Hail to the Thief: Linguistic Agency Increases the Persuasive Power

Agency and Persuasion 1 Hail to the Thief: Linguistic Agency Increases the Persuasive Power of Consumer Education Materials about Identity Theft Matthew S. McGlone Joseph McGlynn, III Maxim V. Baryshevtsev Kathleen G. Blackburn Leah E. LeFebvre Nancy L. McCallum Max Aaron Wartel Alexandra Abbott Department of Communication Studies, The University of Texas at Austin
Dramatic increases in the incidence and severity of Message design is at the core of health education identity theft have prompted many observers to campaigns (Cho, 2012; Maibach & Parrott, characterize the problem as an “epidemic.” Combating 1995). Health educators spend considerable time the threat will require a multimodal intervention of new constructing persuasive messages about health technologies, innovative business practices, a threats, prevention measures, and treatments for medical strengthened legislative framework, more engagement conditions. When doing so, they often turn to research on from law enforcement, and expanded victim assistance. It message effects – i.e., empirical demonstrations of will also require comprehensive consumer education persuasive message features on attitude and behavioral about the nature of the crime and measures people can change (e.g., Ajzen, 1991; Bandura, 1998). take to reduce their vulnerability to it. In this respect, One variable that has received recent attention from intervention efforts might benefit from the lessons learned message effect researchers is linguistic agency (McGlone, in education campaigns about actual epidemiological Bell, Zaitchik, & McGlone, 2013). Linguistic agency is threats such as influenza or HIV/AIDS. defined as the ascription of action or change to one or Healthcare practitioners rely extensively upon more entities involved in an event. For example, a flu educational materials to inform the public about sufferer might assign agency to herself (e.g., I came down epidemiological and prevention measures (Schwartzberg, with the flu) or to the illness (e.g., the flu laid me out for Cowett, & VanGeest, 2007). Such materials can take the days). The ascription of causality to an entity involved in form of physical leaflets and posters or digital documents an event, such as a disease, implies passivity of other disseminated via email and the internet. These materials, entities in the same event (Dowty, 1991). Thus, the which can be used alone or to reinforce provider assignment of agency to a health threat implicitly conveys counseling, have been shown to significantly reduce the notion that people who have been “afflicted” have people’s susceptibility to epidemic threats (Johansson, limited control over their circumstances. Salantera, Katajisto, & Leino-­‐Kilpi, 2004). Agency and Persuasion 2 The differing conceptions of a health threat that result McGlone (2014) thus examined the effects of agency from ascriptions of agency shape message recipients’ assignment for a nonliving health threat, radon gas. They perceptions of the threat. McGlone, Bell, Zaitchik, and found that the agency experimental manipulation needed McGlynn (2013) found that assignment of agency to a to assign sentience (i.e., intentionality) to radon gas to virus (in this case, the H1N1 virus) rather than to people produce a significant effect on perceptions of the threat. heightened message recipients’ beliefs about the severity For example, a literal description of radon gas of the virus, their perceived susceptibility to it, and their dissemination into a home (e.g., Radon gas seeps into intention to seek vaccination. Bell, McGlone, and people’s homes) was less effective than a figurative Dragojevic (2013a) assessed the effects of linguistic agency portrayal of the gas as deliberately targeting a home assignment in messages about four unique bacterial (Radon gas invades people’s homes). threats. People who read a warning about a bacterial Messages warning people about the threat of identity threat rated the threat as more severe and self as more theft need not rely on figurative portrayals of sentience, of susceptible when agency was assigned to the bacterium course; the individuals and groups who commit this crime (e.g., STEC kills people every day) than to humans (e.g., are literally and intentionally targeting their victims. people die from STEC every day). Bell, McGlone, and However, inspection of the various online resources Dragojevic (2013b) examined linguistic agency effects in designed to educate consumers about this crime reveals a the context of health policy editorial advocating pattern of “nominalization” that can potentially mandatory vaccination of girls and boys against the human undermine the persuasive force of these resources. papilloma virus (HPV). Research participants perceived Nominalization occurs when descriptions of activity, HPV to be more severe when agency was assigned to it typically performed by action verbs (evaluate, invade, etc.) (e.g., HPV infects millions of people) rather than to humans are instead articulated as abstract nouns (evaluation, (e.g., millions of people contract HPV). In addition, the HPV invasion, etc.). Although nominalization is a natural and vaccine was perceived to be more effective when agency common language phenomenon, its effect on discourse is was assigned to it (e.g., vaccination guards people from often to attenuate the sense of agency associated with an HPV) than to the vaccinated person (e.g., people can guard activity. To illustrate, consider the following nominalized themselves from HPV with vaccination). activity descriptions: The aforementioned studies focused exclusively on threats that are living (e.g., bacteria) or likely to be perceived as alive (e.g., a virus). Dragojevic, Bell, and •
There was a discussion about raising the minimum wage. Agency and Persuasion 3 •
•
An investigation revealed that building about cybercrime created by the U.S. Department of inspectors were bribed to overlook code Justice (http://www.justice.gov/criminal/fraud/websites/ violations. idtheft.html) and Federal Bureau of Investigation The attack killed 50 people. (http://www.fbi.gov/about-­‐us/investigate/ In each case, causation is linguistically assigned to the cyber/identity_theft). Although all of these resources activity itself (discussion, investigation, attack) instead of offer valuable advice to consumers about how to avoid the actual agent(s) who performed the action (legislators becoming a victim of this crime, overreliance on the discussing policy, police officers investigating a crime, or nominalized term detracts from their persuasive force. terrorists preying on a village), who are omitted entirely Psychological studies of human responses to threat from the description. The replacement of these “true indicate that we consider threats to be the most severe agents” with nominalizations artificially distances readers and ourselves as most susceptible to them when we can from the activities these sentences describe. transparently perceive it as deriving from an antagonist – i.e., an agent that intentionally and actively seeks to do us The term “identity theft” is also a nominalization. Below are several examples of its usage in a 2012 harm (e.g., Lazarus, 1991). Thus by characterizing a publication by the Federal Trade Commission titled criminal threat chiefly in terms of an abstraction called “Talking about Identity Theft: A How-­‐To Guide”: “identity theft,” the FTC guide artificially mutes •
•
Identity theft occurs when someone uses your perceptions of the threat posed by the agents who personally identifying information to commit perform the actual criminal activity. To illustrate, compare fraud or other crimes. the example sentences above to the versions below in You or someone you know may have experienced which agency is restored: some form of identity theft. •
•
•
“thief” or “thieves” (34 instances). A comparable partiality for the nominalized term is apparent in online resources Awareness is an effective weapon against identity thieves. The term appears 125 times in the 53-­‐page pamphlet, roughly four times as often as any reference to an identity You or someone you may know may have been the victim of an identity thief. Helping employees fight identity theft makes good business sense. Identity thieves steal your personally identifying information to commit fraud and other crimes. Awareness is an effective weapon against identity theft. •
•
•
Helping employees fight identity thieves makes good business sense. Agency and Persuasion 4 These versions convey the same information as their (“requesters”) post small human intelligence tasks (“HITs”) nominalized counterparts, but in a way that expresses the and employees (“workers”) perform those tasks for inherent antagonism of the crime more directly. compensation (in this case, $0.50 for a 15-­‐minute time The purpose of the reported research was to commitment). 486 participants were randomly assigned demonstrate the difference in persuasive impact between to read one of the aforementioned fact sheet versions and consumer education materials that employ agentic or afterwards completed a questionnaire assessing their nominalized descriptions of identity theft. We beliefs and attitudes regarding identity theft. Analyses of hypothesized agentic the questionnaire responses indicated three statistically descriptions of the crime would create stronger significant differences between the responses generated perceptions of personal susceptibility and intentions to by participants exposed to the different sheet versions. take preventative action than materials referring to the Specifically, participants who read the agentic description crime in nominalized form. To test this hypothesis, we (“identity thief”) version a) perceived the threat of identity created different versions of a consumer education “fact as more severe, b) perceived themselves as more sheet” that defined the crime, provided recent statistics personally susceptible, and c) reported stronger intentions about its incidence and associated costs, and offered to take preventative action than participants who read the advice for how consumers can protect themselves from its nominalized activity (“identity theft”) version. These occurrence. One version of this sheet referred to the findings support our hypothesis that agentic descriptions crime consistently using agentic descriptions of the of identity theft portray the crime in a way that makes criminal activity (e.g., Identity thieves can ruin your credit consumers more apprehensive about the prospect than and your reputation) and the other consistently employed nominalized descriptions, and thus more inclined to the nominalized term for the activity (e.g., Identity theft engage in behaviors that will reduce their susceptibility to can ruin your credit and your reputation). Each sheet this crime. The implications of these findings for the version was approximately 300 words long and contained development of educational materials are clear: 16 distinct references to the activity (see Figures 1 and 2). consumers are more apprehensive about a tangible “thief” Participants for our study were recruited through a than an abstract “theft,” and this subtle but nonetheless post on Mechanical Turk (MTurk; www.mturk.com), an significant linguistic difference can be used to enhance the online platform owned by Amazon.com. MTurk is a persuasive impact of cybercrime information campaigns. that materials employing crowdsourcing labor market in which employers Agency and Persuasion 5 References Ajzen, I. (1991). The theory of planned behavior. Organizational Behavior and Human Decision education. Patient Education & Counseling, 52, 175-­‐181. Lazarus, R. S. (1991). Emotion and Adaptation, New York: Processes, 50, 179–211. Bandura, A. (1998). Health promotion from the Oxford University Press. Maibach, E., & Parrott, R. (Eds.). (1995). Designing health perspective of social cognitive theory. Psychology and messages: Approaches from communication Health, 13, 623–649. theory and public health practice. Thousand Oaks, CA: Bell, R. A., McGlone, M. S., & Dragojevic, M. (2013). Bacteria as bullies: Effects of linguistic agency Sage. McGlone, M. S., Bell, R. A., Zaitchik, S. T., & McGlynn, J. assignment in health messages. Journal of Health (2013). Don’t let the flu catch you: Agency Communication. doi: 10.1080/10810730.2013.798383. assignment in printed educational materials about the Bell, R. A., McGlone, M. S., & Dragojevic, M. (in press). H1N1 influenza virus. Journal of Health Vicious viruses and vigilant vaccines: Effects of linguistic agency assignment in health policy McGlone, M. S., & Harding, J. L. (1998). Back (or forward?) advocacy. Journal of Health Communication. Cho, H. (Ed.). (2012). Health communication message design: Theory and practice. Thousand Oaks, CA: Sage. Dowty, D. (1991). Thematic proto-­‐roles and argument Communication, 18, 740–756. selection. Language, 67, 547-­‐619. Dragojevic, M., Bell, R. A., & McGlone, M. S. (2014). Giving to the future: The role of perspective in temporal language comprehension. Journal of Experimental Psychology: Learning, Memory and Cognition, 24, 1211–1223. McGlone, M. S., & Pfiester, R. A. (2009). Does time fly when you’re having fun or do you? Affect, agency, Radon Gas Life Through Language Effects of Linguistic and embodiment in temporal communication. Journal Agency Assignment in Health Messages About of Language and Social Psychology, 28, 3–31. Inanimate Threats. Journal of Language and Social Psychology, 33, 89-­‐98. Schwartzberg, J. G., Cowett, A., VanGeest, J., & Wolf , M. S. (2007). Communication techniques for patients with Johansson, K., Salantera, S., Katajisto, J., & Leino-­‐Kilpi, H. low health literacy: A survey of physicians, nurses, and (2004). Written orthopedic patient education pharmacists. American Journal of Health Behavior, 31, materials from the point of view of empowerment by 96-­‐104. Agency and Persuasion 6 Figure 1. Agentic Description (“Identity Thief/Thieves”) Version of the Fact Sheet. Agency and Persuasion 7 Figure 2. Nominalized Description (“Identity Theft”) Version of the Fact Sheet.