Abstract
This study examined the third-person effect (TPE) hypothesis (Davison, 1983) in offensive advertising in the Chinese cultural context. Based on a survey of 1,539 Chinese Internet users about the third- and first-person effects among offensive ads, neutral ads, and public service ads, the study inquires into the relationship between the TPE and respondents’ levels of acceptance toward advertising. Besides confirming the TPE existence in an Eastern cultural context, the results suggest that the TPE predict word-of-mouth (WOM) spreading for both offensive and neutral product ads, but not for PSAs. Theoretical contributions and managerial implications of these findings are discussed.
Introduction
The Third-Person Effect (TPE) is a psychological phenomenon where individuals believe that mass media messages exert a greater influence on others than on themselves. While extensively studied in Western, individualistic societies, its application in Eastern, collectivistic cultures—particularly regarding offensive advertising—remains under-explored. This research investigates how Chinese consumers perceive the influence of ads for stigmatized products (e.g., cigarettes, liquor, private clinics) compared to neutral ads and socially desirable Public Service Announcements (PSAs).
Methods
The study employed a two-phase methodology. Initially, an exploratory phase with 132 college students identified six offensive product categories: cigarettes, private clinics, health supplements, specialty drugs, liquor, and female private products.
In the second phase, a quantitative survey was administered to 1,539 Chinese internet users. The sample was constructed using quota sampling to reflect national internet demographics across gender, age, education, region, and income levels.
Respondents rated the perceived influence of ads on their own purchase intent versus that of "others" using 5-point Likert scales. TPE was calculated as the difference between these scores. The study also analyzed behavioral outcomes, including willingness to engage in word-of-mouth (WOM) communication and support for censorship.
Results
The findings strongly support the TPE hypothesis for offensive advertising. Respondents consistently rated the influence of offensive ads on others as significantly higher than the influence on themselves.
As shown in the data, cigarettes produced the highest TPE score (0.52), while female private products produced the lowest among the offensive categories (0.15). In contrast, PSAs exhibited a "First-Person Effect" (FPE), where respondents believed they were more influenced by the positive messaging than others (difference score of -0.08, $p < .001$).
Furthermore, TPE served as a significant predictor of behavior. For offensive ads, TPE explained 16.2% of the variance in WOM spreading and 7.0% of the variance in support for censorship. Interestingly, the "gift-giving" culture in China may leverage this bias; consumers may purchase "vulgar" products as gifts, believing the recipient is influenced by the ad even if they claim they are not.
Discussion and Limitations
This study confirms that social desirability is a primary driver of perceptual bias in media influence. The more an ad is perceived as socially unacceptable, the stronger the TPE. Conversely, socially desirable messages like PSAs reverse this effect.
However, the study is limited by its reliance on recall rather than real-time exposure to specific ad stimuli. Future research should utilize experimental designs with controlled exposure to specific brands and creative executions to better isolate the causes of offensiveness and explore why the FPE for PSAs does not translate into behavioral predictors like WOM.
Summary updated