 2017) can make online users feel that they are communicating with a person rather than a machine (Choi et al. 2019). However, while in journalism studies this framework explained the ways in which people experience a relationship with television journalists or anchors, it overlooked the relationships that users might develop with non-human entities such as news brands, news sites or news mediums/devices.
Interestingly, marketing scholars have made important theoretical and conceptual advances in this regard. Fournier's (1998) empirical work on consumer-brand relationship showed that people develop relationships with brands and/or objects in a humanized and personalized way. According to Fournier (1998), the depth, quality and strength of the relationship consumers develop with brands can be measured in terms of six facets: love and passion (the attachment and affection that the consumer has for a specific brand), self-connection (the degree to which the brand delivers on important identity concerns, tasks, or themes), commitment (the consumer's intention to behave in a manner supportive of relationship longevity), interdependence (frequent brand interaction and consumption rituals), intimacy (psychological closeness between the brand and the consumer) and brand partner quality (consumer's evaluation of the brand's performance in its partnership role). Such insights have served marketing scholars to uncover the ways in which consumers attach value to and construct meaning from their relationship with inanimate branded objects, such as cleaning products or packaged goods. However, given that news and journalism are real living entities with which users can be in permanent interaction via platforms such as Twitter or Instagram, it seems relevant to examine how loyalty works in the context of a two-way relationship.
Finally, loyalty has gained importance in the last decade as a potential driver of the sustainability of the profession (Hansen and Goligowski 2018). Various studies have addressed how the relationship between users and journalism can yield into subscriptions and provided insights into loyalty by addressing how marketing strategies serve to engage news users (Kim, et al. 2022;Neilson and Gibson 2022). For instance, Nechushtai and Zalmanson (2021) analysed how widely read newspapers in the U.S. frame their value proposition to get audience subscriptions. Rather than focusing purely on an informationbased transaction, subscriptions were often promoted and framed as a relationship with social and normative dimensions such as "sharing content with friends, strengthening ties with the local community and attending exclusive events" (2047). These findings resemble those of Ross, Lester, and Konkes's (2021) case study in a local newsroom, which identified that fostering reader loyalty by merely promoting the media outlet as a community advocate was not enough to drive subscriptions. Following Fournier (1998), Ross, Lester, and Konkes (2021) asserted that readers need to feel that the news organisation is a partner in an interdependent relationship in which they receive something they value in return for engaging with the brand. The research suggests, therefore, that news publications should be conceptualised as partners in a relationship, an approach that, while common in the marketing literature (Fournier 1998), has been underutilised in communication research. Although these findings are relevant to the profession, as they illustrate the ways in which marketing can help news organisations increase their audience revenues., they are insufficient to shed light on how loyalty works as a single phenomenon.

Method
We conducted 35 semi-structured interviews with regular news users (aged 18-65) from two of Chile's largest cities: Santiago and Concepción (see Appendix). Participants were selected through snowball sampling (Atkinson and Flint 2001), initially recruiting people attending the self-convened citizen councils that were regularly held throughout the country at the time of fieldwork. We selected only those individuals who recognized themselves as regular users of newsthose who had an "expert view" (Lindlof 1995) -and were willing to describe their motivations and interest in journalism. The selection criteria of "expert view" ( (Lindlof 1995) allowed us to better capture how loyalty works in journalism as interviewees complied with "the special nature of what they have experienced", resulting in a "special articulation: words that can only be expressed by someone who has been there" (Lindlof 1995, 171). The fieldwork was concluded once we reached the point of saturation in the variety of responses.
In-depth interviews lasted between one and a half and two hours and were conducted in Chile during the months of December 2019 and January 2020 before the corona crisis settled in. This period was beneficial to the research because at the time, the country had just regained some sense of normalcy after the social unrest of October 2019, when citizens publicly manifested their discontent with the social inequalities engendered by the economic model imposed by Augusto Pinochet (Garcés 2019). Although the present research did not focus on this issue, the socio-political context appeared to have stimulated and expanded people's reflexivity on the place of journalism in their lives. The Chilean media system is characterized by its dominantly commercial and concentrated ownership (Hallin and Mellado 2018) its historical link to the Chilean political right (Gronemeyer and Porath 2015) and its predominantly online audience. In Chile, the online population represents more than three-quarters (77%) of the total population, and 82% of internet users use their smartphone for getting news (Fernández 2018). Interviews were recorded and transcribed in their original language (Spanish) and later translated into English.
The study was presented to participants as aiming to examine their relationship with journalism. Interview guidelines were composed of three sections, which we called 'stages. The first stage consisted of a day-in-the-life interview where participants were asked to describe their use of journalism throughout the week. This allowed us to identify the news media or platforms that participants preferred or used most frequently. Given that this research sought to examine what loyalty with journalism means from the perspective of its users, the second stage of the interview focused on understanding what participants were referring to when they talked about "news stories", "news media" and "journalism". An important consideration of the interviews was avoiding imposing definitions or labels on what constitutes journalism and what does not. Therefore, a common question asked was: "when you talk about journalism or news, what are you talking about specifically? How do you make the distinction between what is news and what isn't?" The distinctions each participant made allowed us to leave aside those instances in which participants reflected on other types of content they use, such as documentaries or television series, and to concentrate our analysis on what was regarded by them as journalism.
To initiate participants' reflection on their relationship with journalism in terms of loyalty, in the third stage of the interview we asked them to reflect upon their bonds in terms of human connections. Since loyalty is at the core of relationships (Fournier 1998), we invited participants to think of their relationship with the journalism/news they use as if they were real relationships. This methodological approach was inspired by Wayne Booth (1988) use of the metaphor of friendship to analyse the kind of relationships readers are looking for in texts. The question to participants was posed as follows: if the news media were friends or any other type of relationship, what kind of friends would they be? Since we were not interested in imposing a priori relationship categories, there was no pressure for participants to answer this question accordingly. Consequently, most participants proposed their own types of relationships. One of them, for example, said that he could not regard any of the news media he uses as a friend, but that he could regard it as a classmate. Others made the same point but proposed the metaphor of family relationships, whilst a few of them combined friends and family dynamics. Overall, this stage of the interview aimed to allow participants to freely talk about the relationships they have with the varied array of news sources they use, and what these mean for them, including talking about how these relationships have evolved over time and what are the relevant aspects of them.
To recognize the connections, intensities, multiplicities, and transformations involved in their relationships with journalism, the fourth and final stage of the interview consisted of the think-aloud protocol (Ericsson and Simon 1993). We conducted the interviews in the environment where they interact the most frequently with journalism: their houses. We asked them to use their cell phones, listen to the radio or watch television and to verbalise their actions while doing this. This exercise encouraged them to comment on elements that they appreciate or not in their relationships, bringing out the processes and actions that are part of their loyalty to particular news programs or news brands. For example, when Pablo sat down on his sofa to use his tablet, he immediately described how important it was for him to have a moment alone with the news at home, something not mentioned in the previous stages of the interview.
Since this research set out to examine how loyalty works from the perspective of users themselves, rather than testing how their relationship with journalism were created and what affects them over time, we did not centre our analysis on variables such as users' demographics or frequency of use of a particular device, news format or genre. Instead, the analysis of the interviews was conducted using relational metaphors as sensitizing concepts. According to Bowen (2006), sensitizing concepts "draw attention to important features of social interaction and provide guidelines for research in specific settings" (14). Thus, we used relational metaphors as an "interpretive device and as a starting point for a qualitative study" following an inductive approach (Strauss and Corbin 1967) that allowed the emergence of types of relationships individuals' have with journalism, and a detailed understanding of how loyalty works in each relationship dynamic.
We concluded the analysis by conducting axial coding with the purpose of organizing and reassembling participants' responses into labels based on the variety and depth of the descriptions they made. For reasons of confidentiality and anonymity, all names of participants were replaced by pseudonyms.

Results
The first finding of our research is that loyalty can be maintained with different journalism entitiesincluding TV news programmes, journalists, news brands, news formats, and news platforms. While some participants were loyal to the YouTube account of a specific news TV channel; others were loyal to a journalist who hosts both a news TV and to the written format, both in paper and on screen.
A second finding of our research is that loyalty is an absolute value in the relationships users have with journalism. As we will show in this section, participants hold different types of relationships with journalism, each of which seems to play a different role in their lives. Yet, while loyalty can be a process that is forged and developed over time, from a user perspective, loyalty cannot be scaled or quantified, and while it may require time to develop, there is no grey area once it has been attained; either someone is loyal or is not.
For providing a clearer explanation of how participants make sense of their relationships with journalism, we will employ the actual words and labels participants themselves used during the interviews. We identified four main types of relationship dynamics through which participants made sense of their connection with journalism. The first relationship dynamic that we found is that of good friends, in which users hold a close and intimate relationship with journalism. The second dynamic is that of party friends, characterized by a more circumstantial linkage. The third dynamic is members of a family, either father or uncles and aunts, which is characterized by a sense of familiarity and tradition. The fourth dynamic is classmates, marked by a utilitarian and habitual use of journalism.
These relationship dynamics are in themselves mutually exclusive but can account for different positions and roles journalism has in people's lives. For instance, the same participant who classified his relationship with a news outlet in terms of friendship, also compared his relationship with news platforms to a classmate he had at school. As such, the "relationship dynamics" that participants described capture and reflect the particular ways in which their relationship with journalism unfolds and becomes meaningful (or not).
The analysis of our interviews allowed us to observe that loyalty is present in three of these four dynamics. Below, we will show how each relationship dynamic illustrates a particular way in which loyalty works and will provide details of the complex interplay of connections that constitute loyalty. The implications that these relationship dynamics may hold for the profession will be discussed in the conclusions of this paper.
Good Friends: "I'm Loyal to Them and They are Loyal to Me" The first relationship dynamic we identified was that of "good friends". Within this dynamic, we found a strong sense of loyalty to journalism entities such as news channels, television/radio journalists or news radio stations, that is characterised by, first, an active preference for these entities, second, a liking for the news brand or journalists over the actual content, and third, a sense of assurance of the continued presence of them. These three characteristics are illustrated in the way Pablo, a 29-year-old psychologist, described his relationship with Canal 13's YouTube channel.
I could say that CANAL 13 news is like my best friend (…) … it's the one I watch the most regardless of whether I like what I see or not … if something happens, I turn on CANAL 13 news … always … I'm not going to choose another one … so yeah. It's a loyal friend because it's always there … I can always see it … and it informs me the way I like it, in a brief and precise way, with information that is relevant, that interests me. (Pablo,29 1 ) Pablo also said that Canal 13 is like a friend "they are loyal to me and I'm loyal to them". He acknowledged that while he was constantly checking the news on Instagram via his smartphone, he preferred to get news from Channel 13's YouTube channel via his iPad because "the smartphone gets too hot when I hold it for too long". When holding his iPad, he mentioned that he mainly uses YouTube because he is a "very visual person and is not a good reader" and because "the sound quality of the iPad is much better". This finding correlates with Fournier's (1998) relationship indicator of "quality of association", since it shows that the appropriateness of news presentation to the device conveys a sense of constant availability that users welcome and might perceive as demonstration of loyalty from journalism. With this, we see how the news organizations' adaptation to users' primary device or news presentation preferences can be appreciated demonstrations of news outlets' loyalty to users.
In this sense, our data revealed that in the good friends' dynamic, the loyalty that individuals attach to journalism strongly correlates with the loyalty they feel from journalism to them. Within this two-way dynamic relation, editorial transparency and commitment stand out as the two main ways in which people recognize the loyalty from journalism. For instance, Germán, a 37-year-old human resources analyst, compared his relationship with a radio news programme to that of good friends by saying that he felt assured about the transparency of his journalists: they will not hide things from me (…) If the information is somewhat questionable, they will wait for the source to confirm it, or they will keep the information under scrutiny until they can check how accurate it is. They will tell me 'This is what we know' In turn, commitment, was perceived both by the number of hours the participants see journalists working and in the sense of social responsibility they detect in them: These last three months, I've listened to Matías Del Río on the radio in the morning, then on the television news in the afternoon, and in the evening as well. As a journalist he gives me the commitment I need. I can tell that he doesn't go to sleep until he's finished delivering the information, said Catalina, a 35-year-old engineer. On his part, Miguel, 64-year-old retired air force officer, said he saw commitment on the side of journalists when they demonstrated a sense of social responsibility: I think Chilevisión is a good friend, a close friend, with journalists that always try to go a little bit further … I can see that in them … they are committed with what is happening … They give me security and reliability, because if they are not there, I could have been fooled for a long time.
Interestingly, the ways participants showed their loyalty in this "good friends" dynamic was not aligned with previous literature on the notion. Frequency of attention was seldom mentioned as a characteristic of this friendship. Instead, participants talked about decisions they would make to keep the relationship alive. This entailed, responses such as adapting, tolerating, and forgiving particular journalism entities. Mónica, a 47-yearold biology teacher, for example, explained that her relationship with the radio news programme she has been listening to for almost 8 years is so strong that, when the show was bought by another broadcaster, she followed them and therefore changed her habitual radio station. She also acknowledged starting to use Instagram to be able to interact more with the journalists who host it. Mónica described this radio program as "part of my life", and therefore, adapting to the changes or innovations they bring forward, appears natural to her.
Likewise, loyaltyjust like in good friend relationships -also seems to elicit tolerance to certain aspects of journalism that users do not particularly enjoy. Mónica, for example, admitted liking the radio news show she listens to everyday "even though they are a bit posh" and Pablo admitted watching Canal 13's YouTube channel despite not liking everything they show. This is an interesting finding since is shows that loyalty is not necessarily linked with having a pleasant news experience or a quality moment with the news, but rather to maintain an ongoing and meaningful relationship with journalism entities. In fact, participants acknowledged that they are patient and forgive certain mistakes of the journalism entities they are loyal to. As Gastón, a 37-year-old agronomist, put it: Mónica Rincón and Daniel Matamala are like a friend I trust but who I know can make mistakes. They are my friends, I love them … I haven't cancelled them because I know their history and they are always nice to me.
The duration of the relationship therefore stands out as an important factor behind participants' decision to remain loyal to journalists or news brands. Carmen, a 52-year-old English teacher, like many other participants, was confused by the changes in the political orientation of her news station in recent months. However, she said she is going to give them "some more time" to make up their minds, because she still can't explain their "erratic behaviour". Party Friends' Dynamic: "My Relationship with the News Media that I'm Following Now is More Like This … There's a Party So I'm Hanging Out with Them" The second relationship dynamic we found is that of "party friends". This refers to those relationships that participants saw as circumstantial, but which had the potential to develop into good friendships. This is illustrated in the words of Carolina, a 23-year-old equine therapist, who equated the relationship she has been forging with these Instagram accounts to the dynamic she has with those friends she only sees when she goes to a party: There are certain contexts where I usually go with certain friends. There are friends with whom I will hang out regardless of the context and there are other friends whom I'll call if I want to go to a party. I think my relationship with journalism now is more like this … There's a party so I'm hanging out with them a lot because of their characteristics, because of my "bias", and because of the information I expect to receive in this precise context. Like Carolina, many other participants acknowledged that they started following different Instagram news accounts or began paying more attention to journalists on social media because they wanted to better understand what was happening on the streets. The series of social protests that occurred from October 2019 led to changes in Chileans' news use practices (Lazcano-Peña, Gálvez-Pereira, and Zuleta-Cavalli 2021) which were characterized by a widespread audience interest in seeking information via digital news sources. For several of these participants, the relationship dynamic they built with online news is highly linked with self-definition and social identity (Papacharissi 2013). This is consistent with findings on selective exposure that showed the tendency of individuals for preferring likeminded political outlets (Holbert et al. 2010;Knobloch-Westerwick 2014). However, although participants described this relationship in a highly positive manner, they did not seem comfortable with associating it with a good friendship. The main difference between "party friends" and the good friend's dynamic is the feeling of security that the partner offers. Similar to what happens when one becomes friends with a person, the beginning of the relationship between participants and journalism seems to heavily depend on shared affinities. In the specific case of the relationship with the news, this translates into like-mindedness with the editorial line of the news outlet, its accessibility through a userfriendly interface and a taste for the presentation of the content. As Carolina explained: I have a fluid relationship with them, the information comes in small capsules and is illuminating a lot of things. I don't want to read a news story on a large sheet of paper, I want it to be precise, to be concrete.
Interestingly, for the relationship to evolve, a sense of safety should develop. For example, Gastón claimed that the news site Interferencia was like "that friend you're just getting to know, you like him, but it's not yet a deep friendship". When asked how this friendship could move forward, he replied: the relationship I have with them now seems too fragile because they are new and not widely known, so they could disappear any minute. If they keep doing things like this and get more mainstream, I think we can become good friends. The third relationship dynamic we found is that of family members. This is characterised as a relationship with journalism that is linked to a sense of familiarity that has been built up over time with particular news brands or journalists. Many of the participants that identified their relationship with journalism with family ties reported having had a personal relationship with the news since childhood, especially with television. Interestingly, not all of them referred to this relationship as particularly positive. In fact, the way in which they describe certain news outlets or journalists reveals a distinctive kind of loyalty based on profound familiarity. This feeling of familiarity underlined a high sense of intimacy that is manifested in feelings of respect and tenderness. For instance, Carlos, a 39-years-old university lecturer, compared a TV journalist with a father because of the respect he inspires: I respect [Fernando] Paulsen a lot, he is a good journalist, as if he were my father, I like him even with his flaws … I think that he has a lot of conflicts of interest, but he is a good journalist … he is intelligent.
For Tomás, a 35-years-old educational psychologist, on the other hand, the relationship he has with news programmes is like that of the sweet aunt who constantly visits him at home, "I'm very fond of that aunt, she inspires tenderness in me. Sometimes she gets a bit harsh, and I stop listening to her".
Frequency of attention stands out as a differentiating factor between this dynamic and the previous ones. Interestingly, loyalty here does not seem to be measured according to practices that involve frequency of attention. Carlos, for example, noted that although he trusts and respects a journalist, he does not see him every day: Normally, when he says something on television, I analyse it, process it and give it more importance because he's been doing this for many years and has seen a lot (…) I don't listen to him religiously, but when I do, I listen to him. I'm 40 years old and I don't have to talk to my father every day.
Interestingly, loyalty is recognizable in participants expectations of journalism that range from understanding to betrayal or disloyalty. Miguel, for example, recognized that although the radio station he listens to has made several mistakes in recent times, he has not stopped listening to it because he "knows them" and understand the "these mistakes were not intentional". This sense of understanding, however, comes with boundaries that when broken are perceived as a betrayal, just like in real families. Germán commented that he felt betrayed by TVN because of the way they covered the social outbreak: "They [TVN news] have made us feel they are part of the family but when they had to show it, they turned their back on us". Interestingly, audiences can also commit betrayal. For instance, Belén, a 33-year-old physiotherapist, acknowledged that in her house they are loyal to the news show on Mega. It was difficult for her to explain why, but she claimed that it was something she learned while growing up and it would be a demonstration of disloyalty if anyone in her family would switch to another channel. Overall, the different ways in which loyalty is expressed in this type of relationship show how this notion is illustrative of "frequency fallacy" (Costera Meijer and Groot Kormelink 2021) as frequency of use did not seem to be an indicator of loyalty.
Classmates Dynamic: "I See Them Every Day, But We are Not Friends" The fourth dynamic we identified was one of classmates, either at school or university. A common reason behind this comparison was that, even though participants visit, listen to, or watch certain news sources or formats daily, they would never consider them as friends or family. As Laura, a 34-year-old engineer, said: the "written format" would be the peer who sits next to her at school because "we can have a dialogue (…) we study together, we discuss topics, we exchange opinions (…) I feel that I can read at my own pace, but we wouldn't be friends outside of the classroom". Interestingly, although loyalty was not present in this type of relationship, participants' responses seem to illustrate two main reasons why frequency of news use does not involve loyalty.
The first reason why this frequent use of news is not related to loyalty is because of a lack of political like-mindedness with the news outlet. For example, Marco, a 46-year-old PR strategist, emphasized that he felt conflicted about visiting a news site on a daily basis whose political viewpoint differs from his own. I know I am being contradictory in what I am saying and that I should say that the news site I visit the most is close to me and that I like it, but this is not the case. This has a purely functional purpose: to know what happened here, there and that's it.
The second reason we identified was lack of credibility. A group of participants compared the news accounts they follow on social media, specifically on Twitter and Instagram to the "popular" members of the class because they "are always talking everyone pays attention to them, but you can't really trust them". For instance, Germán said he checks news accounts on Twitter every day but recognized that it cannot be taken seriously because "everything that is posted on Twitter has to be corroborated". As he put it: "I talk to them, but I don't actually listen to what they are saying". Participants recognized that they interact (as scanning their tweets and favorite them every now and then) with them every day because it was "a fun way to stay informed". Such frequent attention to social media news accounts that users do not trust can be explained by the fact that they rely on other sources of information as well (Nelson and Kim 2021). Users' relationship with the "popular classmates", thus, is closely conditioned by the entertainment they get from knowing and learning what others are talking about. To consider them a friend with whom they could be loyal, however, they would first have to trust them.
Interestingly, the link between trust and loyalty is not as straightforward as previous literature on loyalty has suggested. A group of participants compared investigative journalism sites, specialised journalists or long form to the classmate that is the smartest in the classroom and gets good grades. These news outlets, journalists and news formats were characterised as trustworthy, and responsible, but serious and rather distant. Therefore, the dynamic with this classmate only revolved around the need for truthful information. Francisco, a 23-year-old nurse, for example, said that he usually talks to these classmates when he "needs to study for an exam" because he knows that they are serious and the information they provide will be truthful, especially on relevant political or social issues. However, the seriousness and sense of formality attached to being responsible and truthful hindered participants to feel they could develop a friendship with these journalism entities. Francisco admitted that he would not consider CIPER or Mónica Rincón -journalist-as friends because he does not feel "smart enough to have anything to talk to them about". Likewise, the style and presentation of the news also seemed to create a distance between them. As Carmen described it: "I know that CIPER has the information I need, but I can't read them every day. I don't have the time or energy".
Overall, the classmates' dynamics demonstrated that frequent visits to news sites or placing trust in them does not necessarily mean that users had a sense of loyalty to these journalism entities. Instead, this is more likely to reflect news use practices that are shaped by the need for reliable information (Costera Meijer 2020), ease of use (Davis 1989), and entertainment motivation (Costera Meijer and Groot Kormelink 2021).

Discussion
This article sought to examine loyalty as a separate phenomenon and from an audience perspective, developing an in-depth understanding of how news users hold a relationship with journalism. Our results have implications for journalism studies and practice in five different ways.
First, our findings question the prevalent way in which loyalty has been operationalised. From the users' point of view, loyalty does not equal "profitable measures" such as clicks or shares and is not automatically correlated to donations or memberships. While from a producer perspective, frequency of attention is what counts as loyalty, from a user perspective, we identified a frequency fallacy (Costera Meijer and Groot Kormelink 2021), as daily visiting news sites was not indicative of participants' feelings of loyalty. Instead, adapting their routines or behaviour, tolerating certain personality traits of some journalists and forgiving the mistakes of some media outlets emerged as cues of loyalty that show deliberate choices to maintain a relationship with journalism. Interestingly, as loyalty is a two-way process, loyalty of users can be enhanced by journalism's demonstrations of adaption to user needs and preferences, editorial transparency of news outlets, and work and social commitment of journalists. Thus, from a user perspective, a better indicator of loyalty might be what users are willing to do for journalism, rather than the type or duration of attention they pay to it, making it hard from the point of view of individuals, to quantify it.
Second, our findings confirm the idea that loyalty is associated with people's belief that journalism helps them to navigate life and to participate in society (Broersma 2019;Nelson and Kim 2021). The fact that one of the participants reflected on one of his favourite journalists as a father is a clear example of this. This comparison is particularly striking because it showed the impact of news on people and what they value about it. More importantly, it highlights that relationships with certain news media or journalists are relevant because they elicit feelings of support and security; they find their relationship to be truly valuable (Costera Meijer 2020). In this sense, loyalty to journalism is akin to feeling that one is living a life in the company of an ally, someone with whom it is not necessary to talk every day, but on whom one can lean on when it is necessary. Someone who is on one' s side and who will look out for one's welfare, not in a condescending or benevolent way, but rather openly and transparently looking out for one' s own best interest. These findings are highly relevant for journalism studies and practice because they illustrate how the kinds of attitudes and emotions constituting loyalty go beyond measures and indicators of profitable news attention.
Third, we discovered that while loyalty involves trust, trust does not necessarily involve loyalty. This has implications for journalism studies in two ways. On the one hand, it corroborates the findings of previous investigations indicating that users who are loyal to a given news media have trust in it (Nelson and Kim 2021;Goyanes 2020). And on the other hand, it shows that trust alone does not signal loyalty. Loyalty, as we suggest in this article, involves a set of feelings and practices that reflect the meaning people attach to journalism in their lives. This is not to say that news users want to develop a loyal relationship with all the news outlets they access. For many, the best way to stay informed is to hold loyalty towards some news outlets and to frequent other news sites out of a need for cognition (Tsfati and Cappella 2003). For others, however, not having a news medium with which they can develop a friendship, or a family relationship can be a source of disappointment and even loneliness.
Fourth, we identified that the concepts developed in the brand-consumer relationship are different from those encountered in the audience-journalism relationship. In brand relationships, the objects of loyalty -the entities to which people relate to -are often "non-living" objects, such as sneakers or packaged goods. While both possess attributes that can change over time, journalismas an object of loyaltyis of an entirely different nature; journalism is "alive". As one participant stated, journalism is constantly "talking to people", whether it is through broadcasting TV or publishing a news story on Instagram. This means that the relationship that users maintain with the news is essentially dynamic as it is part of a continuous process of lively interaction that is often felt as a real connection. For instance, participants who followed journalists on social media, noted that they not only "thought they knew" journalists from listening to them on a daily radio show, but that they felt they "really knew" them because they saw what they posted on Instagram or Twitter about their private lives.
Fifth, we found that users seem to be clear about what to expect from journalism, but they are somehow unaware of what journalism expects from them. The actions that loyalty might fuel or motivate-such as liking, sharing the news or even subscribing and donating money-are not self-evident for new users. For instance, Pablo, who watches news on YouTube, said that he had never thought of liking a news video because journalists, unlike influencers, never ask for that. And, similar to findings on audience-supported business models (Hansen and Goligowski 2018;Nechushtai and Zalmanson 2021;Kim, Wang, and Malthouse 2022), users' decisions to engage in subscriptions involve elements that are not necessarily linked to loyalty. Participants claimed that they would subscribe or donate money to news organizations if they would see that they really need it. As Francisco put it: "I wouldn't give money to Canal 13, they seem to have enough". Thus, while loyalty implies a high sense of emotional connection with news organizations or journalists, this bond does not automatically invite a willingness to like or subscribe to the news outlets to which they are loyal to. In this sense, the benefits that journalism can get from a loyal audience could be enhanced if news organizations communicated better what they would want from users (Nechushtai and Zalmanson 2021).

Conclusion
The present study advanced our understanding of loyalty by adding width and depth to the notion. By taking relational metaphors as sensitizing concepts, we identified four main types of relationship dynamics participants had with journalism. These are: good friends, party friends, family members and classmates. The analysis of these dynamics allowed us to understand, first, that loyalty is a multidimensional phenomenon that is personal, situated and temporal. This means that loyalty to journalism entities does not unfold and becomes meaningful in a universal and predictable way. It varies according to people's personal experiences and tastes in relation to journalism, the situations of everyday lifeboth spatial and socialin which journalism is embedded, and the length of time they have been in contact with a particular journalism entity. For example, the dynamics of Pablo's relationship with the Youtube News Channel were linked to personal experiences and tastes: he described himself as a visual person and claimed to have fond memories of watching television growing up. Everyday life situations were also part of this association, as he considered it a moment of pleasure to watch this YouTube channel on his iPad at home, more specifically on his sofa. Likewise, he also referred to the temporal dimension: he said that Channel 13 "was always there", thus implying a prolonged presence of this news channel over time.
A second finding is that loyalty is an absolute value that people attach to their relationship with a particular journalism entity. Participants did not talk about being "slightly" loyal to a news brand or "highly" loyal to a journalist. Instead, they talked about their ongoing relationship with journalism through the description of feelings and practices evoked by it. Feelings of affinity, companionship, respect, safety or tenderness, and practices such as adaptation, tolerance, or forgiveness, were common in participant's described experiences of loyalty.
Third, we found that loyalty functions as a dynamic two-way relationship. This means that the loyalty individuals give to journalism entities is often experienced as a relational response to the loyalty they receive from them, even when the device itself (television, newspaper, radio) is seldom recognized in scholarly studies as "interactive". The reason for this has to do with users' experiencing news use as an interaction. As we saw in the results section, for individuals, watching the news on a given TV channel may involve not only an interaction with the medium or device (iPad, TV, smartphone), but also with the type of content (newscast, interview, news talk show), the format (audiovisual, text, audio) or with the news brand. Each of these different elements "speaks" to users through the different attributes they display and convey, which are perceived and evaluated by individuals according to their tastes, needs or interests. Users then respond to journalism entities through a number of different actions, such as recommending them to friends or family, visiting the medium in times of need or viewing it frequently.
Overall, the broad array of insights of our research speak to the relevance of studying loyalty from a relational perspective. This approach sheds light on how people maintain different types of relationships with journalism and the meaning they attach to this bond.

Limitations and Suggestions for Further Research
This study is not without limitations. Our analysis relied on interviews with a relatively well-educated sample group of news users. We believe that future studies could advance research on loyalty by employing methodological innovations that allow a more detailed observation of specific aspects of news use, such as frequency or time spent, as well as exploring the experiences of different demographics and more diverse types of users. Likewise, future studies could further the exploration of how the relationship users develop with journalists, media outlets, or news programs is conditioned or shaped by social media in terms of professional branding (Molyneux, Holton, and Lewis 2018;Olausson 2018) as a driver to loyalty.
Note 1. For more information about the informants, see Appendix.

Disclosure Statement
No potential conflict of interest was reported by the author(s).