The Effects of Social Media Content Created by Human Brands on Engagement: Evidence from Indian General Election 2014

Abstract

With human brands or individual celebrities in fields ranging from sports to politics increasingly using social media platforms to engage with their audience, it is important to understand the key drivers of online engagement. Using Twitter data from the political domain, we show that positive and negative-toned content receive higher engagement, as measured by retweets, than mixed or neutral toned tweets. However, less popular human brands generate higher social media engagement from positive-toned content compared with more popular human brands. Therefore, we recommend that popular human brands (e.g., popular politicians or chief executive officers) keep their content objective rather than emotional. Furthermore, the tone of related brands (i.e., human brands who belong to the same political party) has a strong reinforcement effect; that is, social media engagement is higher when the tone of the focal human brand and related brands are the same and lower when the tones are different. Therefore, we prescribe that human brands actively coordinate their social media content with related brands to generate higher engagement. From human brands’ perspective, our findings recommend a comprehensive social media strategy, which takes into account the tone of content, tone of related brands’ content, and human brands’ popularity.

Crafting Audience Engagement in Social Media Conversations: Evidence from the U.S. 2020 Presidential Elections

“…Due to reputation, posts by highly followed people are likely to produce more resonance (e.g., retweets [10]) than those of an ordinary information sender regardless of their content. Following [3], who found a negative moderating effect of popularity on the relationship between positive sentiment and engagement for the Indian election in 2014, we suspect that online fame may intrude upon the above links stemming from the ELM and hypothesize: H8: The relationship between a SM post's characteristics and the engagement produced is weaker for SM users with a large number of followers.…”

“…Political tweets having a negative sentiment were insignificantly related to retweeting frequency and speed in [10]. A recent paper by R. Mallipeddi [3] submits that both positive and negative toned tweet content by candidates in the Indian election is associated with higher engagement operationalized by retweet numbers. Theory-wise, [9] represents an exception regarding both central and peripheral route cues.…”

“…Today, social media is firmly sewed into the public discourse and permanently accompanies civic and political life (for meta-review, see [1]), e.g., events like festivals, demonstrations [2], elections [3], as well as crises [4]. In the past decade, the popularity of Twitter, Facebook, Instagram, and, more recently, TikTok skyrocketed and turned these platforms into digital "barometers" of the community's mood and intentions, simultaneously representing the shortest and most straightforward way to reach and convince an audience [5].…”

“…Indeed, extensive evidence suggests that linguistic features matter for acceptance and sharing of news articles [6] [7], the support of crowdfunding projects [8], or patients in online health communities (OHC) [9]. Among others, ICT-mediated conversations in the context of political events received significant attention [10] [3]. Especially Twitter has established itself as a forum for political deliberation, with posts validly mirroring offline political sentiment [5].…”

“…In spite of multiple scholarly attempts to connect message sentiment and audience engagement in terms of feedback and willingness to share [3][4], a closer examination of their results shows mixed evidence (see Section 2), even within the political domain. Concept-wise, most past studies take a one-sided perspective and focus on the affective dimension of the message, i.e., sentiment valence and intensity, often ignoring the cognitive dimension.…”