Every muggle knows the overwhelming persuasive impact of fear appeals. Just scare the Other Guy and you will get a change you can count. Of course, when you actually do count the change you typically find almost no change to count. Yet muggles persist (warning labels, anyone?) in declaring and doing something that is disproven.
Until now. Here’s the best available evidence we’ve got for understanding fear appeals and, better still, for doing them effectively. This comes courtesy of a recent meta-analysis of experimental (!!!) studies published in Psychological Bulletin. If you do anything remotely approaching fear appeals (risk appraisal, for example), you need to read this paper and carefully. Again, I want to underline that the studies included in this meta are experiments which provides not only control, but clarity, for understanding this surprisingly complex and tricky persuasion play.
The authors essentially replicate the problems we’ve noted with other fear appeal studies or metas. First, a fear appeal has several moving parts, not just one, and those several individual moving parts typically show Small Windowpane effects on outcomes like intention or behavior. You might recall a previous meta we discussed on the Health Beliefs Model that found this same outcome. Any one moving part is weakly related to the TACT you want to hit. Here’s that same analysis in this meta of experimental studies.
I’ve highlighted the d effects (Small = .2, Medium = .5, Large = .8) for both intention and behavior. All are in that Small+ range which means there’s something going on, but you have to count carefully. So far, we’ve got nothing new. Then the authors start testing the impact of interactions between the individual moving parts of a fear appeal. This table provides a nice illustration.
Take a minute and read each panel to get oriented because each panel is presenting an interaction between two variables. The y-axis reports the d effect size for behavior change. The x-axis displays the level of one moving part (e.g. heightened risk perception) and at the top of the panel you find the second variable (e.g. perceived severity). The interaction clearly demonstrates that while any one part may have little or no effect, when you properly combine them, you’ve got some change to count. It’s still Small+, but you see what kills fear appeals. When you fail at one moving part, it reduces the effectiveness of other moving parts.
That point is made most strongly with this table. It focuses upon two prominent variables Kim Witte has been pounding on since the 1990s and her EPPM, response efficacy and self efficacy.
Please note that far right panel with the proper combination of response and self efficacy. That’s a Medium+ Windowpane on intention and a Small+ Windowpane on behavior. The researchers note other combinations and provide this compelling summary.
Crucially, the impact of heightened risk appraisals was boosted when self-efficacy or response efficacy was enhanced, or when response costs were reduced. In fact, the largest effect sizes in the review obtained when interventions heightened risk appraisals, response efficacy, and self-efficacy simultaneously ( d+ = 0.98 and 0.45 for intentions and behavior, respectively).
When you do fear appeals correctly and implement all the moving parts, you can get Large intention change and Medium behavior change. Here’s how the researchers put it.
The theoretical significance of these findings resides in the empirical support they appear to offer key ideas in the EPPM and early versions of PMT. Both of these theories predicted interactions between risk appraisal and coping appraisal but, with the exception of G.-J. Y. Peters et al. (2012), evidence for such interactions has proven elusive ( Maloney et al., 2011; Ruiter, Verplanken, De Cremer, & Kok, 2004; Witte & Allen, 2000). By meta-analyzing a larger number of interventions than was available heretofore, and by restricting analyses to interventions that generated significant effects on risk appraisals or coping appraisals, the present review enabled us to undertake relevant empirical tests. The findings not only supported the broad hypothesis that risk and coping appraisal combine to influence outcomes but also indicated the specific elements of risk and coping appraisal that are important. In particular, both risk perception and perceived severity effects were augmented by elements of coping appraisal; and response efficacy, self-efficacy, and response costs each augmented the impact of elements of risk appraisal.
Underline that last sentence.
In particular, both risk perception and perceived severity effects were augmented by elements of coping appraisal; and response efficacy, self-efficacy, and response costs each augmented the impact of elements of risk appraisal.
All that is a fear appeal that works. Anything less than all that is what you get with the FauxItAll clamor from the Cool Table with warning labels, scared straight, and Sinners in the Hands of an Angry God: Nothing. Fear appeals are not as simple as they sound or as simple as common sense would have them be. Kim Witte’s EPPM describes all this in great detail and with excellent practical illustration. The problem is that most people don’t execute the EPPM correctly.
On purely a practical basis, you can understand my cringing reluctance to use fear appeals. Sure, if you do a great EPPM implementation you will get great effects. But, you’ve got to get a lot of things moving in the right direction to get the effect and if you screw up one part, it can kill everything else. The complexity of the play makes me nervous as a practical persuader.
Now, if Kim is holding my hand and speaking to me in a soft quiet voice and using little words, then I’m okay. And, in the times I’ve worked with Kim, it has been okay. Nothing blew up. We got a change you can count. She ran the EPPM play and I cheered from the sidelines or ran interference with those who would interfere. So, I’m all in for fear appeals as long as I’ve got Kim Witte on a contract. If not, I’m doing something like Implementation Intentions or a Standard Model design or building low WATT Boxes and Plays with CLARCCS Cues. Those plays, as I prove, are something an idiot can do and get some change to count.
Past our learning about how to do fear appeals correctly, realize how we have learned it. See the value of well-done science for understanding practical persuasion. We’ve had a field of workers doing fear appeals under a wide variety of conditions for a very long time, long enough now that we have a lot of true experiments with random assignment, controlled conditions, smart comparison, and careful counting. That science deftly illuminates what works, what doesn’t, and why or at least how.
Compare that against the Tooth Fairy Tales from the Four HorsePersons of the PostModern Apocalypse. No randomization. Huge samples. No control. Adjusted and debiased data sets. Shifty comparisons. Tricky effect size presentation. And always that parade of statistical significance. Science doesn’t prove everything, but when you do it right, you can produce more change you can count with it than with Tell Me A Story Of Statistical Significance.
Let’s get out of here.
While you’ll always finding me lagging behind, waiting for Kim, you can execute the fear persuasion play and produce obvious, practical, and observable changes in the Other Guys. It is complicated and anyone selling simple fear is selling sand and ice along with ignorance. They will cash your check before you count the change and will not be in the room when you have to explain why you failed. You know what to do: Get Interactions! It’s the combination, stupid. And, you know how to do this. Just chase down Kim Witte and the EPPM! She explains it well enough for even a fool like me to understand it, so imagine how well you’ll do!
Sheeran, P., Harris, P. R., & Epton, T. (2013). Does Heightening Risk Appraisals Change People’s Intentions and Behavior? A Meta-Analysis of Experimental Studies. Psychological Bulletin, 140(2), doi:10.1037/a0033065
P.S. Background Sidebar: I’ve known about the EPPM since Kim has been in the field in the 1990s. By dumb luck, I was on a persuasion panel at a conference with Kim. I’d been out for awhile, but Kim was just finishing her doc and feeling a bit nervous in the presentation. And since it was the EPPM, it is was complicated and even though this was a professional conference with really smart people in the room, most of them were dazed and confused. Hey, it’s a fear appeal; it’s simple, right?
Kim stopped in the middle of the presentation and politely asked, “That makes sense, doesn’t it?” and the room fell silent. So, I chirped out, “Makes perfect sense to me!” and she smiled, then rocked on. Of course, I only understood it enough to be scared of it and realize that if I was going to do fear appeals, it would be as a cheerleader and blocker for Kim.
P.P.S. Found this fun and well done senior thesis project on the EPPM and Sinners in the Hands of an Angry God. If only zealots persuaded this well with fear!