Emotion Arousing Communications
From the 1950s onward, study of emotional or fear-arousing messages has been a feature of scientific research in psychology and communication science (Janis & Feshbach, 1953, 1954; Maloney, Lapinski, & Witte, 2011; Witte, 1998). The outlines of the effect of fear arousal on message persuasiveness are now well-known. Communications that augment already high levels of anxiety are likely to be shunned or discounted, and consequently are not effective means of evoking attitude change. This is especially true if the source of the message is not credible or if the means of avoiding the dire outcomes promised in the message are not provided. Emotion arousing communications can be persuasive, especially with receivers whose initial emotional involvement in the proscribed activity is minimal. This difference in audience features renders use of emotion arousing communications in mass media campaigns problematic, if the target sample includes individuals from both extremes of the anxiety spectrum.
The general messaging model anticipates this problem. It holds that those already participating, or inching toward the targeted injurious behavior are not likely to process strongly threatening information, as it is punishing and hence will be avoided or counterargued. If the mass mediated campaign is developed for purposes of cessation, fear arousing messages probably are ill advised. Those who have categorically rejected the disparaged activity (smoking marijuana, getting high on inhalants or methamphetamines, and so on) might be affected by a fear arousing message, as it is unlikely that they would counterargue it, but they are likely to be persuaded by almost any other message form. On the negative side, a media blitz involving an illicit substance, as in the Montana Meth campaign (http://www.montanameth.org.), might produce iatrogenic results (Erceg-Hurn, 2008), as it could suggest to naive targets that meth use was normative, or stimulate curiosity, reactions considered in our discussion of iatrogenic effects.
Over all, then, one must wonder whether the loss of credibility generated in the user community, and the possibility of iatrogenic effects, is worth the effect that may be found in audiences of resolute nonusers who might have been influenced by most other ad formats, many of which would not have immediately scared off or pushed users to greater resistance or higher usage. Preaching to the choir is easy; converting the fallen is considerably more difficult. To maximize rare mass media prevention opportunities, it is well to understand that one size probably does not fit all, though some sizes may fit all better than others.
Overpromising
Conceptually related to the (over-) use of emotion arousing messages is the common practice of overpromising the (dire or sometimes positive) consequences that will ensue if the recommendations contained in mass media communications are, or are not heeded. Our research suggests that creating negative drug-relevant expectations that subsequently are demolished through countercommunication or simple life experiences is a recipe for disaster. Skenderian, Siegel, Crano, Alvaro, and Lac (2008) showed that changes from one year to the next in subjects’ expectancies regarding the effects of marijuana use predicted their immediate marijuana intentions and subsequent usage behavior. A positive change in nonusers’ expectations concerning the outcome of marijuana usage (that is, change in a pro-marijuana direction), was followed by a statistically significant uptake in marijuana usage relative to that of nonusers whose expectations remained constant or changed in a negative direction. The guidance to be absorbed from this result is that overpromising the dire consequences of drug usage is a risky game, because if the receiver at some point decides that the warning was not truthful, the rebound effect may be severe. Promises about negative effects can be effective in the short run, but the long-term price to be paid for this short-term gain might exceed the costs of having done nothing at all. Avoid overpromising: the tactic will fail, and when it does, it will produce outcomes even more extreme than if it had never been used. In this case, the cure is worse than the disease.
(Socio-personal) Expectancies
An expectancy is a perceived relationship between a behavior and its anticipated outcome (Tolman, 1932). Expectancies may be considered if/then statements: “If I watch this PSA, I will learn accurate information about drugs,” or “If I smoke marijuana, I will lose (gain) friends.” Expectancies have strong predictive validity for a host of socially problematic behaviors (Siegel, 2011; Siegel et al., 2012). For example, expectations regarding positive or negative outcomes of alcohol use significantly predict drinking behavior over and above demographic factors, alcohol-related attitudes, and prior drinking (Carey, 1995; Stacy, 1997). Siegel and colleagues (2008) emphasized the importance of socio-personal expectations (SPEs: expectations concerning the relation between specific behaviors and desired social outcomes) in an adolescent sample (also see Siegel, Alvaro, Patel, & Crano, 2009). Their research disclosed that a brief media intervention could affect SPEs, which were significantly predictive of inhalant use.
Research on SPEs suggests that antidrug messages that focus on physical harms of drugs may produce suboptimal results. Such ads, especially those targeting adolescents, are based on the premise that users will cease and nonusers will never initiate if they are made aware of the physical harms of substance misuse. Milam and colleagues’ (2000) finding that tobacco smokers perceived smoking as more harmful than nonsmokers calls this assumption into question (see also Crano et al., 2007). Given the importance of social outcomes for most adolescents, SPEs appear to provide a more optimal means of influencing potential drug misuse. Studies that have compared threats of social versus physical harms (Schoenbachler & Whittlerm, 1996; Siegel et al., 2008) consistently reveal that SPEs associated with drug use are susceptible to social influence, predictive of drug use, and can affect drug use attitudes and intentions. As social goals become prominent during adolescence (Erikson, 1968; Hogg, Siegel, & Hohman, 2011), it is reasonable to infer that SPEs should play a greater role in adolescent mass mediated drug prevention (see Blanton et al., 2001, 2008). A media focus on messages that satisfy the general message criteria laid out earlier, and focused on readjusting the socio-personal expectations associated with drug use, offers good prospects for success.
Siegel and J. Burgoon (2002) examined the importance of adolescents’ expectations of antidrug media in a wide ranging integrative discussion of expectancy states theory (EST; Berger, Conner, & Fiske, 1974), expectancy violation theory (EVT; J. Burgoon, 1993), and language expectancy theory (Burgoon & Miller, 1985; Burgoon & Siegel, 2003). They detailed the importance of prior anti-drug messaging efforts on newly encountered anti-drug messages, and supplied a framework for understanding how the negative impact of earlier antidrug ad failures could be overcome. Siegel and J. Burgoon (2002) held that creators of antidrug PSAs had to overcome the shortcomings of past campaigns, which resulted in negative expectations for antidrug PSAs. The stronger the resisted counterattitudinal message, the more likely is resistance to subsequent change efforts. Combining the insights of the expectancy approaches leads to the conclusion that prior experiences can create expectations that strongly affect resistance. The more frequently people encounter ineffective antidrug messages, the more confident they become that future messages will offer little benefit. Thus, future messages will be ignored or encoded with a negative bias.
Misdirection and Indirect Influence Effects
“It seems plausible to suppose that an attempt to persuade a person to change his opinion on some issue would be more effective if the persuasive communication were unexpected than if the person anticipated the influence attempt” (Allyn & Festinger, 1961, p. 35). This supposition is compatible with that of Hovland and his colleagues (1953, p. 23), who theorized that a persuasive communication would be particularly effective when the “unintended” receiver believed that the communicator had “no intention to persuade.”2 Obviously, receivers unaware that they are the targets of a persuasive attack are not likely to defend themselves against it. For this reason, indirect change effects have proved intriguing to many researchers involved
in minority group influence, whose results support the counterargument explanation (Martin & Hewstone, 2008). With reasonable consistency, minority influence researchers have produced highly unusual but replicable results in which persuasive messages on one topic appear to have failed miserably, but on examination are shown to have powerful, systematic, and predicted effects on attitude objects that were never mentioned in the persuasive communication (e.g., Crano & Seyranian, 2009; Perez & Mugny, 1987).
Alvaro and Crano (1997) for example, showed that counterattitudinal communications attributed to an in-group minority usually produced little direct influence; however, these messages predictably affected receivers’ attitudes on (indirect) issues that had been established in advance as related to the focus of the persuasive message, even though this association was not accessible to the receivers themselves. The researchers established that research participants’ attitudes on gays in the military and gun control were strongly associated, but the participants themselves seriously underestimated the strength of this connection. Consistent with their leniency contract model, analysis disclosed that when attributed to an in-group minority, a persuasive message arguing against gays in the military had no apparent effect; however, subjects receiving this source-message combination became significantly more conservative in their attitudes toward gun control.
A second study that reversing the focal and indirect attitude objects supported the original results: A strong anti-gun control message attributed to an in-group minority resulted in less favorable attitudes toward gays in the military. Relevant for present purposes was Alvaro and Crano (1996) findings that counterargumentation in response to an in-group minority’s communication, operationalized as the proportion of negative thoughts assessed in a post-message thought-listing task, was significantly attenuated relative to that observed in the majority or out-group minority message conditions. If conditions motivate receivers not to counter a persuasive message, there is a good chance an associated attitude will change, even if the original message appears to have failed to impart any influence whatsoever.
Crano and Chen (1998) replicated and extended these results in a study that made use of different attitude objects (tuition increase and a proposal that students donate 10 hours of their time, weekly, to university needs). The research replicated the in-group minority effect on associated attitudes, and also demonstrated that large scale changes on the indirect attitude were associated with changes on the focal belief after a delay of one to two weeks. These results suggest persistence when measuring effects of mediated persuasion. Although recommended changes might not occur immediately in response to a persuasive communication, they may arise with the passage of time. This is not conceptualized as a passive process. We hypothesize that information, once introduced successfully, with little counterargumentation or source derogation, may have a long-term effect, even in the absence of immediate changes in the recommended direction. If messages can be delivered in a way that they arouse little counterargumentation and source disparagement, there is a good chance they will come to have the intended effect, even if the circumstances at the time of message delivery render immediate acceptance unlikely.
These results help explain why Crano and colleagues’ (2007) adolescent viewers were particularly susceptible to ads that apparently were intended for their parents. Recall the persuasive message was embedded in an antibullying video, and presented as an advertisement. Middle school participants who saw a message that began with “Are you in the sixth, seventh, or eighth grade?” were significantly less persuaded than those who saw the nearly identical ad that began, “Parents, do you have a young teen at home?” Apparently, youth in the parent communication condition did not feel the need to counterargue the message, and this observation held regardless of their drug use status. Importantly, the message delivered in the communication satisfied Hovland’s three central requirements for persuasive messages.
Epilogue
* * *
Neglect of fundamental principles of messaging may be at least partially responsible for the less than outstanding record of mass media effectiveness in drug prevention trials. To be sure, some campaigns have performed admirably, but they are rare, and probably could have produced better results with greater attention to principles of persuasion that have been part of the common language of social psychology and communication science for the past 60 years. This review has focused on campaigns that did not live up to their initial promise, not out of a sense of smug self-satisfaction, but rather to encourage future campaign planners to seek out hard won knowledge based on strict adherence to strong social science. Openness on the part of mass media campaign directors to the lessons of research is essential if we are to make progress, but it cannot be taken for granted. Voices of concern about the “direction and execution” of the Campaign (DeJong & Wallach, 2000, p. 77), for example, which has featured heavily in this review, were raised in sufficient time to alter the course of the work. Further, during the campaign itself, unease with the messages being delivered and their lack of connection to established theories of messaging or persuasion were voiced repeatedly (Crano, 2010), but to no avail. A hallmark of the good scientist and the responsible science administrator is recognition that one does not know everything and that established knowledge must be considered seriously. Throwing lots of money at a mass media campaign devoid of a reasonable idea of how persuasion works is unwise at best.
In the abstract, the advice provided here is relatively straightforward: Pay attention to established theory, realize that intuition does not trump the insights drawn from years of intense research, ensure campaign messages are based on theory and congruent empirical results, maintain the integrity of the research design, consider the beliefs, norms, and behaviors of the respondents to be studied, attend to the possibility of iatrogenic effects, respond accordingly, and assess the outcome of the effort in a scientifically credible manner. This is not an esoteric formula, but judging from past practice, it is not one that is easy to follow. The guidelines laid out are designed to improve mass communication outcomes. They are offered in a constructive voice to answer the simple question, “Can the mass media play a role in prevention of drug misuse?” In our view, there is no doubt that they can, and have. Can the record of accomplishment be improved? As before, there is no doubt that it can, and that it must. Openness on the part of campaign directors is essential, as is a continued and unrelenting insistence by scholars of communication and persuasion that publicly funded campaigns adhere to well-established, empirically based models in the design and delivery of persuasive messages. Nothing short of unremitting perseverance will work, but when lives hang in the balance, such dedication is worth the effort.
Acknowledgments
* * *
Our students in the Health Psychology and Prevention Science Institute at Claremont Graduate University were of incalculable value in their willing support of this work, giving freely and unstintingly of their time in facilitating our review of the massive literature concerned with the use of mass media in drug prevention. We could not have done it without you. Preparation of this research was supported by a grant from the U.S. National Institute on Drug Abuse (R01 DA030490), which we gratefully acknowledge. The contents of this chapter are solely the responsibility of the authors and do not necessarily reflect the views of the Institute.
Notes
* * *
1. A choice of audience features is required to maximize outcomes. At the most basic level, we must decide if the goal is prevention or cessation. And, the possibility that targeting one group may disadvantage other groups should be recognized. Until technology progresses to the point that messages can be tailored to each individual’s specific proclivities, the mass media approach will be faced with these alternatives.
2. The effect was thought to work through a reduction in counterargumentation, though this possibility was not tested.
References
* * *
A
llyn, J., & Festinger, L. (1961). The effectiveness of unanticipated persuasive communications. Journal of Abnormal and Social Psychology, 62, 35–40.
Alvaro, E. M., & Crano, W. D. (1996). Cognitive responses to minority- or majority-based communications: Factors that underlie minority influence. British Journal of Social Psychology, 35, 105–121.
Alvaro, E. M., & Crano, W. D. (1997). Indirect minority influence: Evidence for leniency in source evaluation and counterargumentation. Journal of Personality and Social Psychology, 72, 949–964.
Anderson, D. M. (2010). Does information matter? The effect of the Meth Project on meth use among youths. Journal of Health Economics, 29, 732–742.
Atkin, C. (2002). Promising strategies for media health campaigns. In W. D. Crano & M. Burgoon (Eds.), Mass media and drug prevention (pp. 35–64). Mahwah, NJ: Erlbaum.
Atkin, C., & Wallack, L. (Eds.). (1990). Mass communication and public health: Complexities and conflict. Newbury Park, CA: Sage.
Berger, J., Conner, T. L., & Fiske, M. H. (1974). Expectation states theory: A theoretical research program. Cambridge, MA: Winthrop.
Berkowitz, A. D. (1997). From reactive to proactive prevention: Promoting an ecology of health on campus. In P. C. Rivers & R. E. Shore (Eds.), Substance abuse on campus: A handbook for college and university personnel (pp. 119–139). Westport, CT: Greenwood.
Blanton, H., & Burkley, M. (2008). Deviance regulation theory: Applications to adolescent social influence. In M. J. Prinstein & K. A. Dodge (Eds.), Understanding peer influence in children and adolescents (pp. 94–121). New York, NY: Guilford.
The SAGE Handbook of Persuasion Page 62