Threat Appeals: The Fear–Persuasion Relationship is Linear and Curvilinear

ABSTRACT

Drive theory may be seen as the first scientific theory of health and risk communication. However, its prediction of a curvilinear association between fear and persuasion is generally held to be incorrect. A close rereading of Hovland et al. reveals that within- and between-persons processes were conflated. Using a message that advocated obtaining a screening for colonoscopy, this study (N = 259) tested both forms of the inverted-U hypothesis. In the between-persons data, analyses revealed a linear effect that was consistent with earlier investigations. However, the data showed an inverted-U relationship in within-persons data. Hence, the relationship between fear and persuasion is linear or curvilinear depending on the level of analysis.

Acknowledgments

We thank Eric Loken for his sound advice on an earlier version of this article, some of which we heeded.

Funding

The project was supported by the National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health, through grant UL1 TR000127.

Notes

¹ One reviewer took exception to this conclusion, noting that Witte and Allen (2000) assert that both threat and efficacy components must be strong for the message to be effective. We agree that they made this claim. However, as Mongeau (2013) explains, their conclusion does not follow from their results, which showed only main effects for threat and efficacy. In any case, our focus is on fear the emotion, not threat and efficacy as precursors of the emotion.

Additional information

Funding

The project was supported by the National Center for Research Resources and the National Center for Advancing Translational Sciences, National Institutes of Health, through grant UL1 TR000127.