The value affordances of social media engagement features

Abstract

Social media engagement is ubiquitous but contested, simultaneously framed as an everyday form of support and an urgent societal risk. To make sense of these competing claims, we introduce the concept of value affordances, defined as the set of ethical, aesthetic, and relational principles that emerge from the interaction between different stakeholders and technological infrastructures. We develop a novel method involving focus groups and value cards to study the value affordances of engagement features and explore how international students attribute values to the Like, Comment, and Share buttons of TikTok, Instagram, and YouTube. Across platforms, participants agree that engagement features promote expression, care, and community and hinder privacy, mindfulness, peace, and safety. We discuss how our participants navigate value tradeoffs, emphasizing individual agency over structural factors when evaluating the design of platforms, using social media creatively, and assigning responsibility for harm to other users.

As the “first commandment of social media,” engagement seems to be everywhere (Hallinan et al., 2022, p. 209). Like, Comment, and Share buttons appear alongside content, influencers issue ubiquitous calls to action, and platform policies both encourage engagement and prohibit its “inauthentic” forms (Hallinan, 2021). Although engagement can potentially include any type of participation on social media (Jenkins, 2006), it often refers to “the most visible and obvious signifiers” (Hallinan et al., 2022, p. 209; see also Bucher, 2018), the buttons that transform user interactions into data and generate economic value (Gerlitz & Helmond, 2013). In coordination with complex recommendation systems, engagement shapes the visibility of people and content on platforms, producing what Taina Bucher calls “programmed sociality” (2018). While engagement carries the potential for “civic empowerment” (Tenenboim, 2022), there is growing concern that it may create “perverse incentives for polarizing content” (Rathje et al., 2021, p. 6), foster addiction (Bhargava & Velasquez, 2021), and contribute to precarious working conditions for creators (O’Meara, 2019). Simultaneously described as an urgent societal risk (Vaidhyanathan, 2018) and a mundane form of social support (Carr et al., 2016), engagement stands in as a synecdoche for the complicated status of social media (boyd, 2014) and, as such, an opportune site to explore platform values, the underlying principles governing and expressed through social media (Hallinan et al., 2022; see also Gillespie et al., 2020; Leurs & Zimmer, 2017; van Dijck et al., 2018).

While scholars recognize that design is value-laden (Lessig, 2006; Nissenbaum, 2001), they disagree about where to locate the “infrastructural values” of platforms (Hallinan et al., 2022), looking to diverse sources including technical reports and public disclosures (DeVito, 2017; Rieder and Skop, 2021), platform policies (Scharlach et al., 2023; Sybert, 2022), interfaces (Light et al., 2018), conversations with developers (Seaver, 2022), and algorithmic audits (Brown et al., 2021; Sandvig et al., 2014). These studies offer important contributions to understanding the politics of platforms, yet they typically treat values as a structural property of technology or a reflection of the worldviews of developers. In so doing, they risk overdetermining technology and overlooking the role of everyday users in constructing and challenging its significance. Responding to Rieder and colleagues’ call to approach “normativity as performativity… rather than idealized input into design decisions” (Rieder et al., 2023, p. 73), we turn to the concept of technological affordances.

Foregrounding the relationship between people and technology, technological affordances refer to “how objects enable or constrain” particular ways of acting in the world (Davis, 2020). While normative considerations appear in affordance research, particularly in relation to power inequalities, they are a secondary concern and typically come up in the interpretation of results (for a notable exception, see Aharoni et al., 2022). However, just as perceptions of what technology can do are profoundly perspectival (Nagy & Neff, 2015), so too are the values it promotes and hinders. To forefront this normative dimension, we introduce the concept of value affordances, defined as the set of ethical, aesthetic, and relational principles that emerge from the interaction between different stakeholders and technological infrastructures. Given the aforementioned gap in the literature regarding everyday users’ perceptions of such values, we examine value affordances from this perspective. Adapting techniques from the field of values in design (Belman et al., 2011; Flanagan & Nissenbaum, 2014), we develop a novel method for studying value affordances through focus groups to identify which values social media users associate with the Like, Comment, and Share buttons. Through six focus groups comparing Instagram, YouTube, and TikTok, we provide an emic perspective on platform values.

We begin by introducing the contested nature of social media engagement and argue that engagement features offer an ideal site to investigate platform values. To understand the relationship between social media users and engagement features, we turn to the concept of technological affordances, highlighting the importance of context and perception. Synthesizing the two approaches, we introduce our concept of value affordances and outline a comparative method for studying them involving value cards, solicitation prompts, and focus groups. We present our findings, including the values participants associate with engagement features and the perceived mechanisms connecting features to values. Together, these findings lead us to discuss the value tradeoffs that structure how users make sense of social media amidst public outcry about its harmful consequences.

Literature review

The contested value(s) of social media engagement

Defined as core features of “engagement” (Kim & Yang, 2017), opportunities to Like, Share, and Comment are everywhere on social media, providing a global grammar of interaction for interpersonal relationships and political participation. Such features are “objects of intense feelings” (Bucher & Helmond, 2018) that fundamentally shape how people use the platform and even “understand themselves” (Hallinan & Brubaker, 2021). Despite being “lightweight actions,” social media engagement can have real “social value” for users (Scissors et al., 2016): Likes contribute to feelings of support (Carr et al., 2016), self-esteem (Reich et al., 2018), and status (Marwick, 2013), while Sharing and Commenting augment users’ sense of influence (Kim, 2018; Oeldorf-Hirsch & Sundar, 2015). Researchers have also investigated social media engagement as a form of “political engagement” (Hallinan et al., 2022), where Liking, Commenting, and Sharing political content are “civic acts” that meaningfully shape democratic discourse (Tenenboim, 2022). Although researchers occasionally note that the failure to receive social media engagement can negatively affect individuals (e.g., Hayes et al., 2018; Scissors et al., 2016) or organizations (e.g., Tenenboim, 2022), these lines of research overall frame engagement positively and emphasize its pro-social and democratic value.

Another research trajectory paints a very different picture, shifting the focus to infrastructure and economic imperatives. Engagement buttons transform social interactions into data that fuels targeted advertising, algorithmic recommendations, and the economic valuation of social media corporations (Gehl, 2014). Based on the recognition that “that which engages is not always good” (Hallinan, 2021, p. 719), researchers have expressed concern that the content platforms promote contributes to major societal issues like political polarization (e.g., Rathje et al., 2021), disinformation (e.g., Freelon et al., 2020), and toxicity (Massanari, 2017). In the realm of individually realized harms, scholars and users alike have reflected on addiction (Bhargava & Velasquez, 2021) and the contention that engagement undermines mindfulness and quality time (Baym et al., 2020). Finally, given the increasing professionalization of social media entertainment (Cunningham & Craig, 2019), researchers have connected the engagement paradigm to precarious working conditions for content creators (Duffy et al., 2017; O’Meara, 2019). The value of engagement emerging from this line of research is much more negatively inflected, replacing the ideals of community and personal wellbeing with threats to democracy, individuality, and fair working conditions.

While it may be tempting to try and determine which line of research correctly identifies the value, or better, values, of social media engagement, we contend that it is more productive to read the conflicting accounts as an indicator of the broader tradeoffs of social media. We follow Nancy Baym’s (2018) argument that what “is so often blithely called ‘engagement’” is actually a way to manage essential dialectics between public and private (p. 24). The idea that there are tensions between different values has a long theoretical history (e.g., Schwartz, 2012). The situation gets more complicated when thinking about “how values take shape in, and are shaped by, computational systems” because researchers need to attend to “trade-offs between human values, systems design, and social forces that emerge through system use” (Le Dantec et al., 2009, p. 1; see also Flanagan et al., 2005). In the specific context of social media, researchers have identified tradeoffs between efficiency and quality in content moderation (Jiang et al., 2023) and between the “personal values that people bring with them to social media, such as community, creativity, and authenticity” with “what is valued by the infrastructure of social media” (Hallinan & Brubaker, 2021, p. 1564).

The competing connotations of social media engagement make engagement features an ideal site to investigate the value tradeoffs of social media design. Attention to design offers an interesting comparison to other sites of platform values such as public commitments made in press releases, governmental hearings, and policy documents. In a comparative study of the values articulated in social media platform policy documents, Scharlach et al. (2023) identified a mismatch between the public value commitments of platforms and those promoted in policy documents, with prominent values such as transparency, democracy, and accountability largely absent from policies. Yet, public value commitments are central to creating a corporate image that reflects the power of a platform company (Creech & Maddox, 2022; Haupt, 2021). Given their centralized status and formalized disclosures, the public value commitments of corporations are comparatively easy sites to identify the articulation of platform values. However, their connection to other aspects of a platform, including technological design and user practice, remains an open question.

This brings us to the more challenging task of identifying “infrastructural values” (Hallinan et al., 2022) due to the lack of theoretical consensus around what it means for design to have values and the lack of methodological consensus around how to investigate them. Researchers have developed different approaches, locating values within technical documentation and public disclosures (DeVito, 2017; Rieder & Skop, 2021), interpretations of platform interfaces (Bucher, 2018; Light et al., 2018), ethnographies of developers (Seaver, 2022), and algorithmic audits designed to reveal built-in biases (Brown et al., 2021; Sandvig et al., 2014). Similarly, researchers from the field of values in design have been concerned with how to integrate values into the design process of technological systems, focusing on the implicit worldviews of designers and creating tools for self-reflection (Flanagan & Nissenbaum, 2014). While these approaches are essential, they usually focus on the roles of values in the initial design phase and do not consider how everyday users construct and create meaning when using communication technologies. To better explore these perspectives, we turn to the concept of technological affordances.

From technological affordances to value affordances

Providing a middle-ground approach to understanding the relationship between people and platforms, the concept of technological affordances rejects the idea that either people or technology play a fully determining role (Evans et al., 2017; Nagy & Neff, 2015). Instead, technological affordances examine how material conditions shape social actions and vice versa. The relational approach can be traced back to the concept’s origins in environmental psychology, where affordances referred to the “properties of an environment relative to an animal” (Gibson, 1977, p. 285). Affordances have long played an essential part in the philosophical and sociological debates about the social construction of technology (Hutchby, 2001). Over the years, the concept has been extended, developed, and revised across disciplines such that technological affordances primarily refer to how particular animals—humans—relate to a particular aspect of the environment—technology.

In an updated take, Davis (2020) defines affordances as “how objects enable and constrain,” noting that affordances vary according to the context of use (2020, p. 20). This definition fits with her “mechanisms and conditions framework,” which shifts the question from what objects afford to how objects afford (i.e., the mechanisms involved), for whom, and under what circumstances (i.e., the relevant conditions). The framework responds to some of the unintended consequences of the concept’s interdisciplinary success, where scholars invoke the term inconsistently (Ronzhyn et al., 2022) and conflate affordances either with the features of technology or with the outcomes of its use (Evans et al., 2017). To facilitate comparison across studies, Evans and coauthors propose minimal conditions for something to be considered an affordance: (a) affordances should be distinct from both the object and the features of the object, (b) affordances should be distinct from the outcomes of using an object, and (c) affordances should have the possibility of variation.

Recent conceptual work emphasizes the contextual and perspectival aspects of affordances. Nagy and Neff (2015) argue that affordances are both social and material, a combination of physical features and social practices, and they develop the concept of imagined affordances which “emerge between users’ perceptions, attitudes, and expectations; between the materiality and functionality of technologies; and between the intentions and perceptions of designers” (p. 5). Highlighting the potential gaps between users, materiality, and designers, Shaw (2017) proposes the categories of “perceptible, hidden, and false” affordances (2017, p. 600). Similarly, McVeigh-Schultz and Baym (2015) propose the concept of vernacular affordances, or “the ways people themselves identify and make sense of affordances” (2015, p. 2). The emphasis on user perspectives is significant because it shifts the focus from technology as an abstraction to tangible experiences of sensemaking in digital environments.

While the literature on technological affordances demonstrates the importance of user perspectives, values often come as an afterthought, usually discussed when interpreting the significance of affordances rather than as a core object of empirical investigation. To address this gap, we propose the concept of value affordances which we define as the set of principles that emerge from the interaction between different stakeholders and technological infrastructures. By set of principles, we refer to a bundle of ethical, aesthetic, and relational values. Depending on the context, these values may generate tensions or contradictions. By emerging, we refer to different modalities of affordance identified in previous research where technology can, for example, refuse, encourage, or discourage (Davis, 2020). By stakeholders, we refer to actors such as designers, users, advertisers, or policymakers (Gorwa, 2022). Technological infrastructures refer to large-scale sociotechnical systems and are not limited to social media platforms.

In this study, we explore the value affordances of social media engagement features. Given the polarized accounts and potential value tradeoffs of engagement, we ask: Which values do people think the Like, Comment, and Share features promote and hinder? What mechanisms do people invoke in explaining their decisions?

Method

Data collection

To investigate the value affordances of social media engagement features, we conducted focus groups with everyday users. The interactive dynamics of focus groups make them ideal sites for studying collaborative sensemaking (Lunt & Livingstone, 1996). Although the design and use of features vary across platforms, prior research suggests that people often “think about affordances within the context of their overall assessment of all available platforms,” understood as the broader social media ecology (Zhao et al., 2016, p. 97). Accordingly, we compare value affordances across three popular, visual-centric platforms (Literat & Kligler-Vilenchik, 2021): Instagram, YouTube, and TikTok. For each focus group, we prompted participants to think about a particular platform and then reflect on their experiences with other types of social media.

Inspired by Belman et al.’s (2011) approach we provided each participant with a set of value cards (see Figure 1). Our value cards included the name and basic definition of 32 values (see the Supplementary Appendix for the full list of definitions), drawing from values in design research (e.g., Belman et al., 2011; Flanagan & Nissenbaum, 2014) and inductive studies of values on social media (e.g., Scharlach et al., 2023; Trillò et al., 2023). While research on values in design primarily focuses on ethical commitments such as transparency and equality, research on platform values highlights the importance of relational and communicative values like creativity and authenticity. We also included blank cards and invited participants to add other values, although no participant opted to do so. To surface potential tradeoffs in the value affordances of engagement features, we asked participants to choose three values that each feature promotes and three values that it hinders. We then had participants share the values they selected and respond to the values selected by others.

Given the global reach of social media platforms, combined with the tendency toward digital universalism in affordance research (Willems, 2021), we wanted to speak with people from diverse cultural backgrounds. To do so within the constraints of focus groups, we recruited international students currently studying at Israeli universities. We circulated an invitation to participate with a link to a short survey with questions about social media usage, age, gender, and field of study through university channels and self-organized WhatsApp groups of international students. From 166 responses, we selected participants for each focus group based on platform use, while aiming for diversity in gender, country of origin, and field of study. Our participants include 30 international students from 16 countries (see Table 1 for a breakdown of participant demographics). Despite diverse cultural and educational backgrounds, our participants are relatively homogenous in terms of age, with all but four participants between 18 and 35. We conducted six focus groups in English during June 2022. The first author led the focus groups, assisted by the second author. We audio-recorded each session, which lasted an average of 120 min, and gave participants the local equivalent of 75 USD for their time.

Data analysis

After each focus group, we independently drafted reflection memos and then met to discuss them. The lead author used otter.ai to transcribe the interviews, manually checking and correcting each transcript. To identify the values promoted and hindered by social media engagement, we tabulated the values for each feature and for engagement features as a whole. To capture the consensus around each value, we took the higher total (as more promoted or hindered) and subtracted opposing viewpoints. From these modified totals, we set a threshold of 5 for inclusion among the value affordances diagrammed and discussed in the Results section. To identify the mechanisms people invoked in their explanations of value affordances, we employed inductive thematic analysis to systematically identify patterns of meaning within the discussions of values (Braun & Clarke, 2012). We constructed a codebook to ensure a shared understanding of the analytical process and codes. Following the pilot round of coding and discussion, the first author coded each transcript in MAXQDA, followed by a close reading by the second author. Any unclear parts were marked and discussed until we reached a consensus (Hill et al., 1997). We have lightly edited the language of participant quotes included in the article for readability.

Results

What values do social media engagement features promote and hinder? In line with prior research on the social media ecology (Zhao et al., 2016), participants across the YouTube, TikTok, and Instagram focus groups talked about engagement features in broadly consistent ways. When we asked them to compare platforms, differences tended to be a matter of degree rather than kind. Instead, features emerged as the most salient locus of difference. Following our participants’ lead, we structure the results around features rather than platforms. We first discuss the overall values promoted and hindered by engagement features before turning to the specific value affordances of Like, Comment, and Share. After that, we describe the perceived mechanisms participants invoked when assigning values oriented around individual agency and structural determination.

Engagement features

Overall, participants agreed that engagement features promote expression, care, and community (see Figure 2). Each value aligns with how social media companies advertise the purpose of their platforms. Expression, for example, is prominent in corporate mission statements and platform policies (Maddox & Malson, 2020; Scharlach et al., 2023). Similarly, care has long been associated with sharing on social media through platform branding (John, 2016). The centrality of community is reflected in both “Community Guidelines” as a collection of policies and the brand identities of platforms like Instagram (Leaver et al., 2020) and YouTube (Snickars & Vonderau, 2009). Conversely, participants agreed that engagement features hinder privacy, mindfulness, peace, and safety, matching longstanding public concerns about the harmful consequences of social media (e.g., Vaidhyanathan, 2018). When identifying the values that engagement features hinder, our participants painted a picture of social media as a place filled with judgment, conflict, and risk, where giving up one’s privacy is the price of admission.

Like

When asked about the Like button, our participants argued that—in addition to expression, care, and community—the feature promotes credit, reputation, positivity, and pleasure (see Figure 3). Credit and reputation are closely related yet focus on different functions. Pushing the Like button is a way to give credit to someone for what they post, while the accumulation of Likes contributes to a person’s reputation on the platform. Such considerations are essential for content creators and anyone who uses social media in a professional capacity.

Our participants also talked about credit and reputation in the context of social interactions with friends and family, especially when posting about significant life events like starting a new job or moving to a new place. Participants associated positivity with the visual design of the button, whether a heart icon or thumbs up. The Instagram and TikTok groups especially emphasized this value, perhaps due to the lack of a dislike button on these platforms, reflecting an attempt to minimize dislike and other “negative” expressions on social media (Gray, 2021). Finally, participants associated Like with promoting pleasure by making others feel good and providing a tool to curate your feed, where the more you Like, the more you’ll like what you see.

Our participants also argued that, in addition to privacy, the Like button hinders authenticity, meaning, and humility. Because Liking is so easy and polysemic (Hallinan & Brubaker, 2021), our participants argued that Likes are not necessarily authentic expressions of approval. Furthermore, given the connection between Likes and visibility (Gerlitz & Helmond, 2013), our participants perceived the feature as part of an incentive structure that can work against authentic expression, with people chasing the latest fads for attention. As a shallow form of interaction that trades off with deep discussions, the Like button also hinders meaning. Finally, the quantification of Likes is connected to concerns about humility, where the act of being publicly Liked by others makes it difficult for users to stay humble.

Comment

In addition to expression, participants explained that Comments promote improvement, diversity, and meaning (see Figure 4). The value of improvement comes from understanding Comments as a feedback mechanism where the audience can express their thoughts and opinions. In an ideal world, participants explained, creators will use this information to improve their content. However, participants readily acknowledged that many Comments do not contain useful information due to the prominence of flattery and provocations. The values of diversity and meaning both relate to the comparatively open structure of Comments, which is not restricted to the positive connotations of the Like button. As a result, Comments can facilitate more diverse viewpoints. Similarly, the ability to express yourself is associated with the potential for greater depth of meaning, even if—especially on Instagram—people talked about the Comment section as an extended Like or reaction button, filled with emojis and little else.

The hindered value affordances of the Comment feature point to the negative consequences of openness. Indeed, participants strongly associated Comments with all four general value affordances hindered by engagement features, especially privacy and safety. Our participants also agreed that Comments hinder civility and positivity, framing the comment section as a negative place filled with conflict, with some invoking the specter of a “comment war.” In line with previous research showing the entertainment motive of Comments (Springer et al., 2015), many explained that they enjoyed reading the comments despite, or perhaps because of, all the conflict. Regardless of platform, participants agreed that people generally do not behave civilly in the Comments, although some described attempting to intervene when a discussion got particularly heated or they disagreed with what was being said. Yet, even these participants expressed doubts about the futility of their efforts. Just as politeness can be hard to find in the Comments, so too can positivity, especially in some YouTube and TikTok communities. Even the types of Comments that people enjoy reading—like a good joke or witty remark—often appear at someone else’s expense.

Share

Perceived as less expressive than the other engagement features, participants contended that the Share feature promotes care and community, along with the values of togetherness, efficiency, and pleasure (see Figure 5). Togetherness reflects a common use of the Share feature to send a photo, video, or post to friends and family. This form of social interaction can bring people closer and form a repertoire of (literally) shared references and jokes. However, there are undoubtedly other ways of using the Share feature, including calling out bad behavior or promoting an important cause. Regardless of approach, the Share feature provides an efficient form of communication, facilitated by clicking a button or copying a URL. Finally, participants also invoked pleasure, reflecting the experience of someone Sharing something entertaining with you or a sense of satisfaction when someone else appreciates what you Share.

On the other side, participants believed that Sharing hinders choice, privacy, and mindfulness. While potentially counterintuitive, participants across focus groups invoked choice to describe an obligation to interact with the content that friends or family members send to them. Although people mostly appreciated the thought, they did not necessarily appreciate the content, especially when it is low quality (e.g., an extremely filtered photo), politically inflammatory (e.g., videos related to the Israel-Palestine conflict), irrelevant (e.g., cat content), or simply too long. Together, these actions seem to impinge on someone’s choice of what to pay attention to and how to use their time (a point we will elaborate on below).

Although we found a diverse set of values associated with engagement features, 12 of the 32 value card options did not fit within the schema of value affordances, either because they were rarely invoked or because there was significant disagreement as to whether the value was promoted or hindered. Rarely mentioned values include excellence, passion, equality, beauty, wealth, happiness, and perseverance (which was the only value not mentioned at all). Among these values, the minimal presence of wealth is somewhat surprising given the increasing professionalization of social media (Cunningham & Craig, 2019). However, most of our participants did not perceive such concerns as relevant because they did not use social media in a professional capacity (cf. Duffy et al., 2017; O’Meara, 2019).

The group of contested values includes trust, power, beauty, creativity, conformity, and accountability. Contested values align with the idea that people negotiate the meaning of technology (Shaw, 2017). For example, users argued that engagement features both promote and hinder creativity. Evelyn explained that she uses the Like button whenever she encounters content she finds creative, and it helps her discover more inspiring content. However, according to Liam, engagement features undermine creativity because viral videos always spawn copycat content. Competing claims about the value affordances of the engagement features highlight the difficulty of neatly ascribing values to complex technologies: the Like button can simultaneously reward creativity and incentivize conformity. People understand and use technology differently, and when such technologies are connected to algorithmic systems and bound up in massive social networks, the diversity of intentions and opportunities for unanticipated outcomes multiplies. Given all of this, we argue that value affordances are better conceived of as more or less prominent rather than definitively present or absent. With enough contortions, the same technology may be able to promote and hinder very different values. And yet, our focus group research revealed significant areas of relative consensus. To make sense of this stability, we now turn to how people conceptualized the process of affordance—what does it mean for a technology to promote or hinder values?

Perceived mechanisms

Although each of the value affordances we identified represents a sense of relative consensus, some affordances seemed to be stronger and more intractable than others. Drawing on the theoretical foundations of prior research (Davis, 2020; Duffy et al., 2017), we distinguish between stable and contextual value affordances. While all affordances exist at the intersection between humans and technology, stable value affordances depend more heavily on technology while contextual value affordances depend more heavily on human elements. These differences align with how people perceive the mechanisms of value affordances: Stronger affordances have more structural explanations, and contextual affordances have more personal explanations. While both types of affordances surfaced in the focus groups, our participants tended to emphasize contextual ones and attributed most value affordances to personal choice and individual agency. We discuss both accounts in what follows.

Individual agency

To determine values associated with a particular feature, participants often started from their own experience or the imagined experience of someone similar to themselves. In their explanations, typical practices of use reveal the value affordances of a feature and reinforce a belief in personal responsibility (Maddox & Malson, 2020; Scharlach et al., 2023). For example, Evelyn argued that Comments enable power, explaining that

the words that you say or the comments that appear in a video have a lot of power, either in a positive or negative direction. I think it can definitely be used as a way of supporting someone, but I also think that if someone says something negative, it definitely can have an effect on the people that are seeing it and the people that it's directed towards.

Evelyn evaluates the power of Comments according to the messages that people leave alongside social media content, which can produce both desirable and undesirable outcomes. The ambivalence of the value is thus explained as a consequence of how people choose to use the feature rather than the feature’s formal qualities.

The influence of personal choice is not restricted to the risk of negative outcomes. In the accounts of our participants, it is also central to realizing the benefits of social media. Echoing the idea that “Sharing is caring” (Joanna), James justified the relationship between the feature (the Share button) and the value (care) with the following hypothetical scenario: “Let’s say someone’s having a bad day, or doing something difficult. You can Share something funny or something that might make them happier.” This is certainly not the only way to use the feature, as demonstrated by alternative explanations of sharing content to raise awareness about political issues and call out bad behavior. The resonance of this particular approach is reflected in the strong consensus around care. Even when people disagreed with the attribution of values, personal choice offered ways to explain the differences. As Lathika, reflecting on her disagreement with another participant, noted, “I think it depends on how you use the site” before conceding that the participant’s answer made sense within “a very specific context.”

Accounts that emphasize human agency need not focus on the agency of the participant. Indeed, our participants discussed other actors within their social networks, aligning with their primary use of social media for interpersonal interactions. Although we had two participants who used Instagram to promote organizations and an artist who produced content for TikTok and YouTube, we mainly talked with people who used social media for personal connections or to consume entertainment. Within these personal networks, the behavior of other users also contributes to the production of particular value affordances. For example, as discussed above, multiple users across focus groups explained the Share button creates a social obligation to watch content shared with them, restricting what they can pay attention to and hindering choice.

Structural determination

Although most explanations of how engagement features promote and hinder values focused on interpersonal dynamics, some accounts directly addressed structural determinants like platform design or geopolitical context. This could take the form of treating the platform as a communicative partner, something (or, metaphorically, someone) to be addressed through the use of engagement features. For example, Saanvi elaborated on her understanding of recommendations on YouTube, explaining that “The only time I have used the Dislike button is when there’s crass content and I don’t want to see that ever again on my playlist.” Saanvi uses the Dislike button to communicate with the platform to curate her recommended videos. Other structural factors include how the lack of face-to-face contact can provide a disinhibiting effect and lead people to say things in the comment section that they might not say otherwise. As Joanna put it, “if it was a discussion on the street and you were in front of the person, I’m quite sure you wouldn’t say ‘you are this, you are that.’ You would be probably punched.” Another technical feature people considered was character limits. Velna, for example, raised this point when comparing the comment sections on different platforms: “I don’t even think on TikTok you can send long comments actually. I think there are… how many characters…? So it kind of blocks some people from commenting too much.” Although neither Joanna nor Velna invoked the formal vocabulary of affordance mechanisms (Davis, 2020), we can interpret their accounts in that framework, with Joanna describing how pseudonymous profiles encourage less inhibited forms of social interaction and Velna pointing to the hard refusal of character limits as something that prevents more in-depth or considerate conversations.

Overall, appeals to structural factors were uncommon and often significantly more implicit than the previous examples suggest. While such accounts downplay human agency, they offer little else by way of explanation in terms of understanding how engagement features afford. This was particularly evident in discussions of the Like button. For example, some participants associated the Like button with pleasure, explaining that the simple act of pressing the button made them feel good, or invoking “dopamine” from receiving Likes from others. When one participant, Ashley, offered a more complex analytic account of the mechanisms of value affordances, she explicitly drew upon academic language to do so:

I think when I said that the Like button undermines authenticity, I meant that the Like button creates a sort of environment where when people create posts, it’s not about what they authentically mean but about what they can say that will get them the most likes. Not that the actual act of liking something is inauthentic. Sometimes it is, sometimes it isn’t… but the fact that liking exists and the way that the Like button and works in… I don’t know, there’s probably a fancy social science word for the socio-cultural context or something, I don't know… it creates an environment that undermines authenticity.

Not only were such explanations rare in the focus group, but when they did appear, they were discursively marked as outside of ordinary speech.

Finally, not all structural determinants had to do with platform design. A few participants invoked the role of geopolitical context and cultural norms, primarily in comparisons of Eastern and Western social media platforms. Jenny, a Chinese social science student, explained that she feels more comfortable posting selfies on Instagram because the app is primarily used by her close friends as compared to the larger and more diverse audience on WeChat. Similarly, Vivian explained that the connection between Commenting and accountability comes from Chinese policies that assign legal responsibility for posting false information. Cultural norms could also be platform-specific (Massanari, 2017). Participants repeatedly compared YouTube comments to Reddit, drawing on a perception of Reddit as a “democratic space” with very little moderation. The potential hostility of these spaces stood in contrast to the prevalence of emojis and positivity on Instagram. Although the consideration of platform conventions and national differences moves beyond individual experiences of social media, invocations of other external factors were rare.

Discussion

While critical design researchers have established that any technology involves value tradeoffs (Flanagan et al., 2005; Le Dantec et al., 2009), our study reveals that casual social media users perceive patterned tradeoffs associated with engagement features. Liking, Commenting, and Sharing content offer ways to express yourself, care for the people around you, and build or sustain communities. At the same time, these “technologies of evaluation” (Hallinan & Brubaker, 2021) can undermine mindfulness, understood as the cultivation of a non-judgmental attitude, as well as peace, privacy, and safety. The tradeoffs we identified largely align with other accounts in the literature, including Nancy Baym’s (2018) discussion of how social media engagement navigates the boundaries between public and private, the “normative paradoxes” of privacy (Helm & Seubert, 2020), and the opposition between free expression and safety in content moderation (DeCook et al., 2022). Yet, we found that users invoke a broader range of values, especially when we examine the value affordances of specific features. Beyond more obvious ethical values like privacy and expression of concern to policymakers and platform corporations, users discussed how social media engagement relates to pleasure, positivity, and togetherness. Expanding our consideration of what counts as valuable, then, is important to understand how users make sense of social media and why they continue to engage amidst growing public discourse about its harms.

The fact that our participants actively use social media despite recognizing its possible normative shortcomings raises a further question: How do they navigate the value affordances of social media? Or, modifying the formulation from Malte Ziewitz’s (2019) investigation of search engine optimization employees, how do users navigate the boundary between “good” and “bad” engagement? There is always the chance that our participants did not care about the values hindered by social media engagement. Given research on the growing public concern with mindfulness (Baym et al., 2020) and privacy (Greene and Shilton, 2018; Helm & Seubert, 2020), this seems unlikely, but our focus group data cannot definitively rule it out. However, the perceived mechanisms of value affordances point to possible alternatives. The emphasis on individual agency suggests that many of the tradeoffs associated with engagement features are less about the design of technology and more about the consequence of user practices. Starting from this premise, our participants described creative strategies to negotiate, downplay, or even resolve these tensions.

One set of strategies concerns how people try to confuse the algorithm behind their TikTok “For You” page or Instagram “Explore” page by Liking content they are not interested in order to discover more diverse content or disrupt the platform’s ability to track their interests. Another strategy involves the selective non-use of features. While all participants regularly used social media, some avoided the tradeoffs of engagement features by choosing not to Like, Comment, or Share. Indeed, very few of our participants reported regularly using all features. Finally, some participants minimized the potential harms of engagement, especially around privacy, by using these features in a more closed setting such as direct messages or small groups. These choices limit the reach of the features and give participants more control over their environment.

Another set of strategies concerns how users assign responsibility for promoting or hindering particular values. While our participants consistently emphasized the agency of users, they differentiated responsibility into categories of “us” and “them,” identifying with positive actions that promote values and blaming others for negative actions that hinder values. Framing Liking, Sharing, and Commenting as deliberate acts, participants were happy to take credit for promoting values such as expression, care, and community. When it came to discussing how engagement undermines peace and safety (Vaidhyanathan, 2018), participants were more likely to invoke stories about other people. However, there were some exceptions to this rule. Values such as privacy and mindfulness, seen as hindered by engagement features, were typically explained as inherent consequences of using social media rather than the fault of a few “bad apples.” Just as platforms consistently offload the responsibility for enacting values onto users in their public-facing materials (Maddox & Malson, 2020; Scharlach et al., 2023), users discursively took credit for promoting “good” values while offloading responsibility for “bad” values onto other users.

This shared vision of responsibility is politically convenient for platform corporations and does not invite critical thinking about the role of technology in society (Markham, 2021). However, our participants did not fully buy into the corporate message of platform values. Of the five governing principles platforms invoked in policy documents (Scharlach et al., 2023), users endorsed the idea that platforms promote expression and community, challenged the idea that they promote safety and choice, and reconfigured the value of improvement, changing the emphasis from how a platform will develop better technology to how feedback through engagement helps creators develop better content. Users also invoked other values that are part of public discourse but not a part of the corporate discourse (mindfulness, peace as hindered). Focusing on value affordances from the perspective of casual users highlights how users attend to the social aspect of engagement while platform policies focus on its technical and legal implementation.

As the overlaps and divergences between platform policies and user experiences suggest, when it comes to value affordances, different stakeholder perspectives matter. We found further evidence of perspectival differences in the responses from a practice focus group session we conducted with social media researchers. While researchers focused on the social implications and political function of platform design, drawing on specialized knowledge and terminology, our participants overwhelmingly reflected on interpersonal interactions. In this way, our participants also contrasted with the more structural analysis of social media reported from studies of professionalized social media users like influencers and creators (Duffy et al., 2017), as well as technologists and designers (Belman et al., 2011; Flanagan & Nissenbaum, 2014). The diverse backgrounds and interests of various stakeholders help construct a rich account of the value affordances of complex systems that might otherwise go unnoticed.

Conclusion

This article presented the concept of value affordances, defined as the set of principles that emerge from the interaction between different stakeholders and technological infrastructures.

Exploring the values that everyday social media users connect with ubiquitous Like, Comment, and Share buttons revealed that casual users mostly think about the social aspects of engagement, attributing value affordances to individual agency. We identified tradeoffs that shape the user-technology relationship and strategies that users employ to manage them, including creative ways of using social media and assigning responsibility for harm to other users.

Certainly, we have only told one part of the story of the value affordances of social media. Everyday users are one stakeholder among many, and Like, Comment, and Share are not the only relevant features of digital platforms. We hope that future studies will build on our concept of value affordances and our novel methodology in three ways. First, value affordances likely apply to aspects of design that are application or platform specific. Future research could investigate distinctive features such as TikTok’s Stitch or Facebook’s emoji reactions to explore possible platform differences. Second, to go beyond how technologies enable and constrain, future research could investigate other modalities of affordances. This would also allow researchers to situate individual perceptions of value affordances within the knowledge infrastructures of platforms and the amount of access these infrastructures provide (John & Nissenbaum, 2019). Finally, future research could employ our methodology to explore the perspectives of distinct populations of users, including creators or older users, as well as other important stakeholders like advertisers, developers, and policymakers. Given the associated tradeoffs, the framework of value affordances could be particularly generative for studying non-use or disconnection practices.

This study of value affordances shows how social media users manage value tensions, forefronts differences in platform values seen as important for engagement by everyday users, and allows us to look under the tip of the “values governance iceberg” (Greene & Shilton, 2018, p. 1641). Just like Greene and Shilton describe how developers are faced with decisions about how to integrate platform values, everyday users are left to decide how to create meaning about using engagement features. We argue that value affordances are part of this iceberg, showing the tensions of everyday engagement with platform values. Although casual users do not adopt the same vocabulary for discussing the political stakes of platform design, our value card solicitation prompts offer a way for researchers to explore their normative concerns. Broadening our understanding of value affordances could ultimately shift perspectives of the responsibility for the enactment of platform values (Scharlach et al., 2023) by bringing the tensions of various stakeholders and technological infrastructures from the bottom of the iceberg to the surface.

Supplementary material

Supplementary material is available online at Journal of Computer-Mediated Communication.

Data availability statement

The data underlying this article will be shared on reasonable request to the corresponding author.

Funding

This project has received funding from the European Research Council (ERC) under the European Union’s Horizon 2020 research and innovation program [Grant Agreement No. 819004].

Conflict of interest statement

The authors declare that there is no conflict of interest.

Acknowledgments

We would like to thank Limor Shifman for her support throughout the development of this article and beyond, the editorial team for their expert management, and the anonymous reviewers for their astute comments. We would also like to thank Tommaso Trillò and CJ Reynolds for their feedback on earlier versions of this article and our focus group participants for sharing their insights with us.

References