The Superstar Social Media Influencer: Exploiting Linguistic Style and Emotional Contagion over Content?

Part of: Digital Marketing Topic 9 — AI in Advertising / Influencers · Reading Citation: Lee, M. T., & Theokary, C. (2021). The superstar social media influencer: Exploiting linguistic style and emotional contagion over content? Journal of Business Research, 132, 860–871. Key concepts: Social Media Influencer, Elaboration Likelihood Model, Language Expectancy Theory, Emotional Contagion, Linguistic Style

---

TL;DR

Why do some social-media influencers become wealthy superstars while most barely cover costs? Using the Elaboration Likelihood Model with Language Expectancy Theory and Emotional Contagion Theory, Lee & Theokary study 27 YouTube automotive-review channels (393 viewer responses) via speech-to-text text-mining, surveys, archival data and structural equation modelling. Contrary to pre-internet persuasion theory, the central route reverses: close/concrete/interactive linguistic style drives views and subscribers, while content and production expertise — traditionally central — become peripheral, acting only as a mediator. Emotional contagion has no direct effect; it works only through expertise.

Why It's on the Reading List

It is the empirical influencer reading for the topic, complementing the AI-influencer material: it shows how influencers persuade and which cues actually drive financial performance (views/subscribers), and demonstrates a rigorous SEM test of the ELM in a digital-video context. Good for the "how it is said vs what is said" exam contrast.

Background & Research Question

The influencer market spent ~US$8bn in 2019; YouTube pays roughly US$0.018 per view (~US$1,800 per 100,000 views), yet influencer success varies hugely. Traditional persuasion research (Friestad & Wright's Persuasion Knowledge Model; regulatory focus theory) emphasises "what is said" (content/consumer knowledge) and "how it looks" (persuasion knowledge). Lee & Theokary ask whether, for video influencers, "how it is said" matters more. Three research questions:

1. Which persuasion theories are relevant for superstar influencers?
2. Which elements make someone a superstar influencer?
3. How should the persuasion process be conceptualised and modelled?

Key Concepts & Definitions

Definition — Elaboration Likelihood Model (ELM)

A dual-mode processing model. Central route: able, motivated, focused viewers critically scrutinise issue-relevant information. Peripheral route: low-elaboration viewers rely on superficial cues (language, emotion, source characteristics). High elaboration → central route dominates.

Definition — Language Expectancy Theory (LET)

People form expectations about appropriate communication styles. Message features that positively exceed expectations increase persuasion; negative violations reduce it. Underpins linguistic style.

Definition — Emotional Contagion Theory (ECT)

An influencer's excitement, enthusiasm and passion create a "contagion": viewers subconsciously mimic and synchronise the displayed emotion, then experience it themselves via physiological feedback — generating joy, zeal and desire to view more.

Definition — Linguistic style

The verbal communication style in persuasion. Operationalised on four dimensions: concreteness (articles, prepositions, quantifiers), preciseness (synonym/three-gram use), interactivity (questions, inviting feedback), and psychological closeness (first-person pronouns: I, me, you, we).

Methodology

Multi-method design

• Context: 27 YouTube automotive-review channels that reviewed the same new vehicle (high-value product; able, motivated viewers — amenable to ELM).
• Data: speech-to-text + text-mining (concreteness, preciseness, closeness); participant surveys on a 7-point scale (interactivity, emotional contagion, content & production expertise); archival data from Social Blade (views, subscribers).
• Sample: 393 responses from 19 volunteer participants; inter-rater reliability (ICC) 0.719–0.848.
• Performance (DV): % increase in views (Vchg) and subscribers (Schg), day-before vs one-month-after upload.
• Analysis: PLS-SEM (SmartPLS 3.0). Controls: video release date, duration, channel age.
• Fit: SRMR = 0.072 (good); NFI = 0.770 (acceptable); all VIF < 3; R² = 0.416 (expertise), 0.277 (performance); 78.28% variance explained.

Framework / Model

Proposed ELM for entrepreneurial social-media influencers — central and peripheral routes are reversed from tradition.

Key relationships:

• Close + concrete + interactive language is a strong direct predictor of views/subscribers (the novel central-route finding).
• Preciseness does not matter — audiences prioritise close, concrete, interactive language over exact word choice.
• Emotional contagion → expertise (0.552, t=15.87) → performance (0.110, t=2.22): expertise fully mediates emotional contagion.
• Expertise was reclassified into a single factor (content + production loaded together) and confirmed as peripheral, working only as a mediator.

Implications for Marketers

• For video influencers, differentiation comes from language and displayed emotion, not content/production quality (which is now table-stakes as competition raises the minimum bar).
• Influencers should cultivate psychological closeness, concrete language and interactivity (ask questions, invite feedback, use first-person pronouns, self-disclose).
• Emotional contagion still matters — but only when paired with expertise; passion alone does not grow numbers.
• Winners "find ways to take advantage of linguistic style and emotional contagion"; losers rely solely on content/production.

Exam Takeaways

Likely exam points

• The reversal of the ELM: linguistic style + emotional contagion become central; content/production expertise becomes peripheral.
• The four linguistic-style dimensions — and that preciseness is NOT significant.
• Emotional contagion has no direct effect — it is fully mediated by expertise.
• Theories used: ELM (framework) + LET (linguistic style) + ECT (emotional contagion); contrasts with the PKM.
• Method: 27 YouTube auto channels, speech-to-text + survey + Social Blade archival, PLS-SEM.

Methodology limitations (per authors)

• Text-data analysis involves judgement on sources/coding; survey perceptions assumed to reflect underlying levels.
• Cross-sectional SEM cannot establish time-lagged/causal effects; direction inferred from theory.
• Single context (automotive YouTube) — other communities/platforms (Instagram, Snapchat, TikTok) may need different cues.

Summary

• Empirical multi-method study of why influencer success varies.
• Uses ELM + LET + ECT; tests via PLS-SEM on YouTube auto reviewers.
• Central route is reversed: language style drives performance; expertise is peripheral/mediating.
• Practical lesson: "how it is said" beats "what is said" and "how it looks".

Related Notes

• Digital Marketing — subject hub
• Social Media Influencer — the focal actor
• Elaboration Likelihood Model — the persuasion framework (central vs peripheral routes)
• Emotional Contagion · Linguistic Style — the central-route constructs
• Persuasion Knowledge — the pre-internet model this study challenges
• 09a-Rodgers - Promises and Perils of AI and Advertising — companion influencer/AI reading