AN EMPIRICAL LOOK AT AN INTERNATIONAL DEBATE
Samuel V. O. Prichard and James C. McCroskey
For several decades American college debaters have engaged in debate with English-speaking debaters from other countries. Such debates have increasingly become tradition on some American campuses. Most of us consider these confrontations to be extremely valuable. They stimulate interest in debating among student audiences on host campuses; they provide an airing of controversial questions that may otherwise not be discussed at public gatherings; they enable our debaters to experience the contrasting debate style of the foreigners. In short, they are a rewarding experience for all concerned.
Much has been written about these encounters by both participants and debate directors. However, the impact of these debates on the attitudes of audiences who hear them as received little attention.1 We know relatively little about what attitude change, if any, occurs in the audiences exposed to this unique experience. In fact, we know relatively little about what happens to attitudes of audiences exposed to any contemporary debates.
The following is a report of a study of the effects of one international debate--a debate between two American debaters from the Pennsylvania State University and two English debaters from the Universities of Birmingham and Nottingham. The debate, held February 10, 1966, on the University Park Campus of Penn State, was on the proposition, "Resolved: That racial integration is an impossible ideal." The English debaters upheld the affirmative position.
It is important to stress at the outset that the effects on audience
attitude produced by this debate should not be taken as suggestive of the
probable effects of other debates. Although this debate was typical of
international debates in many respects, there were two characteristics
of this debate that marked it as atypical. First, the topic was a question
of fact rather than the usual question of policy. Second, by textbook standards
this was an unusually poor debate. Nevertheless, the debate generated considerable
excitement and enthusiasm in the audience and, as shall be noted below,
it produced statistically significant attitude change. The importance of
this study is not that the results indicate what attitude changes should
be expected from audience debates; rather, the study is important because
the results point up the need for much more research on the effects debates
have on the attitudes of audience members.
DESIGN OF THE STUDY
This study was not an experiment in the usual meaning of the term. There was no attempt to control or manipulate any variable. The participants in the debate were not aware an investigation was being conducted. As audience members arrived for the debate one hundred of them were handed a packet of material. This packet included instructions, attitude measures to be completed before the debate, and attitude and source credibility measures to be completed after the debate. Eighty-eight usable packets were returned at the end of the debate. All responses were anonymous.
Attitudes on four concepts believed related to the debate proposition were measured before and after the debate. These concepts were "Racial integration," "Negro," "school segregation," and "civil rights demonstrations." The attitude measures were of the semantic differential type. Scales used were good-bad, beneficial-harmful, right-wrong, fair-unfair, positive-negative, and wise-foolish. The attitude measures were pretested in conjunction with previous research.2 This pretesting indicated that the measures represented only the "evaluative" dimension of attitude.
Semantic-differential measures of source credibility for the two debate teams were taken after the debate. These measures had been extensively pretested in previous research. This pretesting indicated that two consistent dimensions of source credibility would be obtained, authoritativeness and character.3 Factor analysis of the subjects' responses confirmed this expectation.
The normal international debate format at Penn State includes use of
the Woodward Shift-of-Opinion Ballot4 to determine the "winner."
All members of the audience received ballots. Two hundred fifty-six ballots
A group of seven former college debate directors present at the debate were in unanimous agreement that (1) the negative team failed to meet the issues introduced by the affirmative, (2) the negative team failed to establish any significant point that withstood affirmative refutation, (3) the affirmative team, by textbook standards, was the overwhelming winner, and (4) both teams expressed only favorable attitudes toward racial integration and the desirability of working to increase racial integration in society. The group of former coaches also agreed that the affirmative position was probably perceived as "achievement of complete racial integration is impossible, but we should continue to do all we can to achieve as much integration as is possible," and the negative position as "just because racial integration is very difficult to achieve is no reason that we should not continue to do all we can to achieve as much integration as possible." These positions, of course, do not represent a clash. They are, in fact, almost identical views. The subjects were not asked to indicate what they perceived to be the positions of the two teams. We cannot, therefore, be certain that the subjects perceived the debate as the former coaches thought they did. However, comments by students in assigned papers required by some teachers in the basic speech course were generally consistent with the former coaches' expectations.
The results of the Woodward Shift-of-Opinion Ballots, which were used
to determine the winner of the debate, indicated a large shift in favor
of the affirmative. One hundred fifteen people shifted in favor of the
resolution, 59 shifted in opposition to the resolution, and 82 recorded
no shift. Table 1 reports the opinions of the audience members before and
after the debate. Because of the wording of the resolution, these results
give us no indication of the attitude change toward concepts related to
racial integration. As indicated in the debate, it was quite possible to
support racial integration and either favor or oppose the stated resolution.
Thus, although the former debate directors felt the affirmative won the
debate and the audience voted overwhelmingly for them, the data do not
indicate what happened to audience attitudes as a result of being exposed
to this debate.
Woodward Shift-of-Opinion Results
|Position on the Resolution||Before the Debate||After the Debate|
|Favor (or strongly favor) the Proposition||58||110|
|Oppose (or strongly oppose) the Proposition||130||112|
EFFECTS ON AUDIENCE ATTITUDES
Since both sides in the debate expressed favorable attitudes toward
racial and hostile attitudes toward racial segregation, it would be reasonable
to expect that if any attitude change occurred on the four attitude concepts
measured, the change would be in the direction of greater approval of integration
and greater hostility toward segregation. As indicated in Table 2, opposite
changes occurred. On two of the concepts, statistically significant attitude
change was observed. The audience became slightly less favorable toward
the concept "Negro," and shifted from a position of hostility
toward 'school segregation" to a position near neutrality. No statistically
significant shift was observed on the concepts "racial integration"
and the "civil rights demonstrations."
Attitude Change Results Concepts Related to Racial Integration*
|Score Before the Debate*||Score After the Debate||
|Racial Integration||88||35.75*||35.86||.11||.172, p > .05|
|Negro||88||31.16||30.24||.92||2.280, p < .05|
|Civil Rights Demonstrations||88||25.83||26.15||.32||.727, p > .05|
|School Segregation||88||18.66||23.52||5.86||3.084, p < .01|
*A high score represents a favorable attitude. The scores on the measure could range from 6 to 42. The hypothetical neutral point is 24.0.
**Test based on correlated data.
Former debate directors felt the affirmative did a far better job of
debating than the negative. The audience voted overwhelmingly for the affirmative.
In addition, students' comments in assigned papers required by some teachers
in the basic speech course generally endorsed the British team as far superior
to the local team. Thus, one might suppose that the British speakers were
also perceived as more credible sources than the Penn State team. The striking
facts were (See Table 3) that there was no significant difference between
the two teams on either the authoritativeness or character dimensions of
credibility. Both teams were highly rated on both dimensions.
|Credibility Dimension||N||British Team||Penn State Team||D||t**|
|Character||88||33.74*||33.35||.39||.690, p > .05|
|Authoritativeness||88||34.61||34.35||.26||1.872, p > .05|
*A high score represents high perceived credibility. The scores on each measure could range from 6 to 42. The hypothetical neutral point is 24.0.
**Test based on uncorrelated data.
The observed effects of the debate on the audience's attitudes, then,
were contrary to what would seem on the surface to be reasonable expectations.
Both sides spoke in favor of racial integration and in opposition to racial
segregation, but audience attitudes shifted the opposite direction on two
related concepts. The debating skill of the two teams was perceived as
markedly unequal, but the teams were given equal credibility ratings. Why
did this apparently strange result occur?
INTERPRETATION OF RESULTS
There is no clear explanation of the observed effects this debate had on audience attitudes. Whether we base our attitude change expectations on traditional rhetorical theory or on the modern tension-reduction theories, the results simply do not conform to what would be reasonable predictions. Probably the most appropriate interpretation was suggested by one of our colleagues. "A mess of forces tends to produce a mess of results."
As well as this interpretation describes the rhetorical event studied, it is not an interpretation that we can comfortably accept. Any debate may be described as "a mess of forces." How then can we predict what will be the effect of any debate on audience attitudes? Such predictions are crucial to many studies of debates in history. The implications are clear and serious.
Several questions are raised by this study. Were the observed results unique? How often does this type of attitude change occur in exhibition debates? Was the nebulous wording of the resolution the cause of the observed results? Can this kind of "strange" result be expected when resolutions of fact are debated? Are the results of "real life" debates comparable to those of exhibition debates?
We have no answers to these questions, but these are questions for which answers are needed. The study reported above was prompted primarily by curiosity. We wondered what happened to audience attitudes as a result of exhibition debates. Our curiosity has changed to concern. The observed attitude changes on the concepts measured were not what we expected, nor could they even be "predicted" on a post hoc basis. Many critical analysis of debates in history have been written which presumed that the researcher could accurately determine the effect of the debate on audience attitudes. If the effects on audience attitudes of a single one-hour exhibition debate cannot be predicted, some may question the accuracy of estimated effects of historical debates.
Our concern, of course, may be unjustified. The particular debate studied
may have been unique. The observed attitude changes could have been chance
variations. The audience for the debate may have had some elusive but vital
characteristics that made their responses atypical. We do not know. The
only clear conclusion from this study is that further research on the effects
of contemporary debates on audience attitude is badly needed.
1. References to the relevant literature on international debating can be found in Arthur N. Kruger, A Classified Bibliography of Argumentation and Debate (New York, 1964), under "British Debate." Kruger lists one book and 43 articles.
2. James C. McCroskey, "Experimental Studies of the Effects of Ethos and Evidence in Persuasive Communication," unpublished dissertation (Pennsylvania State University, 1966).
3. Ibid., see also James C. McCroskey, "Scales for the Measurement of Ethos," Speech Monographs, XXXII (March, 1966), 65-72. Scales representing the authoritativeness dimension were reliable-unreliable, informed-uninformed, qualified-unqualified, intelligent-unintelligent, valuable-worthless, and expert-inexpert. Scales representing the character dimension were honest-dishonest, friendly-unfriendly, pleasant-unpleasant, unselfish-selfish, nice-awful, and virtuous-sinful.
4. Howard S. Woodward, "Measurement and Analysis of Audience Opinion," Quarterly Journal of Speech, XIV (February, 1928), 94-111. See also Alan H. Monroe, "The Statistical Reliability and Validity of the Shift-of-Opinion Ballot," Quarterly Journal of Speech, XXVII (December, 1937), 577-585.
Click Here To Go Back To PERIODICALS