AN EMPIRICAL TEST OF THE RELATIONSHIP OF‘EVIDENCE TO BELIEF SYSTEMS AND ATTITUDE CHANGE Thesis for the Degree of M. A. MICHIGAN STATE UHIVERSITY B. THOMAS FLORENCE 1974 IIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIIILIIIIIIIIIIIIII 3 1293 10464 Emmi-I MSU LIBRARIES Va... RETURNING MATERIALS: Place in book drop to remove this checkout from your record. FINES wiII be charged if book is returned after the date stamped beTow. W "32325? /‘)57 ABSTRACT AN EMPIRICAL TEST OF THE RELATIONSHIP OF EVIDENCE TO BELIEF SYSTEMS AND ATTITUDE CHANGE BY B. Thomas Florence Though rhetorical theories clearly prescribe the function of evidence in forming individual beliefs, communication research aimed at testing the empirical validity of such prescriptions has been inconclusive. Due to questions surrounding the conceptual and operational adequacy of previous research in this area, it was the purpose of this inquiry to: first, critically examine the existing theory and research surrounding the evi- dence-attitude change relationship; second, reformulate existing theory based on the critique; third, test the empirical validity of the theoreti- cal reformulation; and finally, discuss the implications of this research to theory and subsequent research. Utilizing a theory of persuasion suggested in the writings of Richard Weaver, the research focused on investigating the effects of credibility and evidence usage on desirable and rational beliefs and their subsequent effects on attitude change. The study was in the form of a quasi-experimental design in which students from 15 undergraduate classes at Michigan State University were sampled at two points in time. Using multiple regression statistical procedures, all of the hypoth— esized causal relationships within the model were supported by the data, and all but one of the contingent causal relationships were supported. 4 //7“;~‘/c}p AN EMPIRICAL TEST OF THE RELATIONSHIP OF EVIDENCE TO BELIEF SYSTEMS AND ATTITUDE CHANGE BY B. Thomas Florence A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Communication 1974 Accepted by the faculty of the Department of Communication, College of Communication Arts, Michigan State University, in partial fulfillment of the requirements for the Master of Arts degree. W WWW/r Director of Thesis \ Guidance Committee: W (“fl/9&1 ,Chairman NW ‘54de ACKNOWLEDGMENTS I wish to express my sincere appreciation and indebtedness to my committee for awakening in me the desire for scholarship. To each I extend my gratitude: Dr. Gerald Miller, chairman of my committee, whose intellectual guidance has taught me the function of scholarship. Mr. Donald Cushman, whose friendship and intellectual rigor has taught me the meaning of scholarship. Dr. Joseph Woelfel, whose encouragement and insight has taught me the goal of scholarship. In addition I would like to thank: Louie Bender and Norm Fontes for their sincere interest and advice. Ted Smith for breaking ground in the area of this thesis. Dr. Edward Fink for his methodological assistance. Glee for caring. ii TABLE OF CONTENTS Chapter I INTRODUCTION ..... ......... Statement Of the PrOblem. O O O I O I O O O O O O O O O O O O O O O O O O O O Critique of Theoretical Formulation of Evidence- Attitude Change Relationship...... .......... .... Theoretical Reformulation................... II PROCEDURES........ ..... .... Definitions ................................... ..... Design.... ........ ..... III RESULTSOOOO. ..... O OOOOOOOOOOOOOOO .0... ......... IV DISCUSSIONOOOCOOOOOOOIO... FOOTNOTES 0 O O O O O O O O O O O O O O O O O O O O O O O O 0 APPENDIX A. O O O O O O O O O O O O O O O O O O O O O O O 0 APPENDIX B O O O O O I O O O O O O O O O O O O O O O O O O 0 LIST OF REFERENCES.......... ..... .. 00......OOOOOOOOOOOOOOOOOOCO iii \JUJ 14 14 17 19 25 29 31 33 68 LIST OF TABLES Table l Standardized partial regression coefficients, multiple correlation coefficients, and coefficients of determina- tion of desirable belief regressed on the predictor variables (N=161)............. ..... ..................... 2 Standardized partial regression coefficients, multiple correlation coefficients, coefficients of determination, and increments in the coefficient of determination of truth belief regressed on the predictor variables (N-161).................................... ....... . ..... 3 Standardized partial regression coefficients, multiple correlation coefficients, coefficients of determination, and increments in the coefficient of determination of attitude change regressed on the predictor variables (N=16l)................................................. iv Chapter I INTRODUCTION Much communication theory focusing on the relationship between evidence and belief has a long history of coordinate development in rhe- torical theory (Allport, 1969). Based on certain metaphysical assumptions of man, traditional rhetorical theory holds that the merits of a speaker's claim depend on the merits of the evidence or argument on which the claim is based. Theories of evidence cast within this frame are largely pre- scriptive. And whether these be prescriptions regarding the maxims of psychological reasoning or standards of achievement by which justifica- tory arguments may be assessed, these theories assume the importance of evidence in the establishment of individual belief systems. Though rhetorical theories clearly prescribe the function of evidence in forming individual beliefs, communication research aimed at testing the empirical validity of such prescriptions has been inconclusive. Due to questions surrounding the conceptual and operational adequacy of previous research in this area, it will be the purpose of this inquiry to: first, critically examine the existing theory and research surrounding the evi- dence-attitude change relationship; second, to reformulate existing theory based on our critique; third, to test the empirical validity of our the— oretical reformulation; and finally, to discuss the implications of this research to theory and subsequent inquiry. Statement of the Problem The problematic nature of the evidence-attitude change relationship is reflected in the results of previous empirical research. Though philo- sophically appealing arguments for the effectiveness of evidence are over- whelming, existing research does not facilitate conclusions concerning the isomorphism between theoretical prescriptions and empirical descriptions. A review of prior research reveals a balance between studies supporting and those not supporting the preposition: evidence functions to establish and maintain individual attitude and belief systems. Of the research investigating the effects of evidence-plus-assertion and assertion-only speeches on attitude change, Cathcart (1955), Bostrom and Tucker (1969) and Kline (1969) offer support for the relationship be- tween evidence and attitude change; whereas research conducted by Costley (1958) and Wagner (1958) found no sudh relationship. In studies investigating the relationship of evidence and source cred- ibility to attitude change, Cathcart (1955) and Bostrom.and Tucker (1969) found that speeches containing evidence attributed to qualified authorities were significantly more persuasive than speeches containing only unattri— buted assertions; whereas Ostermeier (1967) and Whitehead (1971) found no such relationship. In addition, though Bettinghaus (1953) discovered that speeches containing assertions supported by evidence and qualified author- ities produced significantly more attitude change than speeches containing only assertions supported by evidence, research by Gilkinson, Paulson and Sikkink (1954), Sikkink (1956) and Bostrom and Tucker (1969) found no such effect. This brief review suggests that empirical research surrounding the evidence—attitude change relationship is at best inconclusive. In an 3 attempt to reconcile the conflicting nature of prior research it is necessary to critically examine the theoretical foundations of such research. Critique of Theoretical Formulation of Evidence-Attitude Change Relationship Though prior research focusing on the evidence-attitude change rela- tionship is conceptually linked to traditional rhetorical theory, the full implications of this link are, in many instances, inadequately developed. t! In translating from the prescriptive language of rhetorical theory, where J relationships are specified in terms of nominal categories, to the descrip- tive language of scientific theory, where relationships are specified in terms of quantitative operations, two potential problem areas emerge: (l) a failure to distinguish between different forms and functions of proof, and (2) a failure to distinguish between different types of beliefs. Historically, rhetoric has been viewed as the counterpart of dialec- tic, and as such, aims at securing judgements in the realm of the contin- gent. Thus, at the very center of rhetorical theory, is the idea of an audience as a judge, as one who is to exercise the act of judgement. The act of judgement functions in the domain of practical reasoning or delib— eration. Rhetoric does not effect persuasion in the sense of manipulation, as is often assumed to be the case, but rather creates an atmosphere wherein the judge or audience may make its own decision (Grimaldi, 1958). Within this context, traditional rhetorical theory distinguishes three general dimensions of proof. The first depends on the personal character of the speaker (ethos). The second focuses on putting the audience in a certain frame of mind (pathos). The third and final dimension of proof focuses on the truth or apparent truth of the arguments suitable to the case in question (logos). The majority of previous research, though 4 espousing a rhetorical base, has often failed to make the distinction between evidence functioning as proof of the second type (often referred to as emotional proof) and evidence functioning as proof of the third type (often referred to as logical proof). There is no reason to assume that the form of each type of proof is the same. In fact, rhetorical theory distinguishes between proof which argues for the truth of a given proposi- tion by presenting evidence in favor of the probable truth of the premises from.which the proposition flows, and proof which seeks to establish the :“7‘ desirability of a given proposition by appealing to the passions or desires from which support of the proposition flows (Aristotle, Nichomachean Ethics, [ “ 1356a).1 However, in previous empirical research, classifications of evi— I dence have been made solely on the basis of criteria such as fact versus opinion, presence or absence of the qualifications of the source, and log- ical consistency, rather than on the basis of its form in relation to some function which it serves. In addition to problems associated with the failure to distinguish between differing evidential forms, a more crucial problem may result from a failure to distinguish between the separate functions of proof and the goals associated with each. Evidence functioning as emotional proof seeks to create a desirable belief toward the proposition in question. In other words, in traditional rhetorical theory, emotional proof acts to establish a judgement on the part of the audience as to the desirability of a given proposition or its consequences. Likewise, evidence functioning as logical proof seeks to establish a rational belief toward the proposition in ques- tion. Put another way, logical proof acts to establish a judgement on the part of the audience as to the validity or truth value of the premises from which the proposition flows. There is no reason not to assume functional 5 differences with regard to proof; however, previous researchers, by failing to make distinctions regarding the functions and goals of evidence, and instead studying the effects of evidence "in general," or structural forms of evidence, on attitudes "in general," have made it difficult to inter- pret research results so as to reconcile conflicting findings. In addition to problems associated with the theoretical formulations of the evidence-attitude change relationship, coordinate problems exist within the realm of conceptual and operational definitions. These problems associated with inadequate operational and conceptual definitions in prior evidence research are of three basic types: (1) inadequate conceptual and operational definitions of "evidence" and "no evidence", (2) inadequate operational definitions of belief, and (3) inadequate operational defini— tions of source credibility. Previous researchers have conceptually defined evidence as consisting of three constituent parts: (1) evidential data; (2) source of the evi- dential data; and (3) the qualifications of the source of the evidential data. Thus any piece of evidence may vary along any of three dimensions, any or all of which may be responsible for observed persuasive effects. For example, Ostermeier (1966) referred to evidence as "a reference in which the communicator reveals and stresses a personal association with others who have had firsthand experience with the topic" (p. 138). McCroskey (1969 and 1970) defined evidence as "factual statements origina- ting from a source other than the speaker, objects not created by the speaker, and opinions of persons other than the speaker that are offered in support of a speaker's claim” (p. 170). Each of these definitions sug- gest a variety of distinct dimensions of evidence which have persuasive potential (e.g., personal association with source, qualifications of 6 referenced source, factual or opinionated evidence, etc.), and which make it impossible to be sure which dimensions are or are not producing an ob- served persuasive effect. Similar problems exist for the definitions provided for no evidence. In previous experiments, the only factor that differentiated an evidence condition from.a no evidence condition was the exclusion of the referenced source in the no evidence condition. It should be noted however, that the data, whether fact or opinion, were still present in the no evidence condi- tion. The presence of these data may confound the results of the experi- ment in a number of ways. First, it is possible that the data themselves can be viewed as evidence capable of changing attitudes or affecting the credibility of the speaker. Second, as McCroskey (1970) found, an unseen, unknown source may be perceived as a highly credible source, either by association with a highly credible source, i.e., the speaker, or as balance theory would suggest, by association with an accepted opinion. Third, manipulation of the source and his qualifications may significantly alter the length of the persuasive message. It is possible, therefore, that a no evidence condition that only excludes the reference source, may act as an evidence condition either because of the presence of data or the exis- tence of an implied highly credible source, or may result in findings that are artifacts of differences in the lengths of treatment messages. Problems associated with the operational definitions of belief and credibility stem from the same issue regarding measurement utilizing se- mantic differential scales. Previous research investigating the relation- ship of evidence to belief and the relationship of the evidence-credibility interaction to belief has relied heavily on measures utilizing semantic differential scales. However, research indicates that the comparability of such measures may be in question. As OSgood, Suci, and Tannenbaum have pointed out, What information we have available . . . indicates an encouraging degree of comparability across the subject. The situation is quite different with respect to com- parability across concepts. Ideal or perfect compara- bility here would require that individual scales main- tain the same meaning, and hence the same intercorrela- tions with other scales, regardless of the concepts being judged. This condition can definitely be shown not to hold (p. 176). In recent research by Bauduin and Davis (1972), this conclusion was sup- ported empirically. Bauduin and Davis concluded that it is invalid to assume comparability of semantic differential scales across concepts. In other words, the research shows a definite concept-scale interaction. Since the evidence, topic and source manipulations in previous research were not common within or across studies, it is possible that conflicting results could be a function of shifts in attitude and credibility dimen- sions within and across studies. Theoretical Reformulation Prior to addressing the question of the specific relationship of evi- dence to attitude change, we must briefly delineate the general theoretical context from which the relationship flows. We will do this by first dis- tinguishing between attitudes, beliefs and values; second, by defining the function of persuasive argument within this context; and third, by speci- fying the function of evidence in persuasive argument. Underlying the following formulation is the view that an understanding of the evidence- attitude change relationship is contingent upon an understanding of the distinction between attitudes, beliefs and values, and their functions in forming an integrated cognitive system. 8 Our conception of the distinction between attitudes, beliefs and values is based on the notion of a cognitive hierarchy. Within this hier- archy the highest level will be referred to as the individual's value system. Values are defined as the individual's abstract ideals, not tied to any specific object or situation, representing a person's conception of the ideal modes of conduct and ideal terminal goals (Rokeach, 1968). Within this context, values provide a system for clarification and organi- zation which assigns significance to physical and social reality, and guides and governs the formulation and maintenance of beliefs. Therefore, the value system functions to provide a set of principles for relating objects in the physical and social world.2 The intermediate level of the cognitive hierarchy will be referred to as the individual's belief system. Belief system is defined as the set of interrelationships of objects in physical and social reality held by an individual. Within this context, the belief system functions to guide and govern the formulation and maintenance of attitudes. Consistent with Rokeach (1968) belief can be said to have affective as well as cognitive components. However, unlike Rokeach this concept of beliefs does not allow the definition of a belief in isolation, but only in terms of the belief object's interrelationship with other objects within the belief system. The lowest level of the cognitive hierarchy will be referred to as the individual's attitude system. An attitude will be defined as the per- ceived relationship between an individual and some attitudinal object or situation predisposing the individual to respond in some preferential manner. Given this definition of attitude, we may define attitude system as a subset of the relationships which comprise the belief system; in particular, the attitude system is that subset in which one of the objects . .— 9 in the relationship is the self. This conception of attitude is compatible with previous conceptualizations (Rokeach, 1968) of attitude in that all attitudes are here assumed to be behaviorally oriented. Regardless of whether the attitude is said to be descriptive, evaluative, or advocative, it contains a predisposition which, when activated, will lead to a response, be it a verbal expression or some form of nonverbal behavior. Given this formulation, persuasion may be viewed as a situation in which the persuader is challenged to translate his reasons for selecting a preferred action into the overall value system of his audience in a man- ner capable of securing action (Cushman and Hauser, 1973). More specifi- cally, a persuasive appeal may be viewed as an attempt to present the sit- uation in such a way as to demonstrate that the proposed object or course of action falls within, or is consistent with, the individual's value sys- tem. Having done this the individual must accept the implicit or explicit conclusion or, turning back upon himself, deny the overarching value system from which it flows (Graham, 1925). Therefore, a persuasive appeal may be viewed as an attempt to establish a belief that the object/course of action stands in relation with a desirable (valued) object or goal (i.e., desir- able belief), and then to demonstrate the truth or strength of such a con- nection within the overarching value system of the individual (i.e., truth belief). Such a conception of the persuasive process has at least one very important implication. If we accept the hypothesis that values provide man with a system for organizing and assigning significance to empirical facts, then it seems to follow that empirical facts and arguments that are presented with the intent to persuade only become personally meaningful and fit material for rational activity when associated with a particular 10 object-object or action-consequence relationship. This implication may be more obvious if we consider the functions of evidence within this paradigm. It is important to note that this conceptualization of the persuasive process implies a change in the traditional way of viewing evidence. Pre- viously, the focus on evidence was a focus on the structural form of the evidence and how alterations in structural form affected attitudes. How- ever, the present conceptualization shifts the emphasis from questions of £2£m_to questions of function. In other words, instead of directing atten- tion to the structural components of evidence, this conceptual framework suggests directing attention to the functional components of evidence. Within such a paradigm the function of evidence is twofold. First, evi- dence functions to associate a particular object or course of action to a desirable object or consequence; and second, having done this, to demon- strate the strength or truth of such a connection. More specifically, evidence presented within a persuasive appeal functions to establish a belief that the prescribed course of action or object is linked to the previously valued object or the attainment of various goals or "valued" states. To accomplish this, evidence must: first, establish a belief in the desirability of a proposition by relating the proposition to the audi- ence's value system (desirable belief); and second, establish a belief in the truth of the proposition by demonstrating the probability or improba- bility that the particular relationship suggested in step one does exist (truth belief). This suggests that the truth value of a proposition only becomes a relevant consideration for the individual when the desirability of the proposition has been established. In addition, only when the de- sirability of the proposition has been established does the individual have a perspective within which to interpret empirical facts and arguments. 11 This does not imply that evidence as defined in previous research (i.e., empirical data and logically consistent arguments) exerts no persuasive influence. It does,however,suggest that evidence as previously defined, exerts influence only insofar as the proposition it intends to support is desirable to the audience. Previous research offers partial support for the preceding conceptu- alization. Studies by Smith (1949), Rosenberg (1956), Carlson (1956), and DiVesta and Merwin (1960) found that subject's judgements of the desira- l bility of a proposition (i.e., desirable belief) exert a significant in- attfl". 231.!“ _»-._ a .-9 ' - fluence on attitude change toward the proposition. In addition, research I4 by Knower (1935), Morgan and Morton (1944), Lefford (1946), and Sherif and Jackman (1966) supports the hypothesized link between judgements of the desirability of a proposition (i.e., desirable belief) and subsequent interpretation of evidential data and arguments supporting the proposition. Regarding the posited link between belief in the truth of a proposi- tion (truth belief) and attitude change toward the proposition, and belief in the desirability of a proposition (desirable belief) and attitude change toward the proposition, again previous research offers at least partial support for our conceptualization. Utilizing Fishbein's measures of: (a) the desirability of the consequences of a proposed course of action (de- sirable belief); and (b) the likelihood of the consequences of a proposed course of action (truth belief), Infante (1972) found that the multipli- cative interaction of desirability and likelihood judgements was a signi- ficantly better predictor of attitude change toward a proposition than desirability or likelihood ratings alone. It should be noted that Infante's _ research does not allow inferences concerning the causal sequencing of the variables studied. 12 Based on our reasoning at this point, we may now go on to specify, according to Zetterberg's (1965) classification scheme, a model of a per- suasive appeal, where the appeal consists of evidence cited by a source. (1) (2) (3) (4) (5) The relationship between evidence and desirable belief is assumed to be irreversible, deterministic, sequential, sufficient (assuming the evidence relates the proposition to the audience's value system), and necessary. The relationship between evidence and truth belief is assumed to be irreversible, stochastic, sequential, contingent (upon the desirability of the proposition), and substitutable. ’ The relationship between desirable belief and truth belief is assumed to be irreversible, deterministic, sequential, sufficient and necessary. The relationship between desirable belief and attitude change is assumed to be irreversible, deterministic, sequential, sufficient and necessary. The relationship between truth belief and attitude change is assumed to be irreversible, stochastic, sequential, contingent (upon the desirability of the proposition), and substitutable. In addition, in an attempt to reconcile differences in previous re- search regarding the combined influence of evidence and source credibility within a persuasive appeal, we will posit two additional relationships. (6) (7) The relationship between the credibility of the source of the evidence and truth belief is assumed to be irreversible, stochastic, sequential, contingent (upon the desirability of the proposition), and substitutable. The relationship between the credibility of the source of the evidence and desirable belief is assumed to be irreversible, deterministic, sequential, sufficient, and necessary. These relationships may be represented diagrammatically in a model of the influence of a persuasive appeal: where: X = credibility of the source of the evidence X2 = evidence X3 = desirable belief X4 = truth belief X5 = attitude change a Causal path -------- h-Contingent causal path (contingent upon the desirability of the proposition) Chapter II PROCEDURES Definitions This section will attempt to develop conceptual and operational definitions of the key theoretical constructs so as to capture the full implications of our model. This section will provide definitions for: (I) evidence, (2) desirable belief, (3) truth belief, (4) attitude change, and (5) source credibility. (1) As the previously delineated model would suggest, we will define evidence as any statement of fact, statement of value, or definition of- fered by a speaker or writer which is intended to support a proposition. Evidence may take the form of quotations from authorities, empirical data, logical deductions, analyses, etc.. It should be emphasized that such a formulation is founded upon the attempt of a speaker or writer to provide a justification for some proposed conclusion. This definition is consis- tent with the role of evidence implied by our model, i.e., as the founda- tion for a proposition to which the audience may refer in making their judgements regarding the proposition. Evidence, as a variable within this study, was manipulated by the experimenter through the use of differing message treatments. The evi- dence conditions were dichotomized into an "evidence" message and a "no evidence" message, with the evidence message designed to include components 14 aft—hf: ‘t . 15 appealing to truth belief and desirable belief toward the proposition. Both messages included: (a) a statement of the proposition, (b) a state- ment of the source of the message and his qualifications, (c) three asser- tions relating to the proposition, and (d) a conclusion calling for support of the proposition. The evidence message consisted of a statement of the proposition, a statement identifying the source and his qualifications, three assertions offered in support of the proposition each followed by either authority- based opinion statements or empiricallngrounded factual statements de- 3 signed to appeal to social values commonly held in the sample population, I and finally, a concluding statement calling for support of the proposition. The "no evidence" message consisted of a statement of the proposition, a statement identifying the source and his qualifications, three assertions offered in support of the proposition each followed by a paragraph defining one or more of the terms in thegproposition, and finally, a concluding statement calling for support of the proposition. (2) We have previously defined belief system as the set of interre- lationships of objects in physical and social reality held by an individual. Within this framework desirable belief is defined as a judgement regarding_ the relationship between an object or course of action and a valued object or oal. Desirable belief was operationally defined as the subject's response to the question, "How desirable is the proposition that 'The Council of Graduate Students at Michigan State University should be allowed to tax all graduate students $5.00 per year to cover the operating expenses of the organization?'", on a 5-point Likert scale ranging from very desirable to very undesirable. l6 (3) Consistent with our definition of desirable belief, truth belief is defined as a jgggement regprdipg_the_probability or improbability that agparticular relationship between one object or course of action and an- other object or course of action exists. Truth belief was operationalized as the subject's response to the question."How true is the proposition that 'The Council of Graduate Stu- dents at Michigan State University should be allowed to tax all graduate students $5.00 per year to cover the operating expenses of the organiza- tion.'?" The responses to the question were measured on a 10-point scale ranging from 0% true to 100% true utilizing 10 percentage point intervals.3 (4) We continue to conceive of attitude as the perceived relationship between an individual and some attitudinal object or situation predispos- ipg:the individual to respond in some preferential manner. In order to tap the behavioral orientation of our conception of atti- tude, at the highest possible level of scaling, our operationalization attempted to explicitly associate the proposition to an overt behavior. The proposition utilized within this study was one dealing with the taxa- tion of graduate students at Michigan State University. The subject's initial attitude toward the proposition was operationalized as his judge- ment regarding the appropriate dollar amount of taxation on a scale range from $0 to more than $5 dollars annually. The subject's attitude subsequent to the evidence manipulation was similarly measured and the subject's atti— tude change score was taken to be the difference between these pre and post-treatment measures. (5) Source credibility was defined as a judgement regarding the in- telligence, character and goodwill of a speaker or writer (Cooper, 1932). Three dimensions of this judgement may be delineated as: l7 (1) judgements regarding the extent to which a communicator is per- ceived to be a source of valid assertions (intelligence or expertness), (2) judgements regarding the degree of confidence in the communica- tor's intent to make assertions he considers most valid (character), and (3) judgements regarding the speaker's moral intent toward the audi- ence (goodwill). The procedure for constructing the operational indices of these di- mensions comprised two distinct steps. First, nine questions were con- structed utilizing Likert scales that were hypothesized to measure the three dimensions of credibility. These nine scales were then administered i to the subjects prior to the experimental treatment. Each subject was asked to rate one of two sources, i.e., Dr. Clifton Wharton, President of Michigan State University (person-specific source), and Dr. James Small, Law Professor at Wayne State University (role-specific source), speaking on the topic of "taxation of graduate students". Finally, the data obtained from the credibility scales was divided according to the source evaluated. Each of the two data subsets was then subjected to a factor analysis to determine the degree to which each scale measured the underlying three dimensions of credibility. Based on these factor analyses, factor scores were estimated as indices of expertness, character, and goodwill.4 Design The study was in the form of a multiple regression design utilizing dummy variables to indicate message treatments. subjects for Phase I of the study were taken from 15 undergraduate and graduate classes in the Department of Communication at Michigan State 18 University. During the period between April 16 and April 20, 1973, 302 undergraduate and graduate students completed the initial questionnaire related to Phase I. During the period between April 30, and May 4, 1973, the students in the same 15 communication classes were asked to participate in Phase II of the study. The subjects were randomly assigned to treatments (i.e., role source - evidence, role source - no evidence, person source - evi- dence, person source - no evidence) by randomly distributing a treatment- instrument package to all students present. A total of 253 students com- pleted the Phase II questionnaires. ‘E’T L= "5"“ ”M 1'. Chapter III RESULTS If the relationships within our model are to be cast in terms of mathematical formulations, then we must assume that the relationships ,7 between variables are sufficient and deterministic. Since some of the relationships within the model are assumed to be contingent and stochastic, it is necessary to transform the variables in such a way as to facilitate ‘IIEF”_F* . the assumption of a sufficient, deterministic relationship. In order to test the relationship between truth belief and attitude change, a truth belief index was constructed. By utilizing the simple product of desirable belief and truth belief measures, an index was created so as to transform the observed values of truth belief into values contingent upon the ob- served values of desirable belief. Utilizing the same technique, indices of evidence and source credibility were created to test the relationships between evidence and truth belief, and source credibility and truth belief.5 Given these indices the previously delineated model would hypothesize the following predictive equations:6 x3 = B31exle + BBicxlc + Balgxlg + B32x2 + 83uxu X4 ‘ 34131.2 + Baiéxic + Buéxig + B4ix'2 + B43x3 + B‘mxu X5 " 353x3 + 551119; + 351% where: x1 e = credibility (expertness) 19 20 X1 - credibility (character) X1 - credibility (goodwill) x'le - credibility index (expertness) x'lc - credibility index (character) X' - credibility index (goodwill) X2 - evidence X'2 - evidence index X3 - desirable belief X4 = truth belief X'4 = truth belief index X5 8 attitude change Xu - control variable (i.e., intent of subject regard- attending graduate school) In terms of equation (1) Table 1 indicates that the beta weights for both source credibility (character dimension) and evidence were signifi- cantly different from zero (t = 2.44 and t = 2.85 respectively, d.f. = 159). In addition, the multiple correlation coefficient of .30, was also statistically significant (F = 3.04, d.f. - 5,155).7 At this point it should be noted that certain problems exist in regard to the interpretation of the findings associated with equations (2) and (3). Due to the methods used for constructing the indices necessary to test the contingent relationships posited by the model, a high degree of multicollin— earity would be expected to exist in both equations. In this situation, though the beta weights remain unbiased estimators, they should be expected to be highly unreliable.8 For this reason it may be argued that the most appropriate method for disentangling the separate influences of the explan- atory variables is one which focuses on the increment in variance accounted 21 Table l: Standardized partial regression coefficients, multiple correla- tion coefficients, and coefficients of determination of desir— able belief regressed on the predictor variables (N=l6l). Predictor Regression Regression Regression Variables 1 2 Xu (Grad School Intentions) 8= .01 B=-.02 8= .00 X1 (Source Credibility) x1e (Expertness) =-.02 8= .01 ch (Character) 8= .18* 8= .19** X18 (Goodwill) 8= .10 8= .10 X2 (Evidence) 8= .22*** Multiple R .01 .20 .30 Multiple R2 .00 .04 .09** *denotes p<.05 **denotes p<.01 ***denotes p<.005 T“— 0 , L“ _ {3.1.0: ‘Ilurin- 22 for by the addition of an independent variable. By focusing on the in- crement in variance accounted for, we are provided a basis for judging how much an independent variable contributes additionally to variance accounted for in the dependent variable. Since the size of the increment in R2 depends on the order in which variables are included, in the case of equation (3) the model posits a hierarchy of relationships (i.e., deter- ministic relationship assumed to be stronger than the contingent relation- ship) which dictates the order of inclusion (i.e., desirable belief prior to truth belief); in the case of equation (2) the model dictates that de- sirable belief be included prior to evidence and source credibility, but posits no hierarchy with regard to evidence and source credibility.9 In the case of equation (2), the R2 for desirable belief alone is .16, which is statistically significant (p<.005, F = 15.02, d.f. - 2,158). The addition of evidence increases the R2 to .19, an increment of .03, which is statistically significant (p<.05, F = 5.88, d.f. - 1,157). The addition of source credibility to the set of predictors including desirable belief and evidence increments R2 by .01. The increment added by source credi- bility is not statistically significant (0 - .05, F = 2.0, d.f. = 1,154). It should be noted that the Rz's at each step are statistically significant (p<.005). Table 3 indicates that in terms of equation (3), desirable belief alone accounts for 17 percent of the variance in attitude change (p<.005, F - 16.01, d.f. - 2,158). When truth belief is added as a predictor of attitude change the variance accounted for increased to 27 percent. The increment in R2 of .10 is statistically significant (p<.005, f - 21.74, d.f. - 1,157). w inn—J‘suv [ED 23 Table 2: Standardized partial regression coefficients, multiple correla- tion coefficients, coefficients of determination, and increments in the coefficient of determination of truth belief regressed on the predictor variables (N=l6l). Predictor Regression Regression Regression Regression Variables l 2 3 4 Xu (Grad School Intentions) 8-.01 8=.03 B=.05 B=.05 X3 (Desirable Belief) B=.40*** B=.37*** B=.33*** I 5' Xi (Evidence Index) B=.17 B=.14 xi (Source Credibility) é ‘ Xie (Expert Index) B=.l7* Xic (Character Index) ' B=.00 Xig (Goodwill Index) B=.09 Multiple R .01 .40 .44 .45 Multiple R2 .00 .16*** .19*** .2o*** Increment in R2 .16*** Increment in R2 .03* Increment in R2 .01 *denotes p<.05 **denotes p<.01 ***denotes p<.005 24 Table 3: Standardized partial regression coefficients, multiple correla— tion coefficients, coefficients of determination, and increments in the coefficient of determination of attitude change regressed on the predictor variables (N-161). Predictor Regression Regression Regression Variables 1 2 3 Xu (Grad School Intentions) B=.02 8-.03 83.02 X3 (Desirable Belief) B=.40*** B=.09 XL (Truth Belief Index) g=,45*** Multiple R' A .02 .41 .52 Multiple R2 .0006 .17*** . 27*** Increment in R2 ,17*** Increment in R2 ,10*** *denotes p<.05 **denotes p<.01 ***denotes p<.005 Chapter IV DISCUSSION The previously suggested model hypothesized that the credibility of a source of evidence and evidence usage exerts a causal influence on desirable . belief. The data from this study tend to support such a conclusion. The ' beta weights for source credibility (character dimension) and evidence were both statistically significant (p<.01 and p<.005 respectively). In addi- tion the variance accounted for in desirable belief by both variables was statistically significant (p<.01). It should be noted, however, that the total variance accounted for in desirable belief by credibility and evi- dence usage, though statistically significant, was only 9 percent. There are at least two plausible interpretations of this. First due to the dichotomous nature of the evidence manipulation we might expect somewhat deflated estimates of the strength of the evidence-desirable belief rela- tionship. We would expect an increase in the levels of the evidence mani- pulation to bring about a subsequent increase in the variance accounted for in desirable belief. This, of course, suggests an important area of emphasis in future research on this model. Secondly, however, it would be unrealistic to expect increased precision of measurement/manipulation to account for the remaining unexplained variance in desirable belief. Therefore, we may conclude that the model fails to consider other relevant variables that are determinants of desirable belief. Such a conclusion should not be unexpected given the manifold sources of causal influence 25 26 suggested by previous research (McGuire, 1969). By manipulating only two components of the persuasive appeal, the effects of factors such as appeal "sidedness" and language intensity on desirable belief go unexplained. The model additionally suggested that desirable belief exerts a direct causal influence on truth belief, and that source credibility and evidence usage exert a causal influence on truth belief contingent upon desirable belief. Although the data provide strong support for the existence of the desirable belief-truth belief relationship, only weak support is provided for the remaining two relationships as specified. Whereas the increment in variance explained, provided for by the inclusion of evidence as a pre- dictor of truth belief, was statistically significant, source credibility appears to be a rather weak predictor of truth belief in relation to de- sirable belief and evidence usage. One possible explanation of this would be that evidence usage is causally related to truth belief (contingent upon desirable belief), whereas source credibility only exerts an indirect causal influence through its relation to desirable belief. Such an alternative explanation is indeed feasible within the context of this model, and could be tested using path analytic techniques in subsequent research. Finally, the hypothesized relationships between desirable belief and attitude change, and truth belief and attitude change were strongly sup- ported by the data. Even though the scaling and analysis techniques were something less than ideal, the inclusion of only desirable belief and truth belief as predictors of attitude change succeeded in accounting for 27 per- cent of the variation in attitude change. Overall, all of the hypothesized causal relationships were supported by the data, and all but one of the contingent causal relationships were supported. The model as suggested by the data may be represented as: “IIHP_W--nm-r X - credibility of the source of the evidence X2 - evidence usage X3 - desirable belief X4 = truth belief X5 - attitude change '- Causal path ......... p—Contingent causal path (contingent upon desirable belief). The implications of this study for future research are two. First, in order to explain the effect of evidence on attitude change a model of a persuasive appeal was presented. The relevant components of the appeal, for this study, were evidence usage and source credibility; however, the model suggests that it may be applicable for persuasive appeals in general (i.e., capable of dealing with appeal components other than evidence and source credibility). For example, it may be possible to reconcile problems in the area of language intensity by considering language intensity as an element of an appeal that functions to establish and maintain a truth belief regarding a particular proposition. If this were the case we would expect differential effects of language intensity depending upon the desirability of the proposition. Therefore, it is suggested that future research focus on investigating the generality of the model in terms of other components of persuasive appeals. Second, it is suggested that subsequent research efforts focus on replicating the present study, to include tests of the 28 alternative model suggested by the data utilizing path analytic techniques and operationalizations of evidence that allow manipulation of type and degree of evidence usage. FOOTNOTES 1Aristotle, Nichomachean Ethics, (1356a), pp. 25-35, The Basic Works of Aristotle, (ed.), McKeon. 2The role of the individual's value system as conceived here is conceptually linked to the role of values in an "instrumentally-value" analysis of attitudes (McGuire, 1969). 3The operations utilized to assess desirable and truth beliefs were based on instrumentation developed by McGuire (1960) to measure judge- ments of the truth and desirability of syllogistic premises and conclu- sions. McGuire, W.J. "Cognitive Consistency and Attitude Change." Journal of Abnormal and Social Psychology, 1960a, LX, pp. 345—353. McGuire, W.J. "Direct and Indirect Effects of Dissonance Producing Messages." Journal of Abnormal and Social Psychology, 1960b, LX, pp. 354-358. 4Guilford, J.P. Fundamental Statistics in Psychology and Education, (4th ed.), New York: McGraw-Hill, 1965. 5The evidence index was created by taking the product of observed evidence values and observed desirable belief values. The source credi- bility index was created by taking the product of observed source credi- bility values and the observed desirability belief values for each cred- ibility dimension separately. 6Each equation includes a control variable (Xu) which is not explic- itly considered by the model, but which was used for the specific topic in this study. 7The t-statistic was used to test the significance of the beta weights. An F—statistic was used to test the significance of the multiple correlation coefficient. McNemar, Q. Psycholpgical Statistics, (4th ed.), New York: John Wiley & Sons, 1969, pp. 159-160 and pp. 318-320 respectively. 8For a further discussion of this problem see Kmenta, J. Elements of Econometrics. New York: Macmillan, 1971, pp. 380-391. 29 ‘- Fl 30 9The F-statistic was used to test the increment in R2. For a further discussion of the rationale for this testing technique see: Cohen, J. "Multiple Regression as a General Data-Analytic System." Psychological Bulletin, 1968, LXX, pp. 426-443; and Althauser, R.P. "Mulitcollinearity and Non-Additice Regression Models." in H.M. Blalock (ed.), Causal Models in the Social Sciences. Chicago: Aldine Atherton, 1971. It should be additionally noted that this is a highly conservative test of the effect of an independent variable on a dependent variable. Whereas a traditional test of the significance of a beta weight takes into account the unique and joint effects of a variable, the test of the increment in R2 takes into account only the unique effects of an independent variable. APPENDICES Appendix A PHASE I QUESTIONNAIRE iiM‘)’. an '1 fl 2.1-! .1 Appendix A COMMUNICATION SURVEY The Department of Communication here at Michigan State University is in- terested in obtaining your opinion regarding a current campus issue. The value of the findings of this survey depends upon your careful considera- tion of each question. Please answer all questions. Place an "X" next to the appropriate answer in each of the following questions. .J 1. How much should the MSU Council of Graduate Students be allowed to tax graduate students in order to meet the operating expenses of the organization? _H,- e' -_ 4 I- $0 per year $1.00 per year $2.00 per year $3.00 per year $4.00 per year $5.00 per year more than $5.00 per year 2. How many people have you talked to in the past about this and related topics? 0 1 2 3 4 5 Persons 3. 0n the average, how often have you talked to each of these people about this and related topics? 0 l 2 3 4 5 Hours per month 31 32 While we are not interested in identifying you by name, we would like to identify certain sub-groups within the sample we are questioning. For example, we would like to compare the opinions and interaction habits of graduate students to those of undergraduate students. Therefore, please provide the following information. 1. Student Number 2. Age 3. Sex: Male Female 4. How many years have you attended school? 5. What year are you in at MSU? Freshman Sophomore Junior Senior Grad Student 6. Do you plan to attend graduate school? Yes No 7. If you plan to attend graduate school, where do you plan to attend? 8. Major or preference I Appendix B PHASE II QUESTIONNAIRES ‘IInrM--_vma--r I F! Appendix B OPINION PROFILE Instructions: The purpose of this profile is to obtain your opinions regarding the issues surrounding a proposition. In the following pages you will find a series of questions, followed by a message, followed by another series of questions. Each question will ask for either a numerical response or the placement of an "X" on a scale. The correct form of response will be in- dicated for each question. The value of the findings of this survey de- pends upon your CAREFUL CONSIDERATION OF EACH QUESTION. Be sure to answer all questions. Please complete all of one page before proceeding to the next page. Once you have completed a page please do not return to it. You may, however, return to this page at any time in order to clarify any instruction that is unclear to you. 33 wavy-“wes- 34 Place an "X" above the appropriate answer in each of the following questions. If Dr. James Small, Law Professor at Wayne State University, were speaking on "The Right of Graduate Student Councils to Tax Graduate Students," how would you answer the following questions? 40. I would expect his proposals to be sensible. Strongly Disagree Neutral Agree Strongly Disagree Agree 41. I would expect him to be a reliable source of information on the topic. Strongly Agree Neutral Disagree Strongly Agree Disagree 42. I would consider him to be an expert on the topic. Strongly Disagree Neutral Agree Strongly Disagree Agree 43. I would expect him to provide a clear analysis of the problems involved in the subject. Strongly Agree Neutral Disagree Strongly Agree Disagree 44. I trust that he would tell the truth about the tapic. Strongly Disagree Neutral Agree Strongly Disagree Agree 45. I admire his background. Strongly Agree Neutral Disagree Strongly Agree Disagree 46. I believe that he is concerned with my well—being. Strongly Disagree Neutral Agree Strongly Disagree Agree 47. 48. 35 I believe that he is sincere. Strongly Agree Neutral Disagree Agree I believe that he would follow his own advice. Strongly Disagree Neutral Agree Disagree Strongly Disagree Strongly Agree 36 The Council of Graduate Students at Michigan State University should be allowed to tax all graduate students $5.00gper4year to cover the operating 'expenses of the ogganization. Dr. James Small, Law Professor at Wayne State university spoke about the taxing of graduate students in an address to the National Association of Graduate Student Councils at its annual meeting in Cleveland, Ohio. Dr. Small presented a three-point analysis of the problem. "In the g first place," Dr. Small said, "there are a great many desirable functions L that can be provided by graduate student councils. A council of graduate students is a body whose power is derived solely from the graduate students ‘EEETI‘ of a university and is therefore independent of the graduate school. Ac- tions of a graduate council cannot be reversed except by referendum, as provided for under the constitution, or revocation by the council body, except in cases of personal transactions and matters concerning referenda that are illegal under university regulations or the council constitution or bylaws. These matters may be adjudicated by a judiciary under the auspices of the graduate school provided the judiciary has adequate stu- dent representation. The adequacy of the representation will be established if the council agrees to send members to the judiciary at its initiation." "However," Dr. Small went on, "the currenteconomic situation in the united States has placed an ever increasing financial burden on most grad- uate student organizations. A council of graduate students is charged with certain general objectives, economic and otherwise. It is the duty of a council to promote the academic, social, and economic aims of grad- uate students. An effective graduate student council must also establish continuing communication among graduate students and create channels of 37 communication with other student organizations and with academic and administrative units of the university. It should be noted, however, that a council of graduate students is not a political organization and shall not endorse political issues which do not directly concern the wel- fare of graduate students at the university." Dr. Small stated that by taxing graduate students the funds needed for maintaining a graduate student council could be provided. "In this instance, a tax may be viewed as a fee or due levied on a university grad- Equ uate student population to cover operating expenses of the graduate stu- . dent council. Such a tax may be assessed during normal course registra- ; r tion at the beginning of each academic year, or during course registration at the beginning of each academic term." In an interview subsequent to his speech, Dr. Small commented on the MSU Council of Graduate Students specifically. "At Michigan State Uni- versity the problem is particularly important. Michigan State defines the graduate student population to be all registered Michigan State stu— dents who are classified, according to university records, as being in levels six, seven and nine. This definition allows for the classification of approximately one—fifth of the students at Michigan State as graduate students. Therefore, the Council of Graduate Students represents a sub- stantial portion of the total MSU student population. In addition, the governing board of the council, through elected representatives from all the departments offering advanced degree programs, sends official delegates to all major university administrative committees, including the Academic Council, the Student Senate, and the Presidential Council." For the reasons mentioned, Dr. Small commented, "I would whole- heartedly support the proposed $5 graduate student tax at MSU." 38 PROPOSITION: The Council of Graduate Students 9; Michigan State Univer- sity should be allowed to tax a1l_graduate students $5.00 pergyear to cover the operating;expenses of the organization. 49. Indicate your degree of agreement or disagreement with the preceding proposition. (Place an "X" above the appropriate space) Strongly Agree Neutral Disagree Strongly Agree Disagree 50. How important is this issue to you personally? (Place an "X" above the appropriate space) Very Important Neutral Unimpor- Very Un- Important tant important 51. Suppose an election was to be held on campus in September to vote on the preceding proposition. If it appeared from the pools that the vote would be close, how much TIME PER WEEK.wou1d you be WILL- ING TO WORK for a responsible organization whose POSITION WAS CON- SISTENT WITH YOUR OWN? HOURS PER WEEK 53. In your opinion, how much should the MSU Council of Graduate Students be allowed to tax graduate students in order to meet the operating expenses of the ogranization? $0 $1.00 $2.00 $3.00 $4.00 $5.00 More than $5.00 dollars per year 54. How true is the preceding proposition? 0 - 10% true 11 - 20% true 21 - 30% true 31 - 40% true 41 - 50% true 51 - 60% true 61 - 70% true 71 - 80% true 81 — 90% true 91 -100% true 56. 57. 39 How desirable is the preceding proposition? Very Desirable Neutral Undesirable Very Un- Desirable desirable While we §£g_ngg_interested in identifying you by name, we would like to compare your responses on this questionnaire with your responses on a previous questionnaire. Therefore, PLEASE indicate your student number. Student Number T H A N K Y 0 U Iii-:11- ‘ W“ Lu 0 OPINION PROFILE Instructions: The purpose of this profile is to obtain your opinions regarding the issues surrounding a proposition. In the following pages you will find a series of questions, followed by a message, followed by another series of questions. Each question will ask for either a numerical response or the placement of an "X" on a scale. The correct form of response will be in- dicated for each question. The value of the findings of this survey de- pends upon your CAREFUL CONSIDERATION OF EACH QUESTION. Be sure to answer all questions. Please complete all of one page before proceeding to the next page. Once you have completed a page please do not return to it. You may, however, return to this page at any time in order to clarify any instruction that is unclear to you. “5;“!!177 .-: ‘E— ‘I ‘ 40 41 Place an "X" above the appropriate answer in each of the following questions. If Dr. James Small,_Law Professor at Wayne State University, were speaking on "The Right of Graduate Student Councils to Tax Graduate Students," how would you answer the following questions? 40. I would expect his proposals to be sensible. Strongly Disagree Neutral Agree Strongly Disagree Agree 41. I would expect him to be a reliable source of information on the topic. Strongly Agree Neutral Disagree Strongly Agree . Disagree 42. I would consider him to be an expert on the tOpic. Strongly Disagree Neutral Agree Strongly Disagree Agree 43. I would expect him to provide a clear analysis of the problems in- volved in the subject. Strongly Agree Neutral Disagree Strongly Agree Disagree 44. I trust that he would tell the truth about the topic. Strongly Disagree Neutral Agree Strongly Disagree Agree 45. I admire his background. Strongly Agree Neutral Disagree Strongly Agree Disagree 46. I believe that he is concerned with my well-being. Strongly Disagree Neutral Agree Strongly Disagree Agree 47. 48. 42 I believe that he is sincere. Strongly Agree Neutral Disagree Agree I believe that he would follow his own advice. Strongly Disagree Neutral Agree Disagree Strongly Disagree Strongly Agree 43 The Council of Graduate Studentsgat Michigan State University should be allowed to tax all:g£aduate students $5.00 per year to cover the operating expenses of the organization. Dr. James Small, Law Professor at Wayne State University spoke about the taxing of graduate students in an address to the National Association of Graduate Student Councils at its annual meeting in Cleveland, Ohio. Dr. Small presented a three-point analysis of the problem. In the first place, Dr. Small said, "there are a great many desirable functions that can be provided by graduate student councils. For example, several w: 3mm“ n ... _- graduate councils throughout the country have available for their members free legal services. This includes not only legal advice, but also the services of a lawyer at no cost if the student should find it necessary to go into court. Other councils provide free psychiatric counseling, as well as more general counseling for any personal, family or social problems. Still others make available daycare and nursery school facilities for grad- uate student families that are unable to financially support such services on an individual basis. "However," Dr. Small went on, "the current economic situation in the United States has placed an ever increasing financial burden on most grad- uate student organizations. Secretarial help, office supplies, lawyer's - and social workers' fees and many other costs have increased as much as 48% in the past three years. Even for such a simple task as issuing a graduate student newsletter to help the graduate students keep abreast of current available graduate student services, expenses have increased by nearly a third. To put the matter simply, if graduate student councils 44 are to provide the vital services that are needed, they are going to need more money immediately." It was Dr. Small's conclusion that by taxing graduate students such funds could be provided. "Given the current financial situation, univer- sities are unwilling and often unable to bail out the graduate student councils. Universities are very cautious in how they are allocating funds. Even though most universities are aware of the vital services provided by the councils, the councils have a very low priority with them in regard to financial outlays. The councils don't have many other alternatives. They can begin to charge students for services rendered or they can institute a tax. If they charge students at a rate sufficient to defray their ex- penses, then the cost to the students will become prohibitive. It would become just as expensive for a graduate student to get say, legal assist- ance from the graduate student council as from an independent lawyer. Clearly, if graduate student councils are to continue to function effec- tively, they must increase their funds through the only realistic means - taxation." At an interview subsequent to his speech, Dr. Small commented on the MSU Council of Graduate Students specifically. "In a study recently com- pleted for the Council of Graduate Students at MSU, it was found that by taxing each graduate student $5 per year, it would be possible for the council to offer (a) two full-time lawyers for student legal assitance, (b) a part-time general counseling facility for psychiatric and family counseling, and (c) a 5-day per week day-care center capable of handling 100 to 150 graduate student children between the ages of 1 and 6. Given the magnitude of these services in relation to the cost per graduate stu- dent, I would wholeheartedly support the proposed $5 graduate student tax at MSU." q ll," 45 PROPOSITION: The Council of Ggaaaate Studentsgat Michigaa_State Univer- sigy should be allowed to tax all graduate students $5.00 per_year to cover the operatigggexpenses of the organization. 49. Indicate your degree of agreement or disagreement with the preceding proposition. (Place an "X" above the appropriate space) 'EEFSE§I§ "XEFEE—— 'EEEEPEI 'BIEZEPEZ '§E¥35§I§ Agree Disagree 50. How important is this issue to you personally? (Place an "X" above the appropriate space) Very Important Neutral Unimpor— Very Un- Important tant important 51. Suppose an election was to be held on campus in September to vote on the preceding proposition. If it appeared from the polls that the vote would be close, how much TIME PER,WEEK would you be WILL- ING TO WORK for a responsible organization whose POSITION WAS CON- SISTENT WITH YOUR OWN? HOURS PER WEEK 53. In your opinion, how much should the MSU Council of Graduate Students be allowed to tax graduate students in order to meet the operating expenses of the organization? $0 $1.00 $2.00 $3.00 $4.00 $5.00 More than $5.00 dollars per year 54. How true is the preceding proposition? 0 - 10% true 11 - 20% true 21 — 30% true 31 - 40% true 41 - 50% true 51 - 60% true 61 - 70% true 71 - 80% true 81 - 90% true 91 -100% true 56. 57. 46 How desirable is the preceding proposition? Very Desirable Neutral Undesirable Very Un- Desirable desirable While we are not interested in identifying you by name, we would like to compare your responses on this questionnaire with your responses on a previous questionnaire. Therefore, PLEASE indicate your student number. Student Number T H A N K Y O U OPINION PROFILE Instructions: The purpose of this profile is to obtain your opinions regarding the issues surrounding a proposition. In the following pages you will find a series of questions, followed by a message, followed by another series of questions. Each question will ask for either a numerical response or the placement of an "X" on a scale. The correct form of response will be in- dicated for each question. The value of the findings of this survey de- pends upon your CAREFUL CONSIDERATION OF EACH QUESTION. Be sure to answer all questions. Please complete all of one page before proceeding to the next page. Once you have completed a page please do not return to it. You may, however, return to this page at any time in order to clarify any instruction that is unclear to you. 47 48 Place an "X" above the appropriate answer in each of the following questions. If Dr. Clifton Wharton, President of Michigan State Universigy, were speaking on "The Right of Graduate Student Councils to Tax Graduate Students," how would you answer the following questions? 31. 32. 33. 34. 35. 36. 37. I would expect his proposal to be sensible. Strongly Disagree Neutral Agree Strongly Disagree Agree I would expect him to be a reliable source of information on the topic. Strongly Agree Neutral Disagree Strongly Agree Disagree I would consider him to be an expert on the topic. Strongly Disagree Neutral Agree Strongly Disagree Agree 1 would expect him to provide a clear analysis of the problems involved in the subject. Strongly Agree Neutral Disagree Strongly Agree Disagree I trust that he would tell the truth about the tapic. Strongly Disagree Neutral Agree Strongly Disagree Agree I admire his background. Strongly Agree Neutral Disagree Strongly Agree Disagree I believe that he is concerned with my well-being. Strongly Disagree Neutral Agree Strongly Disagree Agree *- .— ! l "h I?’ ‘3. I: .A' 38. 39. 49 I believe that he is sincere. Strongly Agree Neutral Disagree Agree ' I believe that he would follow his own advice. Strongly Disagree Neutral Agree Disagree Strongly Disagree Strongly Agree 50 The Council of Graduate Studentsfiat Michigan State Universipy should be allowed to tax all_graduate students $5.00 per_year to cover the operating expenses of the organization. Clifton Wharton, President of Michigan State University, spoke about the taxing of graduate students in an address to the National Association of Graduate Student Councils at its annual meeting in Cleveland, Ohio. Dr. Wharton presented a three-point analysis of the problem. "In the first place," Dr. Wharton said, "there are a great many desirable functions that can be provided by graduate student councils. A council of graduate students is a body whose power is derived solely from the graduate stu- dents of a university and is therefore independent of the graduate school. Actions of a graduate council cannot be reversed except by referendum, as provided for under the constitution, or revocation by the council body, except in cases of personal transactions and matters concerning referenda that are illegal under university regulations or the council constitution or bylaws. These matters may be adjudicated by a judiciary under the auspices of the graduate school provided the judiciary has adequate stu- dent representation. The adequacy of the representation will be esta- blished if the council agrees to send members to the judiciary at its initiation." "However," Dr. Wharton went on, "the current economic situation in the United States has placed an ever increasing financial burden on most graduate student organizations. A council of graduate students is charged with certain general objectives, economic and otherwise. It is the duty of a council to promote the academic, social, and economic aims of grad- uate students. An effective graduate student council must also establish “it“... (A! s 1 . V‘ 51 continuing communication among graduate students and create channels of communication with other student organizations and with academic and administrative units of the university. It should be noted, however, that a council of graduate students is not a political organization and shall not endorse political issues which do not directly concern the welfare of graduate students at the university." Dr. Wharton stated that by taxing graduate students the funds needed for maintaining a graduate student council could be provided. "In this instance, a tax may be viewed as a fee or due levied on a university grad— uate student population to cover operating expenses of the graduate stu- dent council. Such a tax may be assessed during normal course registra- tion at the beginning of each academic year, or during course registra- tion at the beginning of each academic term." In an interview subsequent to his speech, Dr. Wharton commented on the MSU Council of Graduate Students specifically. "At Michigan State Univer- sity the problem is particularly important. Michigan State defines the graduate student populationto be all registered Michigan State students who are classified, according to university records, as being levels, six, seven and nine. This definition allows for the classification of approximately one-fifth of the students at Michigan State as graduate students. Therefore, the Conncil of Graduate Students represent a sub- stantial portion of the total MSU student population. In addition, the governing board of the council, through elected representatives from all the departments offering advanced degree programs, sends official delegates to all major university administrative committees, including the Academic Council, the Student Senate, and the Presidential Council." For the reasons mentioned, Dr. Wharton commented, "I would whole— heartedly support the proposed $5 graduate student tax at MSU." ,4 “wr— PROPOSITION: 52 The Council of Graduate Students at Michigan State Univer- sity should be allowed to tax all graduate students $5.00 per year to cover the operating expenses of the organization. 49. 50. 51. 53. 54. Indicate your degree of agreement or disagreement with the preceding (Place an "X" above the appropriate space) proposition. Agree How important is this issue to you personally? the appropriate space) Very Strongly Important Agree Important Neutral Disa atant gree Neutral Unimpor- Strongly Disagree (Place an "X" above Very Un- important Suppose an election was to be held on campus in September to vote on the preceding proposition. If it appeared from the polls that the vote would be close, how much TIME PER WEEK would you be WILLING TO WORK for a responsible organization whose POSITION WAS CONSISTENT WITH YOUR OWN? HOURS PER WEEK In your opinion, how much should the MSU Council of Graduate Students be allowed to tax graduate students in order to meet the operating expenses of the organization? $0 $1.00 $2.00 $3.00 $4.00 dollars per year How true is the preceding proposition? 0 11 21 31 41 51 61 71 81 91 - 10% - 20% — 30% - 40% - 50% - 60% - 70% - 80% - 90% -100% true true true true true true true true true true $5.00 More than $5.00 56. 57. 53 How desirable is the preceding proposition? Very Desirable Neutral Undesirable Very Un— Desirable desirable While we are not interested in identifying you by name, we would like to compare your responses on this questionnaire with your responses on a previous questionnaire. Therefore, PLEASE indicate your student number. Student Number T H A N K Y O U OPINION PROFILE Instructions: The purpose of this profile is to obtain your opinions regarding the issues surrounding a proposition. In the following pages you will find a series of questions, followed by a message, followed by another series of questions. Each question will ask for either a numerical response or the placement of an "X" on a scale. The correct form of response will be in- dicated for each question. The value of the findings of this survey de- pends upon your CAREFUL CONSIDERATION OF EACH QUESTION. Be sure to answer all questions. Please complete all of one page before proceeding to the next page. Once you have completed a page please do not return to it. You may, however, return to this page at any time in order to clarify any instruction that is unclear to you. 54 55 Place an "X" above the appropriate answer in each of the following questions. If Dr. Clifton Wharton,_President of Michigan State University, were speaking on "The Right of Graduate Student Councils to Tax Graduate Stu- dents," how would you answer the following questions? 31. I would expect his proposal to be sensible. Strongly Disagree Neutral Agree Strongly Disagree Agree 32. I would expect him to be a reliable source of information on the . topic. } .r' Strongly Agree Neutral Disagree Strongly 2 Agree Disagree Ls 33. I would consider him to be an expert on the topic. Strongly Disagree Neutral Agree Strongly Disagree Agree 34. I would expect him to provide a clear analysis of the problems in- volved in the subject. Strongly Agree Neutral Disagree Strongly Agree Disagree 35. I trust that he would tell the truth about the topic. Strongly Disagree Neutral Agree Strongly Disagree Agree 36. I admire his background. Strongly Disagree Neutral Agree Strongly Disagree - Agree 37. I believe that he is concerned with my well-being. Strongly Disagree Neutral Agree Strongly Disagree Agree 38. 39. 56 I believe that he is sincere. Strongly Agree Neutral Disagree Agree I believe that he would follow his own advice. Strongly Disagree Neutral Agree Disagree Strongly Disagree Strongly Agree 57 The Council of Graduate Students at Michigan State University should be allowed to tax all graduate students $5.00_peryyear to cover the operating egpenses of the organization. Clifton Wharton, President of Michigan State University spoke about the taxing of graduate students in an address to the National Association of Graduate Student Councils at its annual meeting in Cleveland, Ohio. Dr. Wharton presented a three-point analysis of the problem. In the first place, Dr. Wharton said, "there are a great many desirable functions that can be provided by graduate student councils. For example, several graduate councils throughout the country have available for their members free legal services. This includes not only legal advice, but also the services of a lawyer at no cost if the student should find it necessary to go into court. Other councils provide free psychiatric counseling, as well as more general counseling for any personal, family or social problems. Still others make available daycare and nursery school faci- lities for graduate student families that are unable to financially sup- port such services on an individual basis. "However," Dr. Wharton went on, "the current economic situation in the United States has placed an ever increasing financial burden on most graduate student organizations. Secretarial help, office supplies, law- yer's and social workers' fees and many other costs have increased as much as 48% in the past three years. Even for such a simple tasks as issuing a graduate student newsletter to help the graduate students keep abreast of current available graduate student services, expenses have increased by nearly a third. To put the matter simply, if graduate stu- dent councils are to provide the vital services that are needed, they are going to need more money immediately. 1. l 58 It was Dr. Wharton's conclusion that by taxing graduate students such funds could be provided. "Given the current financial situation, univer- sities are unwilling and often unable to bail out the graduate student councils. Universities are very cautious in how they are allocating funds. Even though most universities are aware of the vital services provided by the councils, the councils have a very low priority with them in regard to financial outlays. The councils don't have many other alternatives. They can begin to charge students for services rendered or they can institute a tax. If they charge students at a rate sufficient to defray their ex- penses, then the cost to the students will become prohibitive. It would become just as expensive for a graduate student to get say, legal assist- ance from the graduate student council as from an independent lawyer. Clearly if graduate student councils are to continue to function effec- tively, they must increase their funds through the only realistic means - taxation." At an interview subsequent to his speech, Dr. Wharton commented on the MSU Council of Graduate Students specifically. "In a study recently completed for the Council of Graduate Students at MSU, it was found that by taxing each graduate student $5 per year, it would be possible for the council to offer (a) two full-time lawyers for student legal assistance, (b) a part-time general counseling facility for psychiatric and family counseling, and (c) a 5-day per week day-care center capable of handling 100 to 150 graduate student children between the ages of l and 6. Given the magnitude of these services in relation to the cost per graduate student, I would wholeheartedly support the proposed $5 graduate student tax at MSU." 59 PROPOSITION: The Council of Graduate Students at Michigan State Univer- sity should be allowed to tax all graduate students $5.00 per year to cover the operatingyexpenses of the organization. 49. Indicate your degree of agreement or disagreenent with the preceding proposition. (Place an "X" above the appropriate space) Strongly Agree Neutral Disagree Strongly Agree Disagree 50. How important is this issue to you personally? (Place an "X" above the appropriate space) Very Important Neutral Unimpor- Very Un- Important tant important 51. Suppose an election was to be held on campus in September to vote on the preceding proposition. If it appeared from the polls that the vote would be close, how much TIME PER.WEEK would you be WILLING TO WORK for a responsible organization whose POSITION WAS CONSISTENT WITH YOUR OWN? HOURS PER.WEEK 53. In your Opinion, how much should the MSU Council of Graduate Students be allowed to tax graduate students in order to meet the operating expenses of the organization? $0 $1.00 $2.00 $3.00 $4.00 $5.00 More than $5.00 dollars per year 54. How true is the preceding proposition? 0 — 10% true 11 - 20% true 21 - 30% true 31 - 40% true 41 - 50% true 51 — 60% true 61 - 70% true 71 — 80% true 81 - 90% true 91 —100% true 56. 57. 60 How desirable is the preceding proposition? Very Desirable Neutral Undesirable Very Un- Desirable desirable While we are not interested in identifying you by name, we would like to compare your responses on this questionnaire with your responses on a previous questionnaire. Therefore, PLEASE indicate your student number. Student Number T H A N K Y O U ‘I-fllP-f-—~ OPINION PROFILE Instructions: The purpose of this profile is to obtain your opinions regarding the issues surrounding a proposition. In the following pages youuwill find a series of questions, followed by a message, followed by another series of questions. Each question will ask for either a numerical response or the placement of an "X" on a scale. The correct form of response will be in- dicated for each question. The value of the findings of this survey de- pends upon your CAREFUL CONSIDERATION OF EACH QUESTION. Be sure to answer all questions. Please complete all of one page before proceeding to the next page. Once you have completed a page please do not return to it. You may, however, return to this page at any time in order to clarify any instruction that is unclear to you. 61 62 Place an "X" above the appropriate answer in each of the following questions. If Dr. Clifton Wharton, President of Michigan State University, were speaking on "The Right of Graduate Student Councils to Tax Graduate Students," how would you answer the following questions? 31. 32. 33. 34. 35. 36. 37. Strongly Strongly Agree Neutral Strongly Strongly Agree Neutral Strongly Strongly Agree Neutral Strongly I would expect his proposal to be sensible. Disagree Neutral Agree Disagree I would expect him to be a reliable source of information on the topic. Disagree Agree I would consider him to be an expert on the tapic. Disagree Neutral Agree Disagree I would expect him to provide a clear analysis of the involved in the subject. Disagree Agree I trust that he would tell the truth about the tOpic. Disagree Neutral Agree Disagree I admire his background. Disagree Agree I believe that he is concerned with my well—being. Disagree Neutral Agree Disagree Strongly Agree Strongly Disagree Strongly Agree problems Strongly Disagree Strongly Agree Strongly Disagree Strongly Agree ‘ullarrr-rmrwnI-Ir I L n. 38. 39. 63 I believe that he is sincere. Strongly Agree Neutral Disagree Agree I believe that he would follow his own advice. Strongly Disagree Neutral Agree Disagree Strongly Disagree Strongly Agree 64 Place an "X" above the appropriate answer in each of the following questions. If Dr. James Small, Law Professor at Wayne State University, were speaking on "The Right of Graduate Student Councils to Tax Graduate Students," how would you answer the following questions? 40. 41. 42. 43. 44. 45. 46. I would expect his proposals to be sensible. Strongly Disagree Neutral Agree Strongly Disagree Agree I would expect him to be a reliable source of information on the topic. Strongly Agree Neutral Disagree Strongly Agree , Disagree I would consider him to be an expert on the tOpic. Strongly Disagree Neutral Agree Strongly Disagree Agree I would expect him to provide a clear analysis of the problems involved in the subject. Strongly Agree Neutral Disagree Strongly Agree Disagree I trust that he would tell the truth about the topic. S tron gly Dis agree N eut ral Agree S t ron gly Disagree Agree I admire his background. Strongly Agree Neutral Disagree Strongly Agree Disagree I believe that he is concerned with my well-being. Strongly Disagree Neutral Agree Strongly Disagree Agree 47. 48. 65 I believe that he is sincere. Strongly Agree Neutral Disagree Agree I believe that he would follow his own advice. Strongly Disagree Neutral Agree Disagree Strongly Disagree Strongly Agree ’— EFF. 66 PROPOSITION: The Council of Graduate Students at Michigan State Univer— sity should be allowed to tax all graduate students $5.004per year to cover the operatinglexpenses of the organization. 49. Indicate your degree of agreement or disagreement with the preceding proposition. (Place an "X" above the appropriate space) WWWWM Agree Disagree 50. How important is this issue to you personally? (Place an "X" above the appropriate space) p1; l-I"> ., . Very Important Neutral Unimpor— Very Un- Important tant important “"‘ “a; ‘ 51. Suppose an election was to be held on campus in September to vote on the preceding proposition. If it appeared from the polls that the vote would be close, how much TIME PER WEEK would you be WILLING TO WORK for a responsible organization.whose POSITION WAS CONSISTENT WITH YOUR OWN? HOURS PER WEEK 53. In your opinion, how much should the MSU Council of Graduate Students be allowed to tax graduate students in order to meet the operating expenses of the organization? $0 $1.00 $2.00 $3.00 $4.00 $5.00 More than $5.00 dollars per year 54. How true is the preceding proposition? 0 - 10% true 11 - 20% true 21 - 30% true 31 - 40% true 41 - 50% true 51 - 60% true 61 - 70% true 71 - 80% true 81 - 90% true 91 -100% true 56. 57. 67 How desirable is the preceding proposition? Very Desirable Neutral Undesirable Very Un- Desirable desirable While we are not interested in identifying you by name, we would like to compare your responses on this questionnaire with your responses on a previous questionnaire. Therefore, PLEASE indicate your student number. Student Number T H A N K Y O U ~d~ L‘Nn LIST OF REFERENCES REFERENCES Allport, G.W. "The Historical Background of Modern Social Psychology." in Lindzey, G. and E. Aronson (eds.), The Handbook of Social Psychology. Vol. 1, (2nd ed.), Reading, Mass.: Addison-Wesley, 1969. Althauser, R.P. "Multicollinearity and Non-Additive Regression Models." in Blalock, H.M. (ed.), Causal Models in the Social Sciences. Chicago: Aldine Atherton, 1971. Baudhuin, E.S. and M.K. Davis. "Scales for the Measurement of Ethos: Another Attempt." Speech Monpgraphs, 1972, XXXIX, pp. 296-302. Bettinghaus, E.P. "The Relative Effects of the Use of Testimony in a Persuasive Speech Upon the Attitudes of Listeners." Unpublished master's thesis, Bradley University, 1953. Bostrom, R.N. and R.K. Tucker. "Evidence, Personality and Attitude Change." Speech Monographs, 1969, XXXVI, pp. 22—27. Carlson, E.R. "Attitude Change through Modification of Attitude Structure." Journal of Abnormal and Social Paychology, 1956, LII, pp. 256-261. Cathcart, R.S. "An Experimental Study of the Effectiveness of Four Methods of Presenting Evidence." Speech Monggraphs, 1955, XXII, pp. 227-233. Cohen, J. "Multiple Regression as a General Data-Analytic System." Psychological Bulletin, 1968, LXX, pp. 426-443. Cooper, L. The Rhetoric of Aristotle. New York: Appleton-Century- Crofts, 1932. Costley, D.L. "An Experimental Study of the Effectiveness of Quantitative Evidence on Speeches of Advocacy." Unpublished master's thesis, University of Oklahoma, 1958. Cushman, D.P. and G.A. Hauser. "Weaver's Rhetorical Theory: Axiology and the Adjustment of Belief, Invention, and Judgement." gnarterly Journal of Speech, 1973, LIX, pp. 319-329. Divesta, F.J. and J.C. Merwin. "The Effects of Need-Oriented Communica— tions on Attitude Change." Journal of Abnormal and Social Psychology, 1960, LXI, pp. 80—85. 68 69 Gilkinson, H., S. Paulson and D. Sikkink. "Effects of Order and Authority in an Argumentative Speech." iguarterly Journal of Speech, 1969, LV, pp. 183-192. Graham, G.M. "The Natural Procedure of Argument." Quarterly Journal of Speech Education, 1925, XI, pp. 317-327. Grimaldi, W. "Rhetoric and the Philosophy of Aristotle." Classical Journal, 1958, LIII, p. 374. Guilford, J.P. Fundamental Statistics in Psychology and Education, (4th ed.), New York: McGraw-Hill, 1965. Infante, D.A. "Predicting Attitude frOm Desirability and Likelihood Ratings of Rhetorical Prepositions." Speech Monographs, 1971, xxxv111, pp. 321-326. Kline, J.A. "Interaction of Evidence and Readers' Intelligence on the Effects of Short Messages." Quarterly Journal of Speech, 1969, LV, pp. 407-413. Kmenta, J. Elements of Econometrics, New York: Macmillan, 1971, pp. 380-391. Knower, F. "Experimental Studies of Changes in Attitudes: I. A Study of the Effects of Oral Arguments on Changes in Attitude." Journal of Social Psychology, 1925, VI, pp. 315-347. Lefford, A. "The Influence of Emotional Subject Matter on Logical Reasoning." Journal of General Psychology, 1946, XXXIV, pp. 127-151. McCroskey, J.C. "A Summary of Experimental Research on the Effects of Evidence in Persuasive Communication." Quarterly Journal of Speech, 1969, LV, pp. 169—176. McCroskey, J.C. "The Effects of Evidence as an Inhibitor of Counter- Persuasion." Speech Monographs, 1970, XXXVII, pp. 188-194. McGuire, W.J. "Cognitive Consistency and Attitude Change." Journal of Abnormal and Social Psychology, 1960a, LX, pp. 345-353. McGuire, W.J. "Direct and Indirect Effects of Dissonance Producing Messages." Journal of Abnormal and Social Psycholqu, 1960b, LX, McGuire, W.J. "The Nature of Attitudes and Attitude Change." in Lindzey, G. and E. Aronson (eds.), The Handbook of Social Psychology, Vol. III. (2nd ed.), Reading, Mass.: Addison-Wesley, 1969. McKeon, R. The Basic Works of Aristotle, New York: Random House, Inc. 1941. 7O McNemar, Q. Psychological Statistics. (4th ed.), New York: John Wiley and Sons, 1969. Morgan, J.J.B. and J.T. Morton. "The Distortion of Syllogistic Reasoning Produced by Personal Convictions." Journal of Social Paychology, 1944, XX, pp. 39-59. Osgood, C.E., G.J. Suci and P.H. Tannenbaum. The Measurement of Meaning. Urbana: University of Illinois Press, 1957. Ostermeier, T.H. "Effects of Type and Frequency of Reference Upon Per- ceived Source Credibility and Attitude Change." Speech Monographs, 1967, XXXIV, pp. 137-144. Rokeach, M. Values, Attitudes and Beliefs. San Francisco: Jossey-Bass, 1968. Rosenberg, M.J. "Cognitive Structure and Attitudinal Affect." Journal of Abnormal and Social Psyehology, 1956, LII, pp. 367-372. Sherif, G.W. and N.R. Jackman. "Judgements of Truth by Participants in a Collective Controversy." Public Opinion Qparterly, 1966, XXX, pp. 173-186. Sikkink, D.E. "An Experimental Study of the Effects on the Listener of Anti-Climax Order and Authority in an Argumentative Speech." Southern Speech Journal, 1956, XXII, pp. 73-78. Smith, M.B. "Personal Values as Determinants of a Political Attitude." The Journal of Psychology, 1949, XXVIII, pp. 477-486. Wagner, G.A. "An Experimental Study of the Relative Effectiveness of Varying Amounts of Evidence in a Persuasive Communication." Un- published master's thesis, Mississippi Southern University, 1958. Whitehead, J.S. Jr. "Effects of Authority-Based Assertions on Attitude and Credibility." Speech Monographs, 1971, XXXVIII, pp. 311-315. Zetterberg, H.L. On Theory and Verification in Sociolqu, Totowa, N.J.: Bedminster Press, 1963. HICHIGRN STQTE UNIV. LIBRQRIES IIIIII IIIIIIII IIIIIIIIIIIEIIIIII I III 311293