Abstract
Social media influencers (SMIs) are important information sources for socio-political topics among adolescents. Adolescents may even ascribe digital super peer perceptions to some SMIs for sharing sensitive information (e.g., LGBTQ+ issues, health) that may be unavailable or uncomfortable to discuss among their regular peers. However, it remains unclear which socio-political topics adolescents encounter and which differential susceptibility factors predict perceived SMI trustworthiness or perceived societal relevance of the information they provide. To address these gaps, we conducted a collaborative citizen science project with N = 386 adolescents aged 12 to 18. Adolescents contributed and indicated their appraisal of socio-political posts by SMIs about a range of socio-political topics, including sensitive issues. Attribution of altruistic motives to SMIs, less exposure to socio-political SMI content, and the presence of opinions in such content emerged as important factors explaining why adolescents may perceive SMIs as trustworthy and their socio-political information as societally relevant.
Keywords
Introduction
They are dancing on TikTok, sharing make-up tutorials on YouTube, or capturing viewers’ attention through gaming streams on Twitch: Social media influencers (SMIs) are digital opinion leaders who have a large follower network on social media (Harff et al., 2025). Contrary to celebrities, SMIs become popular through their successful online self-presentation rather than a certain profession (e.g., athlete, actor; Piehler et al., 2022), making them seem relatable and approachable (J. Lee et al., 2024; Zimmermann et al., 2022).
Some SMIs aptly use their popularity to draw attention to political or societal topics (Zimmermann et al., 2023) like climate change, migration, party politics, or LGBTQ+ issues (Dekoninck et al., 2023; Harff & Schmuck, 2025a; Stocking et al., 2024). Such topics can be distinguished into formal political topics, which relate to more conventional political issues, and lifestyle-based political topics, which are of collective concern (Harff & Schmuck, 2025a; Suuronen et al., 2022). We cluster these under the umbrella term socio-political topics—issues that are political or societal in nature, including those not inherently political (e.g., sports) but that are portrayed in a way that they relate to broader collective concerns, like discrimination in sports (Broussard, 2020).
SMIs have great power to inform people about socio-political topics. This could be particularly relevant for sensitive or niche issues for which role models may be missing in one’s own networks (e.g., sexuality or health; Brown et al., 2005; Elmore et al., 2017). Their content can further have a mobilizing power, increasing audiences’ political engagement (Dekoninck & Schmuck, 2024; Naderer, 2023). Nevertheless, it may also expose people to unfounded claims or conspiracy theories (Guess & Lyons, 2020). Adolescents may be particularly susceptible to SMIs’ socio-political content, since they are often targeted by SMIs (Schmuck, 2021; Van Dam & Van Reijmersdal, 2019), generally rely on SMIs as primary sources for news (Newman et al., 2025), and may still lack necessary media literacy skills to detect misinformation or ulterior motives (Harff & Schmuck, 2025b; Zimmermann et al., 2022).
Scholars have provided important insights into socio-political SMI content (e.g., Gonzalez et al., 2024; Sehl & Schützeneder, 2023) showing, for instance, that such content has become more prevalent over time (Harff & Schmuck, 2025a). However, several research gaps remain. First, prior studies either only investigated SMIs’ socio-political content (Gonzalez et al., 2024; Sehl & Schützeneder, 2023) or solely explored predictors for using SMIs as socio-political information sources (Harff & Schmuck, 2025b). Yet, they did not systematically link exposure to such content to followers’ appraisal of the content or source. Without knowledge on the topics within and characteristics of socio-political content on adolescents’ curated feeds, assumptions about their receptivity to it—or to SMIs—remain speculative. Moreover, it is vital to understand which differential susceptibility factors determine adolescents’ perceptions of SMIs as trustworthy and relevant socio-political information sources.
Second, the channels and SMIs are often predefined in research, for instance, by selecting globally popular SMIs, revealing only limited socio-political posts to analyze (Gonzalez et al., 2024). In the fragmented, high-choice media environment (Van Aelst et al., 2017), it becomes increasingly difficult for researchers to identify SMIs who are popular among adolescents. Additionally, SMIs often post socio-political information solely in ephemeral content like Instagram Stories (Gonzalez et al., 2024), making it undetectable for researchers when only peripherally checking SMIs’ accounts. Consequently, there is a pressing need for methodologies that enable researchers to directly capture the content young users are organically exposed to.
Third, extant research largely focuses on late adolescents or young adults (Harff & Schmuck, 2025b; Schmuck et al., 2022; Soler-i-Martí, 2015). Yet, it is essential to include younger age groups too, as opportunities for political engagement on social media already arise in early adolescence when political identity is forming and has not yet solidified (Eckstein et al., 2012).
To fill these crucial gaps, this study adopted a collaborative citizen science approach with N = 386 adolescents aged 12 to 18 years. We systematically linked the socio-political SMI content adolescents contributed to perceived SMI trustworthiness and societal relevance of such content. We further examined developmental, dispositional, and social factors predicting these perceptions. By directly capturing adolescents’ real encounters with socio-political SMI content, this innovative approach provides valuable insights into the specific content that shapes adolescents’ information diets (RQ1-2). Moreover, by drawing on the Differential Susceptibility to Media Effects Model (Valkenburg & Peter, 2013), we shed light on the factors that determine adolescents’ appraisal of such content (RQ3-4). Our study thus offers important methodological and practical implications for future research on adolescents’ information acquisition via social media.
SMIs and Their Socio-Political Content
Social media influencers (SMIs) have gained growing importance as information sources, surpassing legacy news media in relevance on platforms like TikTok or Instagram (Newman et al., 2025; Wunderlich et al., 2022). Especially young users favor retrieving substantive information through SMIs (Newman et al., 2025), who discuss a wide range of socio-political topics (Gonzalez et al., 2024; Harff & Schmuck, 2025a). Suuronen et al. (2022) distinguished between formal political topics, which relate to political actors, processes, or institutions, and lifestyle-based political topics, whereby “SMIs make a reference to any issue that they think is a collective concern and which has an impact on the wider society” (Suuronen et al., 2022, p. 4). Young people’s identification with formal politics is often low nowadays, while they tend to identify with issues that are personally relevant or are associated with certain lifestyles (Andersen et al., 2021). Given SMIs’ typical focus on lifestyle content, they lend themselves well to communicating information that aligns with youth’s political interests (Soler-i-Martí, 2015). Content analytical studies reveal that SMIs are more likely to address lifestyle-based political topics (57%–67%; Gonzalez et al., 2024; Harff & Schmuck, 2025a). Nevertheless, SMIs may also engage with formal politics, especially during elections (Allgaier, 2020; Schwemmer & Riedl, 2025). Therefore, this study addresses both formal and lifestyle-based political topics, collectively referred to as socio-political content.
Compared to personalities like politicians active on TikTok (Umansky & Pipal, 2023), SMIs often lack formal training or professional expertise in the topics they address (Zimmermann et al., 2023). Hence, the quality of their socio-political content becomes a critical concern. Prior research shows that many of SMIs’ socio-political YouTube videos are opinionated, and among those featuring factual claims, most videos lack a source for at least one of these claims (Harff, 2025). Alternatively, political SMIs may refrain from expressing their own evaluation of socio-political topics and, where present, distinguish it clearly from factual information (Sehl & Schützeneder, 2023). Although scholars have started to explore SMIs’ socio-political communication, there is little knowledge about the socio-political content adolescents encounter or about its content characteristics (i.e., presence of opinions, facts, sources). Thus, we ask:
SMIs and Digital Super Peer Perceptions
Throughout adolescence, parents, peers, and the media shape norms and values regarding appropriate behavior and interactions with others (Genner & Süss, 2017). Brown et al. (2005) argue that adolescents “may turn to the media as a relatively safe and less embarrassing way to be associated with, and perhaps learn from, [. . .] than their real-life same-age peers” (p. 421). More broadly, they suggest that in periods of developmental change (like adolescence), media may serve as an important source of information to reduce uncertainty. This is particularly evident among early-maturing girls, who showed a heightened interest in sexual media content (Brown et al., 2005). As such, they argue that adolescents may turn to mass media personalities who can fulfill a role similar to peers and could thus be classified as “super peers” (Brown et al., 2005, p. 421). This may also be of particular interest to adolescents who are exploring their gender and/or sexual identity or who diverge from society’s gender norms or heteronormativity. Furthermore, the super peer concept has been applied in a health context, assuming that adolescents acquire normative information about alcohol and tobacco use from mass media without necessarily having peers in their own environment who use substances (Elmore et al., 2017). A positive relation between adolescents’ media-related cognitions and their attitudes about alcohol and tobacco use among peers has been found, hence reinforcing the idea of mass media as a super peer.
Moving beyond mass media, the affordances of the digital media environment render the impact of media characters perceived as super peers even more important. These high-choice media environments allow adolescents to curate individualized feeds (Valkenburg et al., 2016; Van Aelst et al., 2017) with content from their favorite SMIs. Additionally, the ubiquitous availability of such content allows for the development of stronger bonds with SMIs (Hoffner & Bond, 2022).
First qualitative evidence suggests that super peer perceptions of media figures may also hold for SMIs, who have a peer-like status and may be particularly relevant for socio-political topics where adolescents lack like-minded others (e.g., sustainability; see Dekoninck et al., 2023). Consequently, SMIs may become key agents of political socialization—a lifelong process whereby individuals acquire political norms, attitudes, and behaviors (van Deth et al., 2011). More specifically, SMIs may model behavior or provide information in a way that is unavailable among adolescents’ own peer group (Brown et al., 2005). Hence, these perceptions may be particularly relevant for sensitive topics like gender, sexuality, and (mental) health, which adolescents may not be able or willing to discuss in their immediate social circles but which SMIs have been found to address relatively frequently (Harff & Schmuck, 2025a). Specifically for news influencers, LGBTQ+ issues (i.e., expressing LGBTQ+ identity or supporting LGBTQ+ rights) are among the most prevalent identity-related topics within SMI content (Stocking et al., 2024). Taking these recent dynamics into account, we transfer the super peer concept to the digital environment by forwarding the novel concept of “digital super peer” (DSP) perceptions: “followers’ views of an SMI as an extension of their peer network (Giles & Maltby, 2004), arising from the provision of novel information, values, or norms” (Van Houtven & Schmuck, 2026, p. 1). The current study investigates the prevalence of topics related to gender, sexuality, and (mental) health, which have previously been identified as issues for which media may be seen as a super peer (henceforth called DSP topics). As such, we ask:
Adolescents’ Appraisal of SMIs and Their Socio-Political Content
Similar to TV personalities, SMIs sway their followers via so-called parasocial relationships (Reinikainen et al., 2020). Parasocial relationships are enduring bonds media users form with media figures based on repeated parasocial interactions—illusionary experiences of “conversational give and take” created by recurrent exposure to a media personality (Horton & Wohl, 1956, p. 215). Researchers argue that parasocial relationships with SMIs may be stronger than with regular celebrities because SMIs are perceived as more relatable (J. Lee et al., 2024; Zimmermann et al., 2022). These personal bonds explain why adolescents consider SMIs as trustworthy sources of socio-political information (Breves & Liebers, 2022; Wunderlich et al., 2022). Similarly, relationships with SMIs may affect whether youth see SMIs’ political content as relevant (Harff & Schmuck, 2025b). However, little is known about potential additional factors that may lead adolescents to perceive SMIs as trustworthy and their socio-political content as societally relevant.
Trustworthiness and Societal Relevance of SMIs and Their Socio-Political Content
The Differential Susceptibility to Media Effects Model (DSMM; Valkenburg & Peter, 2013) posits that developmental stage, individual predispositions, and social context jointly determine susceptibility to media content. Accordingly, this model can be used to examine factors predicting adolescents’ perceptions of SMIs’ trustworthiness and the societal relevance of their content.
Developmental Stage
As teens enter late adolescence, they may be more likely to consider socio-political content as societally relevant, as opportunities for political involvement arise during this life stage (Eckstein et al., 2012). Meanwhile, younger adolescents may be more likely to (uncritically) trust SMIs, as their political identity is not yet consolidated (Eckstein et al., 2012) and they may rely more on SMIs’ subjective content (Zimmermann et al., 2022).
Individual Predispositions
We focus on adolescents’ socio-demographic information, social media use, and media literacy. First, we consider gender and education since young men and those with lower political knowledge generally consider SMIs important political information sources (Harff & Schmuck, 2025b).
Second, regarding social media use, we build on the theoretical approaches of parasocial relationships and opinion leadership (Harff et al., 2025; Stehr et al., 2015): Adolescents who are more frequently exposed to socio-political content by SMIs have greater opportunities to build a close bond with that SMI, increasing trust in the SMI and perceived relevance of their socio-political content.
Third, media literacy may influence adolescents’ susceptibility to SMIs and their socio-political content. For SMIs, two areas of literacy seem decisive: Source-related literacy can be understood as awareness of SMIs’ intentions—guided by profit-driven interests (Kim & Kim, 2021) or by internal factors, hence being more altruistic in nature (Breves & Liebers, 2022). Followers’ altruistic motive attribution has already been linked to greater perceived SMI trustworthiness (Breves & Liebers, 2022; Park & Hoffner, 2024), although SMIs are not necessarily formally educated on the topics they address. Therefore, source-related literacy can also be understood as people’s awareness that SMIs are often merely autodidact experts, meaning that they are perceived as having acquired knowledge by building on reliable sources and self-education (Dekoninck et al., 2023). Additionally, this study considers information-related literacy, which pertains to the ability to understand and evaluate information, and therefore equips individuals with the necessary tools to distinguish factual information from misinformation (Jones-Jang et al., 2021). In the current study, we specifically examine information decision-making which can be described as a central skill that is employed to check whether online information is accurate. It captures the extent to which effortful processing is used when evaluating the credibility of online information (Metzger et al., 2015). Assessing both source-related and information-related media literacy allows us to examine whether these dispositional factors explain adolescents’ appraisal of SMIs as trustworthy and whether they view their socio-political content as societally relevant.
Social Context
Finally, for the social context, we take peers’ and parental influence into account. Regarding the latter, we consider parents’ mediation style and supportive parent-adolescent communication about social media use. Parental mediation is not only an important predictor of the intensity of adolescents’ online media use but also prevents risky online behavior (Appel et al., 2012; Knop et al., 2015). Mediation of children’s digital media use has also been associated with the extent to which youth reflects on news content (S. Lee et al., 2023). For SMIs’ socio-political communication, this could imply that adolescents’ appraisal of SMIs and their socio-political content is influenced by their parents’ mediation strategies. Furthermore, the communication climate at home plays a key role in the development of dysfunctional social media use patterns in adolescence (Appel et al., 2012). Adolescents’ communication with parents about their social media use can protect them from adverse consequences (Schmuck, 2021), while adolescents’ news media use has also been found to relate positively to their communication with parents about socio-political topics (Boyd et al., 2011). However, it remains unclear whether child-parent communication also determines adolescents’ appraisal of the socio-political content they encounter through SMIs.
During adolescence, the focus of social orientation shifts strongly toward the peer group (Baumgartner et al., 2011). As such, peer group norms and perceived peer pressure may impact adolescents’ social media habits. Prior evidence suggests that young people prefer discussing SMIs’ socio-political content with their friends rather than with their teachers (Zimmermann et al., 2022). So, the peer group may determine which SMIs adolescents follow and the importance they ascribe to their content.
Following the DSMM logic (Valkenburg & Peter, 2013), the research presented above highlights (a) developmental, (b) dispositional, and (c) social factors that may be associated with adolescents’ susceptibility to SMIs’ socio-political content. Specifically, the current study aims to investigate the differential susceptibility factors that determine whether adolescents appraise SMIs as trustworthy and rate their socio-political content as societally relevant. Consequently, we ask:
Research Context
Our research was conducted in Flanders (i.e., the Dutch-speaking part of Belgium). In Flanders, most young people follow SMIs (Vanmeldert, 2022) and about a quarter use the internet to find socio-political information (Schulz et al., 2018). A cross-country report on emerging citizenship (Schulz et al., 2018) further shows that Flemish schools integrate civic and citizenship education into various courses rather than teach it as a separate subject. They do so by including a wide variety of learning objectives, ranging from understanding key concepts, over promoting critical thinking, to developing positive attitudes toward participation and engagement. The report also demonstrates that Flemish teachers engaged in civic and citizenship education are mostly trained in responsible internet use (e.g., source reliability, social media), compared to topics like human rights or elections (Schulz et al., 2018). Finally, while pupils generally ascribe low importance to conventional citizenship, most support the endorsement of equal rights for all genders and ethnicities (Schulz et al., 2018). Limited attention to socio-political topics in Flemish schools may foster the importance of SMIs for learning about these issues.
Method
To examine the socio-political content Flemish adolescents are exposed to by social media influencers (SMIs), we conducted a collaborative citizen social science project (see Van Houtven & Schmuck, 2025 for a detailed description of the method) which actively includes citizens in the research process (Purdam, 2014). When working with adolescents, the citizen science approach can deliver unique insights to which researchers would otherwise not have access, as this method “is especially helpful in a new context in which political information and participation have become highly personalized” (Binder et al., 2021, p. 233). Specifically, adolescents can offer novel insights into the specific actors through which they encounter socio-political content and the characteristics of this content.
In the current project, adolescents contributed socio-political SMI content via an online platform, similar to mobile experience sampling (MES, e.g., Valkenburg et al., 2021). Yet, in contrast to MES studies, adolescents are aware of the research goal and actively help shape the research process (see Binder et al., 2021). Based on the information obtained from the citizen science study, we conducted a quantitative content analysis of adolescents’ contributions. Adolescents provided feedback throughout the project and, afterward, two of them also served as our advisory board. That way, we gained insight into adolescents’ participation experience as well their interpretation of the research findings, including the studied population in each stage of the research process. The current study successfully underwent ethical assessment by KU Leuven’s Ethical Review Board: G-2022-5237-R6(AMD).
Design and Procedure
In total, 386 adolescents aged 12 to 18 participated in our project (see OSF for more information about the recruitment and procedure: https://osf.io/ejqwp/overview). The data collection started with a pilot phase (April–June 2023) during which the researchers visited secondary schools across Flanders. In these sessions, the researchers clarified the citizen science method, goals, expectations, and procedure for this specific study. Adolescents were asked to participate in the study and provide feedback. This step was followed by a public phase (September–November 2023) whereby the project was disclosed to all interested Dutch-speaking adolescents in Belgium. The project consisted of three parts: (1) an intake survey, (2) a citizen science survey, and (3) a quantitative content analysis.
Intake Survey
First, we assessed factors that may explain why adolescents appraise SMIs as trustworthy and rate their socio-political content as societally relevant. Apart from adolescents’ socio-demographic information (e.g., age) and their social media use (e.g., usage patterns), we also included measures related to media literacy (e.g., information decision-making), parental (e.g., active mediation) and peer (e.g., group norms) influence. A complete overview, including reliability and descriptive statistics, can be found on OSF. In total, 340 adolescents completed this survey. Adolescents were allowed to indicate that they did not understand a survey item sufficiently, resulting in missing values on some variables.
Citizen Science Survey
The citizen science part was conducted in a separate survey, in which adolescents shared their observations of socio-political SMI content. To this end, adolescents were asked to use social media as they normally would (Purdam, 2014) and could access the study’s online platform at any time to make a contribution. Adolescents could either share the URL to the post or provide a screenshot or screen recording (for ephemeral content, e.g., Snap or Instagram Story). Adolescents were allowed to contribute as much socio-political SMI content as they liked. After each contribution, participants were asked to provide the researchers with simple information about each post (e.g., the platform on which they found the content, the name of the SMI who posted it, and whether they followed this SMI; see also Binder et al., 2021). As such, adolescents’ data collection revealed the names of the most relevant SMIs who post socio-political content, their respective social media channels, and the type of content exemplified by the actual posts. Next, adolescents reported their trust appraisal of the SMI (M = 4.18, SD = 0.87; e.g., “The influencer is [not at all–very] trustworthy”) and the extent to which they found their socio-political content societally relevant (M = 3.83, SD = 0.95; e.g., “The information in this post is [not at all–very] important for society”) with two items each (5-point Likert scales; full item list see OSF). In total, 298 adolescents contributed content in the citizen science survey.
Contributions
The 298 adolescents that took part in the citizen science survey were allowed to contribute as much content as they liked. All together, they shared 574 contributions of socio-political SMI content. Most of the contributions originated from Instagram (228 or 39.72%) and TikTok (196 or 34.15%), followed by YouTube (130 or 22.65%). Posts from Facebook (7 or 1.22%), Snapchat (2 or 0.35%) and X (1 or 0.17%) were least common. Some adolescents also contributed posts (10 or 1.74%) that they found on other platforms like Twitch.
Content Analysis
Finally, the research team conducted a quantitative content analysis of the citizen science contributions. Two student coders manually coded all shared URLs that directed them to a specific social media post still available at the time of coding. This content analysis served to categorize the specific socio-political topic of each contribution and to assess the content characteristics of the posts (i.e., presence of opinions, facts, and sources). Intercoder reliability was successfully established (Brennan & Prediger’s Kappa > 0.75). The codebook and results of the reliability tests are accessible via OSF.
Analytical Sample
Given the study’s citizen science approach, we include all contributions made by adolescents in the data analysis. We provided elaborate information on SMIs and socio-political topics by means of tutorials and definitions in the surveys. Yet, adolescents’ perception of who is or is not an SMI does not necessarily align with that of the researchers in all instances. However, answering the pressing need to look at these actors from adolescents’ perspective and not predefine the accounts that are relevant for this age group, we in the end followed adolescents’ perspective on SMIs. This approach is also in line with the citizen science methodology that valorizes the involvement of the studied population more than the generalizability of the findings (Purdam, 2014).
Sample Description
In what follows, we consider all data from the 252 adolescents that completed the intake survey and made a contribution in the citizen science survey. Thus, these 252 participants constitute our analytical sample. The participants in our study had a mean age of 15.19 (SD = 1.16). Overall, more girls (188 or 74.60%) participated in the project than boys (62 or 24.60%). Two adolescents preferred not to disclose their gender (0.79%). In our sample, 201 adolescents (80.08%) followed a high, 48 (19.12%) a middle, and 2 (0.78%) a low education trajectory. One adolescent indicated to follow a different educational track but did not specify which one. Participants indicated to spend most hours per day on TikTok (M = 2.67, SD = 1.75), followed by Snapchat (M = 2.14, SD = 1.74), Instagram (M = 1.50, SD = 1.27), and YouTube (M = 1.24, SD = 1.34). Our data shows that adolescents spend the least time on Facebook (M = 0.32, SD = 0.77) and X (M = 0.14, SD = 0.39), as well as some other platforms (M = 0.38, SD = 0.82) like Twitch. Finally, participants indicated that they follow, on average, 30.67 SMIs (SD = 27.38). Most also encounter socio-political SMI content on their feeds; only 8 adolescents (3.17%) indicated that they never see such content.
Data Analysis
Data analysis was conducted in R. To answer RQ1, asking about the socio-political content (i.e., topics and content characteristics) adolescents encounter through SMIs, we ran descriptive frequency analyses. The same goes for RQ2, investigating the prominence of DSP topics within SMIs’ socio-political content. We formed a dummy variable distinguishing content dealing with a DSP topic (i.e., gender and sexuality or [mental] health) from content not addressing any of these issues. Furthermore, formal political topics were distinguished from a variety of lifestyle-based political topics following extant literature (Harff & Schmuck, 2025a; Suuronen et al., 2022). To this end, we categorized content coded as “political system, actors and institutions,” and thus discussing conventional politics, as formal political topics, and all other content about topics that are of broader collective concern (e.g., race or gender issues) jointly as lifestyle-based political topics (see also Suuronen et al., 2022).
To address RQ3 and RQ4, asking about the factors explaining adolescents’ perceived relevance of SMIs’ socio-political information and their trust appraisal of those SMIs, we ran multi-level linear regressions using the lme4 (Bates et al., 2015) and lmerTest (Kuznetsova et al., 2017) packages in R. Separate models were used to examine adolescents’ appraisal of the SMI as trustworthy and their socio-political content as societally relevant. In all models, participant ID was used as a grouping variable, since the contributions are nested within the adolescents. All developmental, dispositional, and social factors were included. Additionally, contribution-specific variables (i.e., following that SMI, contributing about lifestyle-based/formal political topics, and contributing about DSP/regular topics) were included in all models. Since SMIs have repeatedly been found to focus more strongly on lifestyle-based political topics (Gonzalez et al., 2024; Harff & Schmuck, 2025a), they may communicate these topics more expertly, enhancing their perceived trustworthiness and the perceived relevance of their socio-political communication. Alternatively, by reducing the complexity of political topics (Schmuck et al., 2022), SMIs’ communication about formal politics may stimulate trustworthiness and perceived relevance. Adolescents may further appraise SMIs as trustworthy and their content as relevant if they are exposed to DSP topics like gender, sexuality, and (mental) health for which information is lacking in their own environment and which are increasingly important in adolescence (Brown et al., 2005; Elmore et al., 2017). Furthermore, the general content characteristics were used as controls. These reflect the degree of opinion orientation as well as the presence of facts and sources about the socio-political information. Existing research in this area suggests that young adults evaluate the opinion orientation and subjective stances in SMIs’ content critically (Zimmermann et al., 2022), but similar findings for adolescents are lacking. Finally, dummy variables were created for gender and education, whereby girls and low-to-middle education were the reference categories, respectively. As such, the analyses combined data from the intake survey with those of the citizen science survey and the content analysis. Since not all contributions were still available when the data analysis took place, the sample size was reduced for some variables.
Results
Descriptive Results
To address our first research question regarding the different topics, we categorized the contributions ourselves, as identifying specific socio-political topics may be too complex for adolescents. For the distinction between formal and lifestyle-based political topics, we drew on work by Suuronen et al. (2022). Frequency analyses revealed that 208 posts contained information about a socio-political topic. Among these contributions, lifestyle-based political content was shared more often (167 or 80.29%) than formal political content (41 or 19.71%). Among the single topics, content about conventional politics like political systems/actors/institutions (41 or 19.71%) was the most prevalent socio-political topic, followed by sustainability/climate change (28 or 13.46%), (mental) health (25 or 12.02%), body positivity/shaming (23 or 11.06%), society (23 or 11.06%), and gender/sexuality (19 or 9.13%). Regarding the digital super peer (DSP) topics (RQ2), these numbers also show that posts about (mental) health and gender/sexuality were among the most frequently shared, suggesting that adolescents often encounter posts about these topics through SMIs. With 44 (21.15%) contributions relating to (mental) health, gender and sexuality, it seems that SMIs engage frequently with DSP topics in their socio-political communication (see OSF for a complete overview of all topics).
In a next step, we examined the content characteristics of the posts containing a socio-political topic (n = 208) by assessing whether the content was opinionated, fact-based, or both, and whether it included sources to support any factual claims about a socio-political topic (RQ1). We found that the socio-political SMI content was strongly opinionated, with 165 (79.33%) contributions presenting opinions about socio-political issues, while 132 (63.46%) stated facts. Additionally, SMIs only sporadically used sources to support factual claims about a socio-political topic (45 or 34.09%). Mind that the presence of sources was only coded for those posts where factual claims about the socio-political information were present (n = 132).
Multi-Level Models
Perceived Societal Relevance
Next, we investigated adolescents’ differential susceptibility to socio-political content from SMIs (see DSMM, Valkenburg & Peter, 2013), explaining why they appraise such content as societally relevant (RQ3). Our first model (Model 1, Table 1) shows that adolescents’ developmental stage (i.e., age) was not significantly related to the perceived relevance of the socio-political information, and neither were the dispositional factors: gender, education, frequency of social media use, number of SMIs followed, exposure to socio-political SMI content, usage patterns, information decision-making, altruistic motives, and perceived autodidact expertise of SMIs.
Regression Table for Perceived Societal Relevance of the Content—Model 1 (RQ3).
Note. Results highlighted in bold designate significant findings. The number of reported participants is lower than the total number of participants in our analytical sample, as this analysis includes variables that were coded in the content analysis. At the time of coding, some of adolescents’ contributions were no longer available and could, thus, no longer be coded.
Female gender = reference category.
Low to middle education = reference category.
Formal political topics = reference category.
Regular socio-political topic = reference category.
As for the social factors, perceived parental mediation and parents’ supportive communication did not prove significant. Next, adolescents’ perceived peer group norms and peer pressure sensitivity were also not significantly related to the perceived relevance of the socio-political information they contributed.
Considering the content characteristics, we found that following the SMI whose content was contributed did not relate to adolescents’ perceived societal relevance of the post. Yet, contributions on lifestyle-based political topics were more likely than those on formal political topics to predict perceived societal relevance (b = 0.36, SE = 0.18, p = .048), while content about DSP topics did not. In addition, our findings show that the socio-political information was not perceived as more societally relevant when it contained facts, while it was when the post contained opinions (b = 0.34, SE = 0.16, p = .029). In a separate model (Model 2, Table 2), we also tested for the presence of sources, which did not act as a significant predictor.
Regression Table for Perceived Societal Relevance of the Content—Model 2 (RQ3).
Note. This model is identical to Model 1 (Table 1), except for the inclusion of the presence of sources, compared to the presence of facts. Results highlighted in bold designate significant findings. The number of reported participants is lower than the total number of participants in our analytical sample, as this analysis includes variables that were coded in the content analysis. At the time of coding, some of adolescents’ contributions were no longer available and could, thus, no longer be coded.
Female gender = reference category.
Low to middle education = reference category.
Formal political topics = reference category.
Regular socio-political topic = reference category.
Perceived SMI Trustworthiness
We further investigated which factors determine the perceived trustworthiness of the SMIs sharing socio-political content (RQ4). Our analyses (Model 3, Table 3) again revealed that age (developmental factor) did not serve as a significant predictor.
Regression Table for Perceived SMI Trustworthiness—Model 3 (RQ4).
Note. Results highlighted in bold designate significant findings. The number of reported participants is lower than the total number of participants in our analytical sample, as this analysis includes variables that were coded in the content analysis. At the time of coding, some of adolescents’ contributions were no longer available and could, thus, no longer be coded.
Female gender = reference category.
Low to middle education = reference category.
Formal political topics = reference category.
Regular socio-political topic = reference category.
Also, gender and education were no significant predictors of adolescents’ trust in the SMI sharing socio-political content. Other dispositional factors—namely frequency of social media use, number of SMIs followed, usage patterns, information decision-making, and perceived autodidact expertise of SMIs—did not significantly predict adolescents’ perceived trustworthiness of the SMI. However, adolescents’ altruistic motive attribution (b = 0.37, SE = 0.10, p < .001) predicted greater trust in that SMI, while more frequent exposure to SMIs’ socio-political content was negatively related to their trustworthiness (b = −0.19, SE = 0.07, p = .010).
Looking at social factors, our model reveals that adolescents’ perceived peer group norms and peer pressure sensitivity as well as parents’ active mediation and supportive communication did not significantly relate to trust perceptions.
Last, regarding content characteristics, following the SMI from the contribution (b = 0.59, SE = 0.13, p < .001) and contributing lifestyle-based political SMI content (b = 0.51, SE = 0.15, p = .001)—but not DSP content—again served as significant predictors. By contrast, neither the presence of opinions nor facts significantly predicted perceived SMI trustworthiness. The same goes for the presence of sources, which we again tested in a different model (Model 4, Table 4). An overview of all multi-level linear regressions can be found in Tables 1 to 4.
Regression Table for Perceived SMI Trustworthiness—Model 4 (RQ4).
Note. This model is identical to Model 3 (Table 3), except for the inclusion of the presence of sources, compared to the presence of facts. Results highlighted in bold designate significant findings. The number of reported participants is lower than the total number of participants in our analytical sample, as this analysis includes variables that were coded in the content analysis. At the time of coding, some of adolescents’ contributions were no longer available and could, thus, no longer be coded.
Female gender = reference category.
Low to middle education = reference category.
Formal political topics = reference category.
Regular socio-political topic = reference category.
Discussion
We employed an innovative collaborative citizen science approach to gain insights into (a) the socio-political content adolescents aged 12–18 encounter through social media influencers (SMIs), (b) the prevalence of digital super peer (DSP) topics (gender, sexuality, [mental] health) within these encounters, and (c) the differential susceptibility factors that determine adolescents’ perceived trustworthiness of the SMI and societal relevance of their socio-political content (DSMM, Valkenburg & Peter, 2013). First, our findings confirm that SMIs’ content is highly relevant in adolescents’ media diets, with adolescents following a range of different SMIs and encountering SMIs’ socio-political content at least sometimes. Prior research shows that SMIs are more likely to discuss lifestyle-based political topics than formal political topics (Gonzalez et al., 2024; Harff & Schmuck, 2025a), which our findings seem to corroborate. Although “political system, actors and institutions” (i.e., formal politics) was the distinct topic category that was most prevalent among the shared SMI content (19.71%), SMI content addressing a variety of lifestyle-based political topics (80.29%) was most represented among adolescents’ contributions. All but one of the formal political contributions revolved around the same news item about an ongoing court trial of the popular Flemish SMI Acid. Since the trial was the subject of many online and offline debates, formal political topics possibly would not have featured as much among adolescents’ contributions if the data had been collected in a different period or country. While socio-political content may only be occasionally covered by SMIs (Harff & Schmuck, 2025a), this example shows that there are certain posts that attract adolescents’ attention and can be recalled by many of them. This provides some evidence for the notion that specific socio-political SMI posts can go viral.
Among the lifestyle-based political topics, more sensitive DSP topics like (mental) health and, unexpectedly, body positivity/shaming were prominently discussed by SMIs. The latter may also classify as a DSP topic, as the issue is relevant for adolescents who find themselves in a period of bodily change and “body-image development” (Kvardova et al., 2022, p. 370) but may be too sensitive to discuss among peers. Additionally, the frequent contributions on sustainability/climate change support previous qualitative work, suggesting that SMIs may also serve as relevant role models for other niche topics (Dekoninck et al., 2023). Due to SMIs’ engagement with DSP topics and adolescents’ susceptibility to such content (see Dekoninck et al., 2023; Elmore et al., 2017), this finding tentatively suggests that adolescents may ascribe DSP perceptions to SMIs who discuss delicate topics or share information that is unavailable in adolescents’ own social environment. Notably, gender and sexuality—topics which we expected would occur frequently—were only the sixth most prominent in SMIs’ socio-political content. Having participated in a classroom setting, some participants may not have felt comfortable sharing such content. Also, sharing (sensitive) DSP topics was not related to greater perceived societal relevance or trust in the SMI. Given these conflicting findings, future research should further consider the potential of digital super peer perceptions and investigate for which SMIs and socio-political topics young people may ascribe such views. Additionally, it would be worthwhile to explicitly measure adolescents’ DSP perceptions, as these may explain the outcomes of exposure to SMI content. Yet, our analyses revealed that adolescents attributed more societal relevance to lifestyle-based political content and higher trust to the SMI who posted it. This highlights the importance of lifestyle-based political topics among young people and their preference for issues that have a stronger connection with personal lifestyles and identities (Gotlieb et al., 2015).
Furthermore, turning to the multi-level models and differential susceptibility factors explaining adolescents’ appraisal of the contributions (DSMM, Valkenburg & Peter, 2013), our findings show that perceiving SMIs as having greater altruistic motives and following the SMI whose content adolescents contributed was positively related to their trust in that SMI. Our findings align with other studies linking altruistic motives to trustworthiness (Breves & Liebers, 2022; Tomlinson, 2018). In addition, our analysis revealed that adolescents who see socio-political SMI content more often attribute less trust to the SMI than other adolescents. We expected that SMIs’ suitability to communicate about lifestyle-based political topics (Gonzalez et al., 2024; Harff & Schmuck, 2025a) and their ability to reduce the complexity of formal political topics (Schmuck et al., 2022) would foster their trustworthiness. Yet, in the current study, we found a negative relation between adolescents’ exposure to socio-political SMI content and the trustworthiness they ascribe to SMIs. However, this relation occurred when following the SMI who posted the content was controlled for. More exposure to socio-political content of SMIs in general seems to make adolescents more skeptical toward their content, which is typical for the age group we investigated (see Zimmermann et al., 2022). However, following the SMI from whom they contributed content is positively associated with trustworthiness, confirming the idea that a close bond with an SMI, like a parasocial relationship, increases their perceived trustworthiness (Breves & Liebers, 2022).
Finally, with regard to adolescents’ perceived societal relevance of the socio-political information in their contributions, our findings show that opinionated socio-political SMI content is perceived as more societally relevant than content that does not contain opinions about a socio-political topic. Extant research suggests that youth prefer opinionated content of YouTubers over more objective TV news coverage (Zimmermann et al., 2022). Our study adds to this finding, suggesting that adolescents also perceive opinionated SMI content as more societally relevant. It is possible that SMIs who express their opinion about socio-political topics are perceived as more invested in these issues. Drawing on the idea of opinion leadership (Stehr et al., 2015), this could give adolescents the feeling that what these SMIs care about and have an opinion on is relevant for them as well.
Limitations
This study has some limitations. First, our analytical sample was skewed for education, in that pupils in the high educational track were overrepresented. Although digital competence and media literacy are skills learned at all levels of the Flemish educational system, more attention is attributed to them in class among those in a higher educational track. As such, it was more suitable to embed our citizen science project within this group’s learning goals. On the other hand, our study likely presents findings about adolescents that are expected to think more critically about what they encounter online. We therefore invite scholars to study adolescents’ appraisal of socio-political SMI content among a more educationally diverse sample. Relatedly, as this study was only conducted in Flanders, the sample may limit the generalizability of our research findings. Yet, the digital landscape and media literacy skills have been found to be similar in different Western countries (De Coninck et al., 2023).
Second, many of the citizen scientists were pupils participating in a school context. This setting may have prevented adolescents from sharing content that is more sensitive in nature or distributed by unreliable sources. In-class data collection not only stimulated participation rates but also allowed for a more extensive information session in which the researchers could elaborate on the conceptualization of key concepts and the study design. This session aimed to increase participants’ understanding of the research goal, which is a crucial element in the citizen science approach. Future studies could consider creating a more private setting (e.g., through privacy shields used for exams) to mitigate the risk of social desirability. Yet, our approach allowed pupils to interact while naturally scrolling through their feeds, which fostered lively peer discussions about the research topics and SMI content they encountered—ultimately encouraging deeper engagement with the study.
Third, some contributions did not correspond with our definitions of SMIs and socio-political content. Yet, this could be seen as a finding in itself, showcasing, in line with prior research on politics (Sosnovik & Goga, 2021), that there may be discrepancies between researchers’ and respondents’ conceptualizations of such concepts.
Finally, although we provided platform-specific instructions on how to share the URL of adolescents’ contributions, many of the TikTok URLs did not lead the researchers to the specific posts. Consequently, we could not assess the specific content adolescents provided. Thus, future research should explore more reliable and easy ways for adolescents to contribute social media content, like data donation (see, e.g., Haim et al., 2023). Relatedly, contributions that were no longer available when the content analysis took place further reduced the sample size. Yet, connecting actual encounters with socio-political SMI content to outcomes among adolescents constitutes an innovative design that offers unique insights.
Implications
Our findings have important practical implications. First, SMIs are considered relevant and trustworthy for various socio-political topics, including sensitive issues, and are more likely to be trusted by adolescents who ascribe altruistic motives to SMIs. Thus, we present valuable insights into groups who may be particularly susceptible to SMI content and could be targeted for media literacy programs.
Furthermore, our study has relevant methodological implications. Systematically linking the contributed content to adolescents’ appraisal can reduce recollection bias (Heiss & Matthes, 2017). Also, our participants provided content from many distinct accounts, highlighting the extremely fragmented and individualized media repertoires of adolescents. Yet, sometimes, multiple individuals shared the same pieces of content, implying that SMI posts can go viral in an over saturated media environment. Last, we have shown that the citizen science approach proves useful in this area, as it allows researchers to get closer to young people’s true media reality instead of imposing preconceived ideas on them regarding the platforms or accounts that are of importance. Thus, we encourage researchers to investigate adolescents’ media experiences through this methodological lens.
Footnotes
Acknowledgements
We thank the students involved in the coding of content for their diligent and thorough work.
Ethical Considerations
The Social and Societal Ethics Review Committee at KU Leuven University approved this project [approval number: G-2022-5237-R6(AMD)].
Consent to Participate
All participants provided online written informed consent prior to their participation.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Research Foundation Flanders [grant numbers: G0B7822N, 11Q2B24N].
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.