Author ORCID Identifier


Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation


Communication and Information



First Advisor

Dr. Marko Dragojevic


Persuasive attempts often fail. One theory that attempts to explain when and why people resist persuasion is psychological reactance theory (PRT; J. W. Brehm, 1966; S. S. Brehm & J. W. Brehm, 1981). According to the theory, people have an inherent desire to maintain their autonomy, or freedom. When they perceive that a persuasive message threatens their freedom, they experience psychological reactance, an aversive motivational state, which motivates them to restore their freedom by resisting the persuasive message. Whereas past studies have demonstrated that various message features (e.g., controlling language) and source characteristics (e.g., social group membership) can, individually, influence perceived freedom threat, few studies have investigated the effects of message and source characteristics on PRT outcomes in tandem. Guided by politeness theory (PT; Brown & Levinson, 1987), PRT, and the intergroup approach – i.e., social identity theory (SIT; Tajfel & Turner, 1979) and self-categorization theory (SCT; Turner et al., 1987) – this dissertation sought to address this gap in the literature by simultaneously examining the effects of controlling language and source social group membership on perceived freedom threat, reactance, and resistance to persuasion (i.e., source and message derogation, attitudes, and behavioral intentions). To achieve this, a 2 (controlling language: low, high) x 2 (source social group membership: ingroup, outgroup) between-subjects experiment was conducted. American nationals (N = 498) read a fictional commentary about global warming, which advocated for greater engagement in eco-friendly behaviors. Based on experimental condition, the commentary contained either low (e.g., could) or high (e.g., must) controlling language and was ostensibly written by either an ingroup (i.e., American) or an outgroup source (i.e., Chinese). Compared to low controlling language, high controlling language posed a greater threat to perceived freedom for both sources, but this effect was more pronounced for the outgroup source. Compared to low controlling language, high controlling language also elicited more reactance, but only for the outgroup source. Resistance to persuasion also differed based on controlling language and source group membership. Compared to low controlling language, high controlling language led to greater derogation of both sources’ warmth but was more pronounced for the outgroup source. High controlling language also increased derogation of the source’s competence and the message itself, regardless of the source’s group membership. Contrary to predictions, the outgroup source increased intentions to engage in eco-friendly behaviors, compared to the ingroup source. No significant effects emerged for attitudes toward the message advocacy. Taken together, the findings suggest that source group membership can moderate the effects of controlling language on perceived freedom threat, reactance, and—to a more limited extent—resistance to persuasion. Theoretical and practical implications are discussed.

Digital Object Identifier (DOI)

Available for download on Friday, April 24, 2026