Reference:
Choi, Y. S., Wu, Q., & Lee, J. Y. (2025). Can You Tolerate Influencer Marketing? An Empirical Investigation of Live Streaming Viewership Reduction related to Influencer Marketing. Journal of Business Research, 188. https://doi-org.devinci.idm.oclc.org/10.1016/j.jbusres.2024.115094
Keywords: Influencer Marketing, Live Streaming ,Persuasion Knowledge Model Propensity Score Matching (PSM), Real-Time Interaction
Summary:
this study suggests that influencer marketing within live streaming may lead to a decline in viewership
as viewers experience resistance toward sponsored content. Building on the persuasion knowledge model, we
analyze this phenomenon using streaming data from Twitch.tv and apply propensity score matching (PSM) to
assess viewership trends.
Development:
The article addresses two central research questions:
- Does influencer marketing in live streaming reduce viewership?
- Which streamer characteristics mitigate or worsen this reduction?
H1 :Influencer marketing in live streaming is associated with viewership reduction.
H2 Moderate levels of real-time interaction mitigate viewership reduction related to influencer marketing.
H3 Topic diversity attenuates viewership reduction related to influencer marketing.
H4 Negative content exacerbates viewership reduction related to influencer marketing.
Moderators :
-Real time interaction
-Topic diversity
-Content negativity
Methodology:
- Propensity Score Matching (PSM)
- Used to reduce self-selection bias between sponsored and non-sponsored streams.
- Regression analysis
- Used to test:
- direct effects,
- moderation effects,
- nonlinear interaction effects.
- Used to test:
- Fixed-effects models
- Controlled for streamer-specific characteristics.
Data was collected from the platform TWITCH
Dataset :
- 87 South Korean streamers,
- 26,657 live streams,
- data collected between:
- July 2020,
- February 2022.
Key findings:
findings reveal a significant decline in viewership associated with influencer
marketing.
We also identify strategies that streamers can employ to mitigate this negative impact. First, moderate
levels of real-time interaction between streamers and viewers help alleviate viewership reduction.
Second,streamers who diversify their content topics experience less viewership declines. Lastly, minimizing negative
content reduces the adverse effect on viewership. These findings contribute to the literature on influencer
marketing and live streaming, offering practical insights for firms and streamers aiming to enhance audience
engagement.
Limitations:
- The study was limited to one streaming platform (twitch)
-
The researchers could not track:
- individual viewer behavior,
- emotional reactions,
- real-time psychological changes