Abstract
This study examined whether a debater's background nonverbal behavior affected audience perceptions of her and her opponent's likeability. Students watched one of four versions of a televised debate. In each, while the speaking debater appeared on the main screen, subscreens displayed her opponent's background nonverbal behavior. In one version, the nonspeaking debater displayed a neutral expression, whereas in the others she displayed occasional disagreement, nearly constant disagreement, or both agreement and disagreement. After viewing the debates, students rated the debaters' likeability. Analysis indicated that background behavior influenced perceptions of the nonverbal communicator but not of the speaking debater.
Notes
For a script of the debate, contact John S. Seiter.
In all versions of the debate, the second speaker's subscreens appeared four times during her opponent's speech (at 11 seconds; 1 minute, 55 seconds; 3 minutes, 40 seconds; and 5 minutes, 26 seconds) resulting in an average of 1 minutes, 21 seconds total time on subscreen appearances.
Longer lasting subscreens in the fourth (constant disagreement) condition permitted us to include more nonverbal disagreement relative to the other conditions. Doing this, however, created a limitation to the study that is addressed in the Discussion section.
The omnibus F test for likeability for Speaker 1 was F(3, 147) = 6.55, p < .001, η2 = .121.
The omnibus F test for likeability for Speaker 2 was F(3, 147) = 1.85, p = .906, η2 = .004. Here, we followed the suggestion by Rosenthal, Rosnow, and Rubin (Citation2000) that contrast analyses should be conducted even in cases when the omnibus F test is not significant.