Probing the Effects of Normative Beliefs, Attitude Strength, and Attitude Certainty on Opinion Expression Behavior

ABSTRACT

The reported study investigated the effects of normative factors and metacognitions on opinion expression behavior. It sought to add to the knowledge base by probing the extent to which they operate additively or non-additively. Participants were placed in an online opinion environment and given the choice about whether and how to express something. Perceived minority opinion status, trait-like fear of social isolation, situationally-dependent beliefs about the sanctions one would receive for expressing something, attitude strength, and attitude certainty were measured. The results failed to support a non-additive effect. Rather, three normative factors independently predicted whether participants commented, and attitude certainty solely predicted the directness of participants’ expressions. We speculate that normative beliefs may contribute to participants’ decisions about whether to express something, and metacognitions may contribute to the form of those expressions.

KEYWORDS:

• Normative influence
• metacognition
• expression behavior

Disclosure Statement

No potential conflict of interest was reported by the author(s).

Notes

1. We also measured length of post using an Excel function that records word length (M = 18.68, SD = 15.35, Mdn = 15.00). Initially, we wondered whether attitude strength and attitude certainty might predict the verbosity of someone’s post. People with low levels of knowledge about a topic sometimes judge themselves as more competent than they really are, while more highly competent individuals sometimes underestimate their knowledge (Anson, 2018; Kruger & Dunning, 1999); we speculated that people with greater attitude strength and attitude certainty might be especially confident, leading them to talk at greater length about their opinion. We found that length of post was not significantly correlated with attitude strength, r(655) = .04, p = .363, or attitude certainty, r(657) = .06, p = .154, and scatter plots did not suggest that any nonlinear relationships existed.

Additional information

Funding

This research was funded by startup funds provided by Purdue University College of Liberal Arts.

Notes on contributors

David M. Keating

Dave Keating (PhD, Michigan State University) is an assistant professor in the Department of Communication & Journalism at the University of New Mexico.

Evan K. Perrault

Evan Perrault (PhD, Michigan State University) is an associate professor in the Brian Lamb School of Communication at Purdue University.

Seth P. McCullock

Seth McCullock is a PhD candidate in the Brian Lamb School of Communication at Purdue University.