"Communicator Discrepancy, Source Credibility, and Opinion Change"
By: Stephen Bochner and Chester Insko

Stephen Bochner, University of New South Wales, and Chester A. Insko, University of North Carolina, attempt to prove the relationship between discrepancy, credibility, and opinion change via an experiment on sleep. The authors incorporate the Judgmental Theory and the Dissonance Theory into their experiment. They believe communication is an "external anchor that produces increasing positive influence with moderate discrepancy and decreasing influence with extreme discrepancy" in a given situation. In other words, the amount of positive influence and moderate discrepancy are directly related and similar. However, as influence decreases, extreme discrepancy will be at an extreme level. Theorists believe that dissonance (a difference of opinion) can be reduced in four ways:

Conformity to the communicator's point of view (Bochner, Insko. p. 614). In this study, conformity is the preferred avenue of dissonance reduction at the slight and moderate levels of discrepancy, but not extreme discrepancy. Theorists have suggested the reason for the problem in the extreme case was due to the fact there was inconsistency from previously held cognitions.

Disparagement of the communicator (Bochner, Insko. p. 614). This reduction method relies heavily on the communicator. It relates to the amount of credibility a communicator currently has.

Persuasion of the communicator that (s)he is incorrect (Bochner, Insko. p. 614). Communication persuasion can only be applied to situations in which subjects are allowed to socialize with each other. In this research experiment the subjects were not allowed to talk to each other, so this reduction method does not apply.

Obtained social support from other like-minded individuals (Bochner, Insko. p. 614). Social support cannot be offered due to the fact that the subjects are not permitted to talk to each other in this study. Lack of communication makes this reduction presently inapplicable also.

The two authors used experimental research to attempt to provide results linked to the amount of discrepancy and disparagement needed for an opinion change. They chose the topic of reducing the number of hours an individual sleeps in one night. They had two reasons for choosing this topic: it has a finite ceiling and the majority of college students agreed on the amount of sleep needed in one night. Upon choosing a topic, the two men developed three dependent variables to use in the experiment.

Three dependent variables were used to measure the received data. They were: "opinion regarding sleep, disparagement of the communication, and disparagement of the source." To gain feedback about opinion, they asked the participants "For maximum health and well being, how many hours of sleep per night do you think the average young adult should get?" Participant answers ranged from 0 to ten hours per night. The other two variables were measured on six 7-point scales that involved questions previously asked to participants. After establishing the dependent variables, they moved onto independent.

The authors also chose to manipulate three independent variables: "discrepancy (nine levels), sources credibility (high or medium), and order of dependent-variable measurement (opinion-disparagement or disparagement-opinion)." The authors manipulated the situation by inserting a number and then using the amount in a persuasive communication style. As for credibility, the participants were asked whether "Sir John Eccles, Nobel prize winning physiologist" or "Mr. Harry J. Olsen, director of the Fort Worth YMCA" had more communication credibility. The mean for Mr. Harry J. Olsen was considerably higher. The two men had a certain procedure that they followed to reach the desired results.

The experiment started out with five hundred and seventeen students from the University of Hawaii who were enrolled in introductory education and psychology courses. Each participant was given a testing booklet that consisted of: a front page (outlining the experiment and instructing the students to act in the capacity of a normative population), the second and third pages consisted of practice pages, then the communication persuasion pages, and the final pages addressed an evaluation of the experiment. 36 different books were needed in this scenario in order to ensure honesty in a classroom setting. Out of the beginning 517, 13 were eliminated either because of not completing the questionnaire entirely or to create an equal distribution. From the remaining 504 students, the two men were able to gather data and come to a few conclusions.

The two men concluded that their hypothesis was confirmed only for the medium amount of credibility, as opposed to the medium and high amount. From a theoretical standpoint, the highly credible person would have scored equal to or higher among the students than the person who was only of medium credibility. The authors stand firm on their conclusion of "medium and high credibility sources do not differ at the moderate discrepancy levels, but that the high credibility source is superior at the extreme discrepancies."

Opinion change is a much harder concept to grasp and change in one attempt. The testing and authors concluded that source disparagement is more important than communication disparagement when dealing with an opinion and attitude change. The discrepancy manipulation must be powerful enough to sway the individual and produce source disparagement. Accurate opinion changing experiments in the future should analyze this experiment and manipulate their variable in a different manner to reach a desired outcome.

Scholars might examine this application/experiment and conclude that the end results are problematic. In the beginning of this experiment, students were recruited to be subjects. They could be considered as convenient samples due to the fact that they were chosen to represent a certain population of people. The majority of the students chosen were located in the same geographic region and were only chosen due to the fact that they were enrolled in introductory education and psychology courses. The lack of people representing the population in this test does not allow for statistical power to be significant to the rest of the population or other tests. It has been said that at least 3000 people are needed to represent the population of the United States. The authors of this article did not acknowledge the problems from a population standpoint, but rather found problems in their scales that were used. The scales were set up in a manner that caused them to suspect discrimination in regards to hours of sleep. The authors realized the error and will attempt to make the next study more psychologically meaningful.

In conclusion, this method of testing has many different possibilities when first starting out in the beginning. An experimenter must examine all possible outcomes and work to manipulate the subjects to reach a hypothesis. The success of one's experiment is determined by the height of the finite ceiling and the subject being dealt with.

Citations

S. Bochner and C. Insko, "Communicator Discrepancy, Source Credibility and Opinion Change," Journal of Personality and Social Psychology, Vol. 4, 1966, pp. 614-621.



Social Judgment Analysis Page Click to read more about the Social Judgment Theory in an online analysis of the theory.
Return to our homepage.

The Relational Dialectics Theory is similar to Social Judgment. This link leads you to research done by Natalie Wright on the related theory.

This page last updated on 03/11/00 by Miranda Valentine.