Lacking omniscience, I cannot predict which, if any, of the five exploratory works will capture the imagination of tomorrow's persuasion researchers or furnish the impetus for major scientific accomplishments. One thing is certain: opportunities for exploration and accomplishment are there for ambitious scholars of persuasion to seize. Hopefully, the books reviewed above will provide them with some of the conceptual, methodological, and empirical tools necessary to expand the frontiers of knowledge about the persuasion process-a process central to man's continued existence and betterment. BOOKS REVIEWED ATTITUDE AND ATTITUDE CHANGE. By Carolyn W. Sherif, Muzafer Sherif, and Roger Nebergall. Philadelphia: W. B. Saunders Company, 1965; pp. v+264. $11.45. ATTITUDE CHANGE. By Charles A. Kiesler, Barry, E. Collins, and Norman Miller. New York: John Wiley & Sons, Inc., 1969; pp. vii+386. $8.95. ATTITUDE, EGO-INVOLVEMENT, AND CHANGE. By Carolyn W. Sherif and Muzafer Sherif. New York: John Wiley & Sons, Inc., 1967; pp. v+316. $9.95. ATTITUDE FORMATION AND CHANGE. By J. D. Halloran. Leicester, Great Britain: Leicester University Press, 1967; pp. 167. $4.00. ATTITUDE. By Marie Jahoda and Neil Warren. Baltimore, Maryland: Penguin Books, Inc., 1966; pp. 375. $1.95. BELIEFS, ATTITUDES AND HUMAN AFFAIRS. By Daryl L. Bem. Belmont, California: Brooks/Cole Publishing Company, 1970; pp. v+114. $2.25. BELIEFS, ATTITUDES AND VALUES. By Milton Rokeach. San EXPERIMENTS IN PERSUASION. By Ralph L. Rosnow and Edward READINGS IN ATTITUDE THEORY & MEASUREMENT. By Martin Fishbein. New York: John Wiley & Sons, Inc., 1967; pp. v+499. $10.95. ROLE PLAYING, REWARD, AND ATTITUDE CHANGE. By Alan C. Elms. New York: Van Nostrand Reinhold Company, 1969; pp. iii+232. $2.50. THE STUDY OF ATTITUDE CHANGE. By Richard V. Wagner and John L. Sherwood. Belmont, California: Brooks/Cole Publishing Company, 1969; pp. v+213. $3.25. THEORIES OF ATTITUDE CHANGE. By Chester A. Insko. New York: Appleton-Century-Crofts, 1967; pp. v+374. $7.50. A MODEL OF COMMUNICATION EFFECTIVENESS* An effort to develop a mathematical model of interpersonal and mass media Since Shannon and Weaver's [1] introduction of a mathematical theory of communication two decades ago, many speech and communication theorists have acknowledged that it would be beneficial to be able to relate quantitatively the factors that compose the communication process. However, the absence of any simple mathematical model of interpersonal and mass media communication effectiveness is not surprising. Most speech and communication theorists shy away from any commitment to a rigid, quantitative approach to an area with so many nebulous variables; nevertheless, several have hesitatingly explored the applications of such a model. For example, Harrah [2] made a definite contribution with his model, which is based primarily on mathematical set theory, and Cherry [3], Osgood [4], Stephenson [5], and others have recognized the desirability of measurement and quantification of some of the variables that are the determinants of effectiveness. But a simple mathematical model, which could improve the understanding and prediction of communication effectiveness, has seemed to elude most scholars in the field. Abelson [6] probably provides the best quantitative insight into the relationships among message determinants of effectiveness in a computer simulation model, but even he does not present a symbolic representation of effectiveness. Although Abelson does not develop a desirable codification for effectiveness, his propositions help clarify the direct and inverse mathematical variations. For example, he makes these axiomatic statements: Individual i will be more apt to accept assertions made by s (sender) the more favorable i's attitude towards and the higher i's receptivity to s. An assertion is especially apt to be accepted by if it is consistent with his predisposition toward that assertion, and under no circumstances will be accepted if it runs counter to his predisposition. An assertion is less apt to be accepted by i if it is inconsistent with his position on the issue. When i's attitude toward s is negative and his receptivity to s is very low, assertions made by s not previously encountered by i and not consistent with his position will tend to promote acceptance by i of converse assertions. The direction of such attitude position change is toward s if i's attitude toward s was initially positive, and away from s if i's attitude towards was initially negative; the degree of such change is a direct function of the difference between the positions of i and s [6]. *Excerpts from "A Model of Communication Effectiveness," The Journal of Communication, XX (March 1970), pp. 81-91. Reprinted with the permission of The Journal of Communication, copyright holder and the courtesy of the author. In transferring the dependent and independent variables into information processing language, Abelson has attempted to satisfy four desiderata: (1) Influence will be successful (i.e., new assertions will be accepted) under a fairly wide range of conditions, particularly when the influencee has a mild position on the issue but is sufficiently interested to have exposed himself to ...stimuli. (2) However, there are definite resistances to influence, particularly when the influencee has a strong position on the issue and, or a negative attitude toward the communicator. (3) If resistances are very strong, there may be a "boomerang effect" whereby the influencee accepts new assertions opposite to those intended and a new position farther away from that of the communicator. (4) However, there is some way of influencing almost everybody to some degree [6]. These four criteria and the 49 axiomatic statements were freely consulted in designing the theoretical framework of the proposed model, which mathematically relates the variables of the communication process, including the amount of information, the sender and receiver's places in the social-psychological system both real and preceived, and the effectiveness of the message, in a simple straightforward manner. The communication represented in the model is restricted to a persuasive message, and the destination of the message must be an individual instead of a machine. A horizontal line, designated AB and composed of many sub-continuums, is used to signify the master continuum on which every person finds himself in relationship to all other people. For example, a sub-continuum a, b, might be considered as the limits of all political positions in this country from "far left" to "far right." Each individual can be placed somewhere on a, b, relative to all others in the population. The political spectrum can then be represented like this: In every case, a person's position on this hypothetical scale is determined by his socioeconomic status, group memberships, heredity, and all the experiences that go to make each one of us different. If an individual's position relative to others can be placed on a hypothetical scale for politics, it can be placed on a hypothetical scale that stands for degrees of differences from "far left" to "far right" for each possible existing variation affecting our relationship with others. The summation of all these hypothetical scales, each weighted according to its importance in influencing the real position of the receiver, is the continuum AB, which can be stated symbolically as: AB = f1 (a,, b1) + f2 (a2, b2) + ... fn (an, bn). A prospective receiver's position on this line is not static, because the individual is in a constant state of readjustment of his position. Many of these changes are so minute that they are quite immeasurable with existing methodology. The forces, which must assume a state of equilibrium with respect to the AB component, determine an individual's location on the continuum and are signified by R1, R2, R3, ... Rn. When a state of imbalance exists the receiver will readjust his position until the AB component reaches required equilibrium. Generally speaking, this portion of the model is in agreement with the tradition of the "balance" theorists, including Heider [7], Newcomb [8], and Festinger [9]. The sender's position on the AB continuum is at a point where it is preceived by the receiver. The subject of the message, and in turn the content used to support it, may be two of the strongest influences determining the social and psychological distance, as perceived by Q. Other variables, including the speaker's introduction, his ethos, personality, etc., influence the distance between the real position of the receiver and the perceived position of the sender significantly. In addition to the horizontal component, each of the forces has a vertical component whose total CD is defined as Q's predispositions on some subject JK. C A. D B As a result of forces R1, R2, R3, ... Rn, there will be predispositions, the CD component, either in the positive direction, in the negative direction, not at all, or in both directions. In reality a CD component probably will not exist for each and every force, since the CD component is present only when the force R, affects an individual's predispositions on subject Jk. For example, membership in a car pool may have absolutely no effect on religious predispositions, and yet, this same membership may have great effect on political predispositions, depending on the topics of discus sion, other group memberships, and a finite number of other variables. From the model it becomes clear that predispositions are completely independent of the social and psychological distance, because prior attitudes can only be determined by the real position of the receiver, a resultant of the R, forces. In other words, predispositions are not altered in intensity by the perceived s-p distance from the sender. Since receiver Q cannot move in the CD direction, neither are the predispositions toward subject Jk of opposite signs necessarily equal. Assume that two individuals are located on the scale AB with S the sender and Q the receiver, keeping in mind that the position of S is important only as perceived by Q. Then, assigning the letter Y as the difference on the continuum between the receiver and the perceived position of the sender, the formula becomes: Also let us assume that a message of magnitude M is sent by the sender S and received by Q. S A Y X - B This message is either reinforced or depreciated by the resultant of the positive and negative predispositions, and therefore, 1 2 P1 and P2 are the magnitudes of the positive and negative predispositions. The positive direction is selected as being in the direction of the intended effect as preceived by the sender. For example, if the sender is attempting to gain votes against a proposed ordinance, the direction remains positive so long as that is the direction of the effect intended by the sender. Likewise, negative dispositions refer to those predispositions whose force is in the direction opposite to the intended effect of the message. Now, with the relationships among message, predispositions and the sender's perceived location on the continuum defined, it is easy enough to define the effectiveness of the communication M. This effectiveness is the ratio X Since mathematically, Tan = (X/Y) and 0 = Arctan (X/Y), the degree of effectiveness of a communication is defined as the number of degrees in the angle 0, and the angle @ is referred to as the Angle of Effect. The size of the angle results directly from combining quantitatively the factors generally accepted as influencing the effect of the message. |