Does Follower Size Matter? Diversity of Sources and Credibility Assessment Among Social Media Influencers

Abstract

This study investigates how follower size influences social media influencers’ sourcing behavior and credibility assessment rigor when producing content for their audiences. Grounded in Social Capital Theory, this study examines the tension between popularity as a form of social capital and its limited capacity to predict rigorous credibility assessment using a global quantitative survey of 500 social media influencers across multiple languages and regions. The findings suggest that influencers with larger follower sizes utilize more diverse sources; however, follower size does not correlate with credibility assessment rigor. This underscores that follower size functions as a symbolic rather than epistemic resource, where source diversity often serves as a visible signal of professionalism rather than having a deeper verification. Additionally, credibility assessment rigor is not a significant predictor of source diversity, and platform type and content genre did not moderate the relationship between follower size and source diversity. These findings contribute to the influencer marketing literature by challenging assumptions linking popularity with higher scrutiny of content credibility. The study holds implications for platform policy, media literacy education, and influencer–brand collaborations. Recommendations are provided for improving transparency and source vetting among digital content creators in increasingly flooded social media platforms.

1. Introduction

Social media influencers (SMIs) are individual content creators who build sizeable followings and influence audience attitudes by producing content for specific social media platforms that many people use as brand information, entertainment, and even news online [1]. Their reach and impact are underwritten by booming adoption of social platforms: TikTok alone reports more than 150 million U.S. users, with daily usage nearing 100 min [2]. Industry investment has followed suit. The global influencer market is estimated to reach over 30 billion in 2025 [3].

In digitally networked environments, people lean on cognitive heuristics and quick cues such as popularity, source cues, or surface design, especially when deciding what and whom to trust [4]. For SMIs, those cues include follower counts, engagement ratios, and visible intimacy with audiences’ signals that may or may not track underlying reliability. This study is motivated by that tension: in a setting where information is abundant and verification is costly, what standards of credibility do influencers themselves apply, and how does their social media capital (e.g., follower size) shape the diversity of sources they consult?

Although SMIs are often perceived as credible by virtue of trust and expertise, such credibility is frequently inferred from readily observable signals, such as follower count, rather than through systematic evaluation of information sources [4]. Follower size functions as a form of social media capital, signalling reach and influence, yet its relationship with audience engagement is complex. Empirical studies indicate that follower count can exhibit diminishing returns, with mid-tier influencers sometimes generating higher engagement rates than their macro- or mega-tier counterparts [5,6]. This raises critical questions about how influencers with different audience sizes approach verification and source selection when creating content for their followers.

Moreover, the ‘mere number effect’ suggests that large follower counts can enhance perceived influence, even when audiences are aware of the presence of fake or inactive followers [7]. This narrative raises critical questions regarding how influencers, segmented by follower size, source and evaluate information to determine its credibility. Given that credibility is grounded in dimensions such as trustworthiness and expertise [8], understanding the diversity of sources consulted and the evaluation criteria applied by influencers has implications not only for marketing effectiveness but also for the broader integrity of information circulating in digital public spheres.

Existing literature shows that online credibility is often judged through heuristic cues such as popularity rather than systematic verification [4]. Influencer research has largely focused on endorsement effects, emphasizing trustworthiness, expertise, and message value [1,9], yet evidence suggests that follower size is not a straightforward indicator of influence. Engagement can decline with larger audiences [5], mid-tier influencers may outperform mega-accounts [6], and abnormal likes to followers’ ratios can harm perceived credibility [10]. Even when fake followers are disclosed, large numbers can still enhance perceived influence of the ‘mere number effect [11]. While social capital frameworks conceptualize influencer–audience ties as relational assets [12], little is known about how follower size shapes influencers’ own sourcing practices or credibility evaluation criteria. This gap limits understanding of how influencers across tiers construct diverse sources and apply credibility standards, a perspective important for both social capital theory and improving information quality in digital spaces.

This study examines how social media influencers with varying follower sizes from nano (1000–10,0090), micro (10,000–100,000), macro (100,000–1 million), to mega-influencers (above 1 million), use diverse content sources and apply credibility evaluation criteria across platforms such as Instagram, TikTok, and YouTube globally. In the context of a rapidly growing influencer economy with market size grew from USD 16.4 billion in 2022 to USD 24 billion in 2024 and 86% of U.S. marketers report intent to partner with influencers in 2025 based on SproutSocial [13]; the study employs a cross-sectional survey and statistical mediated moderation analysis to test hypotheses that follower size (social media capital) predicts source diversity, credibility assessment, and that platform type and content genre moderate these relationships, thereby addressing a critical gap in understanding how follower size shapes influencers’ own sourcing and credibility practices, a dimension overlooked by prior studies that focused on audience responses rather than influencer behaviours.

However, while larger audiences may motivate influencers to diversify their sources, such diversification may not directly reflect deeper verification or fact-checking rigor or even show fact-checking rigor. Prior studies show that as influencers professionalize, they often adopt strategies of sourcing that signal credibility to audiences and brands without engaging in meaningful verification processes [4,14]. In this case, source diversity can serve as a symbolic cue of transparency and professionalism rather than practice of actual diligence. Accordingly, follower size may heighten the visibility of credibility cues through the display of diverse sources without majorly increasing the underlying rigor of information evaluation. Recognizing this distinction between performative and substantive credibility is very important in understanding how social media capital directly shapes influencer information behaviours.

Furthermore, this study addresses the practical concern for the integrity of information in digital environments, where influencers play a central role in shaping public exposure to news, recommendations, and other content. Understanding how follower size, platform type, and content genre influence sourcing and credibility practices can inform policies promoting transparency, guide organizations toward credible partnerships, and help audiences engage critically with influencer content. These outcomes can advance media literacy, curb misinformation, and reinforce trust in digital communication, contributing to a more informed and resilient public sphere, well beyond the limited scope of endorsing a brand.

Grounded in social capital theory, which views relationships and network resources as assets that shape behaviour, this study conceptualizes follower size as a form of social media capital influencing how influencers source and evaluate information. By examining these dynamics across platform types and content genres, the research not only extends theoretical applications of social capital to the influencer context but also offers actionable insights for improving content credibility and information integrity in digital platforms.

2. Literature Review

2.1. Social Capital Theory

This study is grounded in social capital theory [15,16,17], which explains how resources rooted in social networks enable collective action through structural, relational, and cognitive dimensions. In the influencer context, follower size reflects structural capital (reach), while reciprocity, trust and shared meaning build relational and cognitive capital that underpin credibility and influence [12,17]. Viewing follower size as social media capital provides a basis for understanding how network resources shape influencers’ sourcing diversity and credibility evaluation, offering theoretical grounding for the hypotheses tested in this study.

This study defines social capital as the resources obtainable from one’s social network [18]. Structural capital is operationalized as follower size, the number of unique accounts subscribed to an influencer’s content [12]. Relational capital refers to trust, reciprocity, and sentimental bonds cultivated through ongoing interaction and authenticity signals [19]. Cognitive capital captures shared language, meanings, and narratives that align influencer and audience interpretations [20].

Social capital theory was selected for this study because it aligns directly with our focus on how network resources are reflected in follower size, shape influencers’ sourcing diversity and credibility evaluation. The study’s problem centres on understanding how and why influencers with different levels of social media capital vary in their information behaviours, which is consistent with the theory’s emphasis on structural, relational, and cognitive dimensions of network-based resources [17]. The purpose of this study is to link these dimensions to measurable practices such as source diversity and credibility assessment while also accounting for moderating effects of platform type and content genre. Prior applications of social capital theory in influencer research [12,21] demonstrate its effectiveness in explaining relational dynamics and audience engagement, further supporting its suitability for guiding the hypotheses and methodological approach in this study that investigates how follower size and relational dynamics shape information behaviours across platforms and content genres.

Building on Bourdieu’s notion of symbolic capital, this study extends Social Capital Theory to digital influencers. In this context, follower size represents not only the structural ties but also visible recognition that can be performed. Influencers may use ‘diverse sourcing’ as a form of symbolic capital, a performance of credibility and professionalism and the epistemic practice of verification. This addition strengthens the bridge between network-based resources and credibility in online environments.

2.2. Influencer Follower Size and Source Diversity

Follower size is widely acknowledged as shaping influencer behavior and strategic communication. As followers increase, influencers face audience expectations and more public scrutiny, motivating them to diversify the sources they use to maintain trust and authenticity [22]. This aligns with classical persuasion frameworks [23], which emphasize the importance of credible sources in influencing opinion change and message acceptance. Influencers with smaller follower sizes may rely on fewer or less varied sources due to their lower visibility and less public pressure, while those with larger audiences tend to adopt more rigorous sourcing strategies to avoid loss of credibility and to appeal to diverse follower groups [2].

Wies, Bleier, and Edeling [24] identify a ‘Goldilocks’ zone where follower count balances engagement and content quality, highlighting that follower size not only affects reach but also impacts content strategy and source selection. This is further supported by He and Tian [25], who highlight that influencers with large followings tend to be more aware of reputational risks and thus diversify their sources to meet followers’ demands for trustworthy content.

Additionally, meta-analyses reveal non-linear relationships between follower size and influence effectiveness, which suggest that the complexity of source diversity may increase disproportionately with follower size [7]. Influencers navigate this complexity by incorporating diverse types of information, blending personal experience, expert opinion, and third-party verification to support credibility and follower engagement.

2.3. Diverse Sources: Definition, Importance, and Influencer Practices

Diversity of sources refers to the variety and heterogeneity of information inputs that influencers utilize when creating content. It covers differences in origin (peer, expert or user-generated), format (text, video, image), and perspective (neutral, critical, promotional) [25]. For influencers, employing diverse sources enhances message richness, mitigates bias, and supports authenticity, which are vital in maintaining long-term follower trust in an environment saturated with misinformation [2,4].

Research by Lou and Yuan [1] demonstrates that messages with diverse, credible sources prompt higher consumer trust and stronger purchase intentions. Similarly, Saima and Khan [26] highlight the mediating role of credibility in the influence of source diversity on consumer behaviour in influencer marketing contexts. The strategic use of diverse sources can also reduce cognitive dissonance in followers by providing multiple viewpoints, highlighting the influencer’s role as a reliable content gatekeeper [27].

Practically, influencers employ diverse sources ranging from brand-provided content, independent expert reviews, personal testimonials, to follower feedback [19,28]. This blend enriches accounts and satisfies varying follower expectations, therefore increasing engagement and authenticity that is perceived. However, the ability and willingness to integrate such sources often depend on follower size and platform norms [29].

Overall, source diversity acts as a quality signal that can distinguish genuine influencers from ‘fake’ or less credible ones, which is particularly important given the prevalence of fake followers and bots [11]. Consequently, source diversity has emerged as a critical area of interest in understanding influencer effectiveness.

Hence, this study proposes that social capital, manifested through follower size, shapes influencers’ need to use diverse information sources. Based on this, we developed the following hypothesis:

H1: 

Influencers with larger follower sizes are more likely to use more diverse sources than those with smaller sizes.

2.4. Credibility Assessment and Its Role in Influencer Marketing

Credibility remains a core construct in influencer communication research, operationalized through dimensions such as trustworthiness, expertise, and attractiveness [1,4]. Followers’ credibility assessments influence not only their engagement but also the influencer’s content strategies, especially regarding source selection [22].

Several studies underscore the importance of credibility as a mediator in the follower size–source diversity relationship. For example, Kim et al. [22] argue that as influencers attract larger audiences, followers become more discerning, prompting influencers to enhance the credibility of their sources to maintain trust. Similarly Zhou et al. [30] discuss how fake followers affect perceived influence on social media platforms, and this is echoed by Kaye et al. [31], who show that perceived authenticity significantly boosts credibility perceptions, thereby encouraging influencers to carefully curate diverse and credible sources.

Wang and Weng [32] expand on this by revealing how influencer authenticity directly affects followers’ positive word-of-mouth and trust, which is crucial in digital settings characterized by scepticism. The gatekeeping perspective [26] provides a useful theoretical framework here: influencers act as gatekeepers who control the flow of information and shape follower perceptions through their credibility management, including source diversity.

Empirical findings by Weismueller et al. [28] demonstrate that credible influencers generate more effective advertising endorsements, emphasizing the role of credibility in marketing success. Similarly, Omar AlFarraj et al. [29] find that trustworthiness and expertise are significant predictors of purchase intention in aesthetic dermatology influencer marketing, illustrating the practical importance of credibility in content creation. Greater structural capital (larger follower size) is expected to heighten the rigor of credibility assessment due to the higher reputational stakes and audience expectations; such increased rigor in credibility assessment by influencers should increase the diversity of sources used and hence mediate the follower size and source diversity. Thus, this study hypothesizes that:

H2a: 

Follower size will be positively associated with the rigor of influencer’s credibility assessment.

H2b: 

The rigor of influencer’s credibility assessment will be positively associated with the use of diverse sources.

H2c: 

The rigor of influencer’s credibility assessment will mediate follower size and the use of diverse sources.

The rigor of credibility assessment in this study is conceptualized both as a cognitive process and as a display of professionalism through the actual practice of their own source verification when sharing information. This interpretation integrates Social Capital’s cognitive and symbolic capital perspectives, providing a stronger theoretical foundation for understanding how influencers manage credibility through visible information behaviours.

2.5. Platform Type and Content Genre as Moderators

We also postulate that the relationship between follower size and source diversity is dependent on the social media platform and content genre it uses because each platform comes with different affordances, audience behaviours, and content norms that shape influencer practices [33].

For instance, TikTok’s short-video, entertainment-oriented environment prioritizes creativity and immediacy, limiting room for extensive source diversity but promoting innovative content curation [31]. Instagram, in contrast, supports more polished and curated posts, often with embedded external links and diverse content types, allowing for richer source diversity [6]. YouTube, as a long-form video platform, offers another space for deep dives and detailed source integration [34].

Content genre also influences expectations around source diversity. Beauty and lifestyle influencers may rely heavily on product reviews and expert endorsements, while gaming influencers might source from community feedback and live interactions [35,36]. Information-based content genre is more likely to draw from diverse sources, while entertainment-based content genre may have less need to verify via corroboration of sources. This genre-based difference affects how follower size translates into source diversity.

Prior social capital research [37,38] suggests that the nature of follower–influencer relationships, shaped by platform and genre, moderates credibility perceptions and content sourcing strategies, similar to what Taillon [35] found that closeness between influencers and followers moderates the effectiveness of endorsements, indirectly affecting how influencers select sources. Therefore, we hypothesize:

H3: 

Platform type and content genre of the influencer will moderate the relationship between follower size and influencers’ use of diverse sources.

Figure 1 illustrates our mediated and moderated framework of influencer size, rigor of credibility assessment and diversity of sources. Grounded in Social Capital Theory, the framework positions follower size as the independent variable, affecting influencers’ diverse sources both directly and indirectly through credibility assessment, which serves as a mediator. Additionally, the model includes platform type and content genre as moderators, affecting the strength of the direct relationship between follower size and sourcing practices. This structure reflects the study’s aim to understand how social capital, mediated credibility evaluation, and contextual factors shape influencers’ information sourcing behaviour across different online environments.

To summarize, our overall research question for this study is:

How does influencers’ follower size affect credibility assessment rigor and diversity of sources? Do platform type and content genre of the influencers moderate such information practices?

3. Research Methods and Materials

This study employed a quantitative survey design to investigate the relationships among influencer follower size, credibility assessment, platform type, content genre, and the use of diverse sources in influencer content creation. The analysis utilized a large, global dataset collected via an online survey conducted in 2024 with 500 social media influencers. These influencers ranged from nano-influencers (1000–10,000 followers) to mega-influencers (more than 1 million followers), enabling comprehensive coverage of various influencer scales and types.

3.1. Sampling and Participant Recruitment

The anonymous global online survey was conducted, ensuring confidentiality and compliance with ethical research standards. Questions were all close-ended and objectively worded (non-loaded) to facilitate ease of completion. Such design aimed to encourage candid answers from the respondents to avoid social desirability bias. The project received prior approval from the researchers’ university institutional review board before implementation. Participants provided informed consent before completing the questionnaire.

Participants were recruited through an international online panel of a reputable research company, Qualtrics, which offers access to a broad pool of potential respondents. Compensation for participation was provided according to Qualtrics’ standard incentive policies. The survey applied a 50/50 gender quota to achieve gender balance among respondents.

The survey targeted active social media influencers who regularly create public content across major platforms such as Instagram, TikTok, YouTube, and others. Several screener questions were used to ensure the respondent is a qualified influencer eligible for the study, including a minimum follower count of 1000, creating and posting content on a regular basis for audiences other than their friends and family on their main social media platform, with years of experience creating public content, before the respondent can continue the survey. Recruitment employed a quota sampling approach to ensure representation that is diverse across geographic regions and gender. However, we did not set quota by follower size to obtain a representative result of follower size distribution of influencers globally. The survey collected detailed data on demographics (age, gender, education), influencer experience (years of content creation), follower size, content genre, and key variables of interest, such as credibility assessment and use of diverse sources.

To maximize cultural and linguistic inclusivity, the survey instrument was first developed in English and then professionally translated into seven additional languages: Arabic, Chinese, French, German, Portuguese, Russian, and Spanish. These translations were vetted by native communication scholars for accuracy and cultural relevance. Influencers from 44 countries across six continents participated, with 52.4% hailing from Global South countries, classified per UNESCO’s designations.

The geographic quotas allocated for the survey reflected linguistic and economic considerations. English-speaking Global North countries (e.g., United States, United Kingdom, Canada, Australia, Singapore, and Ireland) were assigned a larger sample (100 participants each). English-speaking Global South regions (e.g., India, parts of Africa, Pacific Islands, Southeast Asia) contributed 50 participants. Other language regions (French, Spanish, Portuguese, German, Russian, Arabic, and Chinese) were each assigned 50 participants, ensuring a balanced global distribution.

3.2. Measurements

Follower size was operationalized into four categories: nano-influencers (1000–10,000 followers), micro-influencers (10,000–100,000 followers), meso-influencers (100,000–1 million followers), and mega-influencers (more than 1 million followers). Due to uneven distribution of follower size levels, with much fewer respondents being larger influencers (only 8 mega influencers and 27 macro influencers), these categories were collapsed into two groups: smaller influencers (1000–10,000 followers, n = 340) and larger influencers (10,001–1,000,000 followers, n = 160) for analysis and comparison.

Content genres were classified based on eleven categories commonly identified in influencer marketing [39]: sports and fitness, gaming, comedy, beauty, fashion/lifestyle, parenting, animal and nature, photography, travel and food, shopping/product reviews, and current affairs/politics/economy. For analytical purposes, these were consolidated into three overarching groups: entertainment only, information only, and mixed content creators.

Source diversity was measured using a multiple-response item that asked participants to identify the types of information sources they regularly use when creating content. The item stated: “What sources do you use to create your content? (Check all that apply)” and presented a checklist of seven options, including: (1) online sources not from mainstream media, (2) personal experience or encounter, (3) mainstream news media, (4) government sources, (5) self-conducted investigations or expert interviews, (6) tips or leads from followers and friends, and (7) an open-ended “other” category. Participants were instructed to select all options that applied. Each selected source was coded as 1, and unselected options as 0. The final source diversity score was calculated by summing the total number of selected options, yielding a continuous measure ranging from 0 to 7, with higher scores indicating greater diversity in content sourcing practices.

Credibility assessment rigor was measured using a single behavioural item that asked respondents to indicate how they verify content before sharing it with their audience. The item stated: “What steps do you take before sharing content with your audience? Select which one best describes your practice.” Respondents selected one of four ordinal response options reflecting increasing effort of the influencer’s credibility-checking behaviour: (1) “Just share the content I found entertaining or useful without checking for accuracy,” (2) “As long as I trust the source/creator of the content, I will share the content,” (3) “I only check for accuracy for news content, all other content I don’t check,” and (4) “Check the source of information from fact-checking sites or look up other authoritative sources and only disseminate information that is verified.”

Each response was coded from 1 to 4 by ranking the effort and rigor, with higher scores indicating greater rigor and effort in systematic credibility assessment. The use of a single-item behavioural measure aligns with methodological literature indicating that single-item scales are acceptable when constructs are unidimensional, clearly defined, and narrow in scope [40].

To validate this single-item measure, we examined face validity and predictive validity by testing the correlation between the item and our dependent variable (diversity of sources). The items in the assessment rigor scale met these face validity criteria of relevance, ease of response, no ambiguity, non-distressing or sensitive and not judgmental. In addition, it also meets the predictive validity using the item in the result as hypothesized: significant positive correlation between credibility rigor and use of diverse sources (B = 0.3510, SE = 0.0464, t = 7.5581, p < 0.001).

Demographic variables included age measured in standard intervals, gender (male, female, self-described), education level, and years of content creation experience categorized into four intervals: less than one year, 1–3 years, 3–10 years, and more than 10 years.

3.3. Data Analysis

All analyses were conducted using IBM SPSS Software (Version 29) and Hayes PROCESS macro. The analytic process began with data cleaning and screening for missing values, outliers, and assumptions of normality. Descriptive statistics were done to summarize sample characteristics and distribution patterns of key variables. An independent samples t-test was employed to compare source diversity between influencers with smaller and larger follower sizes. When the assumption of equal variances was violated, Welch’s t-test was used. Additionally, linear regression analyses were conducted to examine the direct effects of follower size and credibility assessment on influencers’ use of diverse sources. To assess conditional and indirect effects simultaneously, moderated mediation analysis was conducted using PROCESS Macro Model 9. This model allowed for the examination of whether credibility assessment served as a mediator in the relationship between follower size and source diversity, and whether this relationship was moderated by platform type and content genre. A bootstrap approach with 5000 resamples and 95% bias-corrected confidence intervals was used to evaluate the significance of indirect and interaction effects. Variables involved in interaction terms were mean-centred to reduce multicollinearity. Statistical significance was evaluated at an alpha level of 0.05. To test for common method bias, a collinearity test was conducted and all the predictor variables’ VIF were low (1.0–1.3), much lower than the 3.3. threshold of concern. This analysis aligned with the study’s theoretical framework, grounded in Social Capital Theory, and was designed to test both direct and moderated mechanisms influencing influencers’ sourcing behaviour.

4. Findings

4.1. Demographic Statistics of the Respondents

The study’s sample comprised 500 respondents with a diverse age distribution. The largest age group was between 25 and 34 years old (33.4%), followed by those aged 35 to 44 years (28.4%). Younger participants aged 18 to 24 years accounted for 19.4%, while those aged 45 to 59 years made up 14.8%. Only 4.0% were aged 60 or older. This age composition suggests that the study primarily reflects the perspectives of young and middle-aged adults, a demographic often most active in digital content creation and social media engagement. See Table 1 for details of the sample demographics.

By quota design, gender distribution was nearly equal, with male respondents representing 49.6% and female respondents slightly higher at 50.2%. Only one participant (0.2%) preferred to self-describe their gender. In terms of education, most respondents were highly educated, with 42.2% holding a bachelor’s degree and 23.4% possessing a master’s degree or graduate certificate. A smaller proportion had completed some college or an associate degree (18.2%), while 13.6% had a high school education or below. Only 2.4% held a PhD, JD, or equivalent. This indicates a predominantly well-educated sample, which could influence attitudes toward sourcing and credibility assessment in media contexts.

Geographically, the largest representation came from Europe (35.2%) and Asia (29.4%), followed by South America (13.0%) and North America (10.0%). Smaller proportions were from Africa (6.4%) and Oceania (3.0%). The wide geographic spread suggests a globally diverse sample, with a heavy concentration in Europe and Asia.

In terms of content creation experience, 45.2% had been creating public content for 1 to 3 years, 36.0% for more than 3 years but less than 10 years, and 6.6% for 10 years or more. Only 12.2% had less than one year of experience. This shows that most respondents are experienced content creators, with a significant portion having sustained engagement over several years.

For income sources, 31.4% reported that digital content was their main source of income, while the majority (68.6%) did not rely on it as their primary income. Interestingly, most respondents perceived themselves as having at least some influence over their audience, with 69.4% answering “yes, sometimes” and 18.6% stating “yes, always,” while only 12.0% reported not influencing their audience.

When examining the types of content produced, the most common categories were beautiful and attractive things (18.0%), entertainment including memes (15.6%), and recommendations/vlogs (13.8%). Other popular formats included blogs/online commentaries (13.4%) and demonstrations/tutorials (12.6%). Less common were news videos (8.8%), video clips curated from different sources (7.0%), documentaries (3.8%), short fictional films or stories (3.6%), and “other” types of content (3.4%). This distribution reflects a preference for visually engaging and entertainment-oriented formats, though informational and instructional content also plays a substantial role.

Table 1. Demographic characteristics of the respondents.

Demographic Statistics	Category	Frequency	Percentage
Respondents Age	Between 18–24 years old	97	19.4
	Between 25–34 years old	167	33.4
	Between 35–44 years old	142	28.4
	Between 45–59 years old	74	14.8
	60 and older	20	4.0
	Total	500	100.0
Gender	Male	248	49.6
	Female	251	50.2
	Prefer to self-describe	1	0.2
	Total	500	100.0
Level of Education	High school or below	68	13.6
	Some college/associate degree	91	18.2
	Bachelor’s degree	211	42.2
	Master degree or Graduate Certificate	117	23.4
	PhD/JD or equivalent	12	2.4
	Total	499	99.8
Continent	Asia	147	29.4
	Europe	176	35.2
	Oceania	15	3.0
	North America	50	10.0
	South America	65	13.0
	Africa	32	6.4
	Total	500	100.0
Do you have a journalistic background	No	393	78.6
	Yes	105	21.0
	Total	498	99.6
How long have you created public content on social media	Less than one year	61	12.2
	1 to 3 years	226	45.2
	More than 3 years but less than 10 years	180	36.0
	10 years or more	33	6.6
	Total	500	100.0
Is digital content your main source of income	No	343	68.6
	Yes	157	31.4
	Total	500	100.0
Influential on audiences	No	60	12.0
	Yes, sometimes	347	69.4
	Yes, always	93	18.6
	Total	500	100.0
Content genre	News videos	44	8.8
	Demonstrations/tutorials	63	12.6
	Recommendations/Vlogs	69	13.8
	Blogs/online commentaries	67	13.4
	Video clips curated from different sources	35	7.0
	Documentaries	19	3.8
	Entertainment including memes	78	15.6
	Beautiful and attractive things	90	18.0
	Short fictional films or stories	18	3.6
	Other (please specify)	17	3.4
Source Type (multiple response)	Online sources	184	36.8
	Personal experience	290	58.0
	Mainstream news media	184	36.8
	Government	63	12.6
	Own investigation & interviews	193	38.6
	Tips & leads from followers	147	29.4
	Other	3	0.6
Follower Size	1–10 K	340	68.0
	10–100 K	125	25.0
	100 K to 1 million	27	5.4
	More than 1 million	8	1.6
	Total	500	100.0

Table 1. Demographic characteristics of the respondents.

Demographic Statistics	Category	Frequency	Percentage
Respondents Age	Between 18–24 years old	97	19.4
	Between 25–34 years old	167	33.4
	Between 35–44 years old	142	28.4
	Between 45–59 years old	74	14.8
	60 and older	20	4.0
	Total	500	100.0
Gender	Male	248	49.6
	Female	251	50.2
	Prefer to self-describe	1	0.2
	Total	500	100.0
Level of Education	High school or below	68	13.6
	Some college/associate degree	91	18.2
	Bachelor’s degree	211	42.2
	Master degree or Graduate Certificate	117	23.4
	PhD/JD or equivalent	12	2.4
	Total	499	99.8
Continent	Asia	147	29.4
	Europe	176	35.2
	Oceania	15	3.0
	North America	50	10.0
	South America	65	13.0
	Africa	32	6.4
	Total	500	100.0
Do you have a journalistic background	No	393	78.6
	Yes	105	21.0
	Total	498	99.6
How long have you created public content on social media	Less than one year	61	12.2
	1 to 3 years	226	45.2
	More than 3 years but less than 10 years	180	36.0
	10 years or more	33	6.6
	Total	500	100.0
Is digital content your main source of income	No	343	68.6
	Yes	157	31.4
	Total	500	100.0
Influential on audiences	No	60	12.0
	Yes, sometimes	347	69.4
	Yes, always	93	18.6
	Total	500	100.0
Content genre	News videos	44	8.8
	Demonstrations/tutorials	63	12.6
	Recommendations/Vlogs	69	13.8
	Blogs/online commentaries	67	13.4
	Video clips curated from different sources	35	7.0
	Documentaries	19	3.8
	Entertainment including memes	78	15.6
	Beautiful and attractive things	90	18.0
	Short fictional films or stories	18	3.6
	Other (please specify)	17	3.4
Source Type (multiple response)	Online sources	184	36.8
	Personal experience	290	58.0
	Mainstream news media	184	36.8
	Government	63	12.6
	Own investigation & interviews	193	38.6
	Tips & leads from followers	147	29.4
	Other	3	0.6
Follower Size	1–10 K	340	68.0
	10–100 K	125	25.0
	100 K to 1 million	27	5.4
	More than 1 million	8	1.6
	Total	500	100.0

4.2. Follower Sizes and Diversity of Sources Used

An independent sample t-test was conducted to evaluate whether influencers with larger follower sizes used more diverse sources compared to those with smaller follower sizes. The results revealed a significant difference in the diversity of sources used between the two groups.

Descriptive statistics in Table 2 indicated that influencers with smaller followings (n = 340) had a lower mean diversity score (M = 2.01, SD = 1.014) compared to those with larger followings (n = 160), who reported a higher mean diversity score (M = 2.365, SD = 1.368).

Levene’s test for equality of variances was significant, F (1, 498) = 23.45, p < 0.001, indicating that the assumption of equal variances was violated. Therefore, Welch’s t-test results were used. The test showed a statistically significant difference in source diversity, t (247.81) = −2.75, p = 0.003 (two-tailed), 95% CI [−0.61, −0.10].

Table 2. Independent samples t-test for Influencer follower sizes by diverse sources.

Following (n = 500)	N	M	SD	t	df	p
Smaller size	340	2.014	1.0147	−2.75	247.814	0.003
Larger size	160	2.368	1.3688

Table 2. Independent samples t-test for Influencer follower sizes by diverse sources.

Following (n = 500)	N	M	SD	t	df	p
Smaller size	340	2.014	1.0147	−2.75	247.814	0.003
Larger size	160	2.368	1.3688

The significant difference in source diversity between influencers with smaller and larger followings suggests that social capital, as indexed by follower size, plays a pivotal role in diversifying the information sources. To assess whether follower size predicts influencer’s diversity of source use, a regression was run with diverse sources as the outcome and follower sizes as the predictor. The analysis showed a statistically significant effect, B = 0.3517, SE = 0.1123, t = 3.1331, p = 0.0018, 95% CI [0.1311, 0.5723], indicating that influencers with larger follower sizes are significantly more likely to use a diverse range of information sources. These results support H1, affirming that influencers with larger follower sizes engage in more diverse sourcing practices, consistent with expectations according to Social Capital Theory.

Influencers with larger followings reported significantly higher use of diverse sources, supporting the notion that increased visibility and reputational risk incentivize broader information sourcing. This aligns with Social Capital Theory, which posits that individuals embedded in larger networks are more likely to access and utilize institutional resources to maintain trust (Putnam, 2000 [41]).

4.3. Follower Size and Credibility Assessment

To test whether follower size is positively associated with rigor of credibility assessment, a multiple regression analysis was conducted with credibility assessment as the outcome variable and follower size, content genre, platform type, and their interaction terms as covariates. The model summary indicated a low explained variance, R² = 0.0069, F (5, 476) = 0.663, p = 0.652, suggesting that the overall model was not significant. Specifically, the coefficient for follower size was not statistically significant, B = −0.6931, SE = 0.5366, t = −1.29, p = 0.197, 95% CI [−1.75, 0.36]. Therefore, H2a was not supported. Larger follower size was not associated with higher rigor of credibility assessment or fact-checking behaviours (

Table 3. Summary of Moderation Analysis of Covariates on Follower Size and Rigor in Credibility Assessment.

Predictor	B	SE	t	p	95% CI LL	95% CI UL
Constant	3.6657	0.7780	4.7117	<0.001	2.1370	5.1944
Follower Size (FSZ)	−0.6931	0.5366	−1.2916	0.1971	−1.7476	0.3613
Content Genre	−0.1082	0.3122	−0.3467	0.7290	−0.7216	0.5052
FSZ × Genre	0.1430	0.2229	0.6413	0.5216	−0.2951	0.5810
Platform Type	−0.3509	0.3504	−1.0012	0.3172	−1.0395	0.3377
FSZ × Platform Type	0.2129	0.2410	0.8837	0.3773	−0.2605	0.6864

R = 0.0832, R² = 0.0069, MSE = 1.2906, F (5, 476) = 0.663, p = 0.652.

).

4.4. Rigor of Credibility Assessment and Use of Diverse Sources

A bivariate regression analysis was employed to test whether rigor of credibility assessment predicts diversity of sources. The model was statistically significant, R² = 0.1189, F (2, 479) = 32.33, p < 0.001. The unstandardized coefficient for credibility assessment was B = 0.3510, SE = 0.0464, t = 7.5581, p < 0.001, 95% CI [0.2597, 0.4423], indicating a strong positive relationship. Thus, H2b is supported. Influencers who engaged more rigor in credibility assessment also tended to use more diverse sources.

4.5. Mediation Analysis

A mediation analysis was conducted using Hayes PROCESS Macro to determine whether credibility assessment mediated the relationship between follower size and diversity of sources. The indirect effect of follower size on diversity of sources through credibility assessment was not statistically significant. The estimated indirect effect was B = −0.073, Boot SE = 0.051, and the 95% bootstrap confidence interval [−0.173, 0.019] included zero, indicating no evidence of mediation. The direct effect of follower size on diversity of sources remained significant (B = 0.3517, SE = 0.1123, t = 3.13, p = 0.0018), suggesting that follower size influences diversity of sources independently of rigor of credibility assessment.

Hence, H2c was not supported. Although both follower size and rigor of credibility assessment were independently associated with diversity of sources, rigor of credibility assessment did not explain the link between follower size and diversity of sources, indicating follower size does not necessarily lead to higher rigor of credibility assessment.

4.6. Moderated Mediation Analysis

To examine whether the relationship between follower size and the use of diverse sources was moderated by platform type or content genre, we employed Hayes’ PROCESS Model 9. The interaction between follower size and platform type on rigor of credibility assessment was not statistically significant, B = 0.2129, SE = 0.2410, t = 0.88, p = 0.3773. Likewise, the interaction between follower size and content genre was also not significant, B = 0.1430, SE = 0.2229, t = 0.64, p = 0.5216. Both moderation indices were insignificant, and the change in R² for the interaction terms was minimal (∆R² < 0.003), indicating negligible explanatory power.

Although the coefficients are in the expected direction, the small effect sizes and lack of significance suggest that influencer practices in rigor of credibility assessment may be relatively consistent across different platforms and content genres. This convergence of practice across platforms could imply that broader norms or professional pressures—such as reputational concerns or algorithmic visibility might be standardizing how influencers approach source diversity, irrespective of the content type or platform they operate on. In other words, platform affordances and genre distinctions may no longer strongly differentiate source evaluation practices among influencers.

Therefore, Hypothesis 3 (H3), which predicted moderation by platform type and content genre, was not supported. This lack of moderation of follower size in shaping rigor of credibility assessment indicates that other factors (e.g., education, audience expectations, or monetization goals) may play a more central role in how influencers vet and diversify their sources. Table 4 presents a summary of the moderated mediation analysis.

Table 4. Summary of Moderated Mediation Analysis of Diverse Sources.

Path	β	SE	t	p	95% CI (LL)	95% CI (UL)
Follower size → Credibility assessment	−0.6931	0.5366	−1.29	0.197	−1.7476	0.3613
Credibility assessment→Diverse sources	0.3510	0.0464	7.56	<0.001	0.2597	0.4423
Follower sizes→Diverse sources	0.3517	0.1123	3.13	0.0018	0.1311	0.5723
Follower size × Genre → Credibility assessment	0.1430	0.2229	0.64	0.522	−0.2951	0.5810
Follower size × Platform type → credibility assessment	0.2129	0.2410	0.88	0.377	−0.2605	0.6864

Note: Dependent Variable = “Diverse sources”. Boldface indicates statistically significant result.

Table 4. Summary of Moderated Mediation Analysis of Diverse Sources.

Path	β	SE	t	p	95% CI (LL)	95% CI (UL)
Follower size → Credibility assessment	−0.6931	0.5366	−1.29	0.197	−1.7476	0.3613
Credibility assessment→Diverse sources	0.3510	0.0464	7.56	<0.001	0.2597	0.4423
Follower sizes→Diverse sources	0.3517	0.1123	3.13	0.0018	0.1311	0.5723
Follower size × Genre → Credibility assessment	0.1430	0.2229	0.64	0.522	−0.2951	0.5810
Follower size × Platform type → credibility assessment	0.2129	0.2410	0.88	0.377	−0.2605	0.6864

Note: Dependent Variable = “Diverse sources”. Boldface indicates statistically significant result.

5. Discussion

Overall, the results showed mixed support for the hypotheses, revealing a consistent relationship between follower size and source diversity, but not between follower size and credibility-assessment rigor.

This study examined how follower size relates to influencers’ use of diverse information sources and how this influences the rigor of their credibility assessment when sharing information. The findings provide insights into how influencers curate and verify content for their audiences. First, the finding that influencers with larger follower counts tend to use a greater variety of sources supports the Social Capital Theory that follower size correlates with access to broader networks and resources. This may be due to partnerships, higher stakes in audience engagement, or a professionalization of content practices. It shows that visibility and credibility online of these influencers are not only tied to popularity metrics but also to how influencers substantiate their messages with different types of information sources, with a mix of personal experience/interviews and official sources.

One main discovery we found is that follower size was not associated with higher rigor of credibility assessment. Instead, influencer uses popularity metrics rather than content validation to determine source credibility when sharing their content with their followers. However, this relationship was nuanced by the finding that credibility was also associated with source diversity. This suggests that influencers try to increase their credibility through displaying the variety of information sources they use rather than the rigorous validation of each individual information source, which is not a process seen by the public.

This finding is consistent with prior literature emphasizing the role of credibility heuristics in shaping information sourcing [4]. In the context of influencers, those who prioritize accuracy and source validation appear to engage in more varied content curation strategies, possibly to cross-check facts or appeal to a broader audience. This aligns with the argument that credibility-conscious influencers adopt practices like journalistic gatekeeping [26], thereby increasing their reliability in the eyes of followers and brands. But there are also influencers who use the routine of source variety in lieu of rigorous checking to boost their credibility. Moreover, the result contributes to the influencer credibility literature by empirically linking cognitive evaluation processes (credibility checks) to behavioural sourcing practices, an underexplored but increasingly relevant area in the fight against misinformation. Diverse sources will be beneficial to increase information accuracy when the content creator uses diverse sources as a checking tool, with an increase in rigor in assessing credibility of information before sharing with others. When the diverse sources are just for display and credibility heuristics rather than cross-checking and are not used as evidence of corroboration, then they may not be helpful to increase quality and accuracy of information.

Third, platform type and content genre were not found to moderate the relationship between follower size and the use of diverse sources. Additionally, the results did not find any correlation between diversity of sources and content genre. This suggests that regardless of the type of content influencers produce, be it political commentary, lifestyle vlogs, or entertainment, the breadth of their information sources does not vary significantly. One possible explanation is that influencers across genres have begun adopting similar sourcing practices, potentially shaped by shared norms, algorithmic pressures, or professionalization of content creation across platforms. This convergence may reflect a broader shift in influencer culture where source diversity is not necessarily conditioned by content type but rather by perceived audience expectations and reputational considerations.

The finding corroborates prior research showing that macro-influencers rely more heavily on credible and institutional sources to uphold their legitimacy [42,43] and reflects earlier assertions by Kim et al. [22] that larger influencers adopt more elaborate sourcing strategies as part of strategic brand management. However, it is important to note that using diverse types of sources does not necessarily imply rigorous verification. Macro-influencers may diversify their sources primarily to reduce reputational risk, as diversity is easier to demonstrate in content creation than the less visible process of careful fact-checking. Verification before sharing requires specialized knowledge, time, and caution, which may not be consistently applied despite a broad sourcing profile.

Furthermore, the results suggest that the link between follower size and sourcing diversity is consistent across various content genres and platform types. This finding is different from Kim et al. [22], who found that genre and platform affordances (e.g., YouTube vs. Instagram) influence influencer strategies. One possible explanation is that influencer professionalism and sourcing practices have converged across platforms, perhaps due to shared norms driven by algorithmic visibility or cross-platform branding. Alternatively, the insignificant effect may indicate that influencer content types have diversified within genres, weakening any consistent moderating influence. These insights caution against overgeneralizing platform-specific strategies in influencer research.

Theoretically, this study contributes to the development of Social Capital Theory in digital environments. Traditionally, social capital refers to the resources individuals gain through social networks, trust, and norms of reciprocity [15,16,17]. In the context of social media, influencers accumulate social capital through their relationships with followers and the trust embedded in their content. However, the result that influencers with larger follower size not assessing with more rigor in information credibility assessment contradicts the prediction from Social Capital Theory, which suggests that individuals rooted in larger social networks (e.g., macro-influencers) are more incentivized to preserve their public image and thus engage in more rigorous information vetting [41]. The current findings suggest that follower size alone may not determine whether influencers scrutinize content for credibility. This discrepancy may reflect the evolving dynamics of influencer culture, where micro-influencers, due to their close follower relationships, may feel equally compelled to verify information. Alternatively, the visibility of credibility assessment behaviour may be more influenced by platform norms or influencer personality traits than by mere audience size.

The findings show that diverse sourcing serves as a symbolic form of capital for influencers; it communicates credibility, responsibility, and transparency, all of which strengthen the influencer-follower relationship. Follower size alone does not guarantee trust; rather, influencers may feel that audiences evaluate them by their reach (follower size), with sourcing efforts (diverse sources). This reinforces the idea that online influence is shaped by both structural capital (follower networks) and cognitive capital (perceived expertise and trustworthiness). Hence, building on Bourdieu’s concept of symbolic capital, source diversity can be understood as a performative signal of legitimacy. It functions similarly to how journalists cite institutional sources to reinforce authority; yet in influencer culture, it can also mask a lack of actual verification. This suggests a hybrid form of symbolic capital that blends informational cues with aesthetic and strategic branding. Future research could explore how audiences decode such signals, whether they perceive diversity of sources as a proxy for trustworthiness, or whether they differentiate between genuine verification and performative credibility.

This unexpected result in this study invites a reconsideration of how Social Capital Theory (SCT) functions in decentralized, algorithm-driven media environments. While traditional SCT emphasizes that individuals with larger networks are more likely to protect reputational capital by behaving responsibly [41], the influencer economy rewards visibility and performance, not necessarily epistemic rigor. In digital spaces, attention often functions as capital in itself [34], and content performance may overshadow the value of source accuracy. Therefore, the non-significant link between follower size and credibility checking may suggest that structural social capital (i.e., large audiences) does not always translate into normative behaviours like fact-checking. This divergence highlights the need to adapt or augment SCT to account for new forms of symbolic and algorithmic capital within influencer ecosystems.

By highlighting the relationship between content practices (like sourcing) and follower trust, the study expands our understanding of how influencers employ their symbolic resources and maintain them in the digital age. It bridges traditional theories of social capital with emerging forms of digital labour and influence.

The findings offer several practical takeaways for influencers, marketers, and platform developers. First, influencers seeking to build or maintain credibility should be intentional about citing a wide range of reliable sources, especially when addressing topics beyond personal experience.

Second, brands and marketers looking to partner with influencers may consider evaluating not only follower metrics but also sourcing practices as indicators of thoughtfulness and professionalism. Collaborating with influencers who consistently use diverse sources may result in higher engagement and trust from audiences.

Lastly, for influencers, platforms, and marketers, these findings offer several tangible takeaways. First, platforms such as Instagram, TikTok, or YouTube could implement source-tagging features enabling creators to link, cite, or annotate their content with source information in a seamless interface. This not only promotes transparency but could enhance platform trustworthiness at scale. Second, influencer marketing agencies may consider including source verification behaviour as part of their vetting criteria when selecting partners for social campaigns, especially in health, news, or educational domains. Some labels of rigorous information checking, such as “fact-checked”, can be added to social media posts. Influencers who post more “fact-checked” posts should be rewarded by marketers. Third, training modules or certifications in digital literacy and source vetting standards (like those similar to media literacy programs for journalists) could be designed to support influencers in improving content accuracy while maintaining engagement.

6. Limitations and Suggestions for Future Research

While this study offers meaningful insights into the information-sourcing behaviour of influencers, certain limitations must be acknowledged. First, the reliance on self-reported survey data may be subject to self-report bias, where participants potentially overstate the rigor of their sourcing practices or credibility evaluations. Second, although the study included a global sample, regional and cultural nuances in influencer behaviour and audience expectations may not be fully captured, as we only obtained the perspectives of influencers. Future research could disaggregate results by region, language, or influencer subcultures to better understand context-dependent practices.

Likewise, the study used a single-item measure to capture credibility assessment rigor, which may not fully reflect the multidimensional nature of the construct. However, credibility assessment practice rigor as a behavioural practice measure with clear face validity, rather than attitudinal or behavioural intention measure, a single-item measure can be justified. As there is a very uneven sample size in follower sizes, we cannot compare specifically each category of influencers, from small to mid-size to mega-influencers. Some content types were also collapsed into broader categories, potentially masking nuanced differences in how influencers engage with different types of sources. Further, participants were self-selected and may represent a more engaged or credibility-conscious subset of influencers, limiting their generalizability.

In addition, integrating multiple methods such as content analysis of influencer posts or experiment could illuminate how influencers internally process credibility and whether stated reported behaviour aligns with actual behaviour. Exploring audience perspectives would also be valuable, especially in understanding how users interpret signals of credibility and source diversity and how that impacts trust and engagement. Longitudinal designs could further reveal whether influencers who consistently cite diverse or credible sources enjoy long-term benefits in audience loyalty or brand partnerships.

Furthermore, examining the audience’s and influencers’ perceived value of different source types (e.g., peer-generated, institutional, journalistic, or academic) across genres and platforms could unpack the symbolic capital attached to certain sources and how such signals are decoded by varying audiences.

7. Conclusions

This study contributes to the growing body of research on influencer credibility by highlighting a key paradox at the heart of digital influence: while larger follower size is associated with greater diversity in information sources, it does not necessarily correlate with deeper credibility assessment. This disconnect suggests that sourcing diversity may serve more as a performative signal of credibility than as a genuine strategy for information verification. Influencer research from the creator’s perspective, especially using quantitative approach like this study, is very rare in the field, as it is very difficult to recruit influencers to participate in a survey study, and most influencer studies are single-country studies [9]. This global study made a unique contribution with a large sample of influencers, enabling statistical analysis.

Theoretically, our study challenges core assumptions of Social Capital Theory that individuals embedded in large social networks (macro-influencers) are more likely to safeguard their reputational capital through rigorous content vetting. Our findings suggest that visibility and scale may drive influencers to diversify content to meet expectations for transparency, but not always to engage in deeper fact-checking or validation. In this context, source diversity may function as a form of symbolic capital, a visible performance of credibility used to maintain audience trust, rather than a reflection of actual epistemic diligence.

The absence of significant moderating effects by platform type and content genre further reinforces the idea that professionalization and algorithmic convergence have standardized influencer behaviours across digital spaces. Regardless of niche or medium, influencers appear to adopt similar sourcing strategies, likely shaped by shared norms and branding logics. This finding both supports and complicates earlier scholarship suggesting that genre and platform affordances shape influencer practice. Convergence may override differentiation by genres and platforms.

Practically, our results offer several implications. Influencers seeking to build or sustain trust should prioritize not only the breadth of sources but also their depth, including verifiability and quality. Marketers and brands evaluating influencer partnerships may consider credibility evaluation habits as an important indicator of long-term reputational alignment. Platforms can play a role by developing tools that facilitate source transparency, such as embedded citation tags, credibility scoring systems, or verification prompts at the point of content creation. Rather than generic calls for quality, such tools can reinforce epistemic responsibility in everyday content production.

Finally, this study reconceptualizes influencers not merely as cultural intermediaries or entertainers, but as information gatekeepers whose sourcing behaviours and credibility assessments negligibly but powerfully shape digital discourse. Popularity and professionalism may drive performance, but genuine trust may hinge on practices that extend beyond display toward verification, intentionality, and responsibility in content curation.