The Effect of Numeracy on the Relationship Between State Reactance and Message Evaluations

ABSTRACT

This study explored the relationship between message reactance potential and individual levels of numeracy in the context of persuasive antiviolence messaging. It was specifically suggested that high levels of numeracy would exacerbate the negative effects of state reactance. The results indicated that high and moderate levels of numeracy strengthened the negative relationship between state reactance and message evaluation. Taken as a whole, the results of this study suggest that for persuasive messages that possess reactance potential and contain statistical/numerical information, individual levels of numeracy exert a nonignorable influence on how the message is evaluated.

Keywords:

• Anti-Violence Messaging
• Numeracy
• Persuasion
• Psychological Reactance Theory

Notes

Space limitations prohibit a full report of the comparative model analyses used to construct the current measure of state reactance. This information can be obtained from the author upon request.

Further details on factor construction can obtained from the author upon request.

Before testing the stated hypotheses, a series of manipulation checks were conducted using independent samples t-tests. The results of these tests indicated that those in the high reactance potential group perceived significantly higher levels of freedom threat, experienced higher levels of anger toward the message, and generated more negative, message-relevant thoughts than those in the low reactance potential group.

Although standardized regression coefficients have been reported for clarity, simple slopes calculations were conducted using the unstandardized regression coefficients. The reported standardized interaction term was manually computed as described by Preacher (2003).

Participants were also asked to sign up for the antiviolence e-newsletter. Consistent with both reactance research specifically and attitude-behavior theories generally, a simple binary logistic regression model suggested that behavioral intentions positively predicted to newsletter signup (0 = no, 1 = yes), B = 0.94, Wald = 18.28, p < .001, Exp(B) = 2.56.

Additional information

Notes on contributors

Toby Hopp

Toby Hopp (PhD, University of Oregon, 2014) is an assistant professor in the Department of Advertising and Public Relations at the University of Alabama.