Yesterday I posted about a study of the growth of “image bites” and the decline of “sound bites” on network television news. After reading that post, my colleague Gina Lambright asked whether any other research has been done on the political impact of video versus audio; this was a natural question for her to ask, for her spouse is a news photographer. Anyway, the answer is yes. As Exhibit A (other exhibits will follow in due course), let me point to a nice piece of work that Jamie Druckman completed a few years ago.
The first Kennedy-Nixon debate in the 1960 presidential campaign has lived on in memory as a turning point not so much because of what the candidates said but rather because of how they looked. Kennedy looked like … well, Kennedy, and Nixon looked like an especially unflattering caricature of himself. Everyone “knows” that Nixon’s unattractive appearance led him to be perceived as the loser of the debate. However, the evidence that supports that conclusion turns out upon inspection, to be somewhere in the range of weak to nonexistent. Until Druckman’s study, the only reasonably credible evidence came from a post-debate survey that indicated that those who had listened to the debate on the radio were more likely to think Nixon had won, but those who watched it on television were more likely to see Kennedy as the winner. That’s a nifty result, if valid, but Druckman questions its validity for a host of methodological reasons that I won’t go into here except to say that they’re pretty compelling.
In an attempt to clear things up, Druckman ran an experiment, the participants in which (mostly undergraduate students) had no prior knowledge of the Kennedy-Nixon debates. These participants were either shown the debate on television or had the audio played for them. Druckman suspected that watching the debate would lead them to base their assessments of the candidates more on personality traits, and that proved to be the case; at the same time, agreeing with a candidate on the issues significantly affected the audio-only participants’ assessments of the candidates, but the same didn’t hold true for those who had had the video as wel as the audio input. Druckman also expected viewers to learn more factual content than listeners, largely because television is more likely to hold people’s attention than audio-only input is. That, too, proved to be the case. As for “bottom-line” evaluations of the two candidates, viewers did turn out to be more likely to see Kennedy as the winner of the debate. As Druckman concluded:
bq. This is compelling evidence that television – by enhancing the impact of image – can make a difference in overall candidate (debater) evaluations. It also is the first clear empirical evidence consistent with the widespread assertion of viewer-listener disagreement in the first Kennedy-Nixon debate. In sum, television images have an independent effect on individuals’ political judgments: they elevate the importance of perceived personality factors, which can in turn alter overall evaluations.”
Here, then, we have an instance in which rigorous social science research supports the conventional wisdom. What everybody “knew” turned out to be right after all. (Click here for an abstract of Druckman’s study.)