Worldwide connections of micro-influencers who promote e-cigarettes on Instagram and TikTok: a social network analysis. (Preprint)

Abstract

Exposure to e-cigarette marketing on social media is a risk factor associated with e-cigarette use among youth. Tobacco brands have been relying on influencers to promote e-cigarettes on social media; however, influencer marketing has not been sufficiently studied.

This study explored network connections and interactions among micro-influencers (influencers with ~10,000-100,000 followers) and their audiences on social media platforms popular among youth: i.e., Instagram and TikTok.

We constructed directed unipartite networks among Instagram (N = 104) and TikTok (N = 100) micro-influencers and users on Instagram (N = 55,622) and TikTok (N = 68,673) who commented on influencers’ posts (including micro-influencers who commented on each other’s posts). We calculated individual node measures: in/out degree and betweenness centrality, and network measures: density, reciprocity, transitivity, and assortativity for Instagram and TikTok networks. Using Kruskal-Wallis and post-hoc Dunn tests, we then compared centrality measures between and within Instagram and TikTok networks based on geographic regions of the micro-influencers (North America, South America, Asia, and Europe) and content micro-influencers posted about (e-cigarette alone versus e-cigarette alongside lifestyle contexts).

The Instagram network was denser (more interconnected) and active compared to TikTok. The Instagram network had 1.48 times as high density, 281 times as high transitivity, and 85 times as high reciprocity as the TikTok network. Micro-influencers on Instagram had 2.5 times as high median indegree or the number of incoming ties (Median = 286) as micro-influencers on TikTok (p-value = .008), higher median outdegree or the number of outgoing ties (Median = 5) than micro-influencers on TikTok (Median = 0, p-value < .001) and much higher median betweenness centrality (Median = 32,958) than micro-influencers on TikTok (Median = 0, p-value < .001). Both Instagram and TikTok networks were characterized by presence of heterophilic ties (i.e., micro-influencers from different geographic regions such as Asia, North America and Europe connected to each other). Among Instagram micro-influencers, those who promoted e-cigarettes alongside lifestyle content had four times as high betweenness (Median = 69,404) as micro-influencers who promoted e-cigarettes exclusively, (p-value < .001). Among TikTok micro-influencers, those who promoted e-cigarettes alongside lifestyle content had higher betweenness (Median = 0, Mean = 1,500, range = 0-49,960) compared to micro-influencers who promoted e-cigarettes exclusively, (Median/Mean = 0, p-value < .001).

Micro-influencers from different geographic regions engage with each other’s content, potentially exposing audiences to a broader e-cigarette content. Micro-influencers who post about e-cigarette alongside lifestyle content occupy more central positions in the Instagram and TikTok networks than those who post about e-cigarettes only, potentially exposing users who are not interested in tobacco-related content to harmful imagery of e-cigarettes. The findings from this project emphasize the need for strengthening influencer marketing regulation on social media platforms popular among youth.