THE EFFECTS OF TASK AND SOURCE? C‘REDIBILITY 0N EVIDENCE USAGE - Thesis for the Degree of M .A. MICHBGAN STATE UNIVERSITY- SANDRA 1L. HLION 1972 ’ THESIS i xbb 22; E33‘ MSW _REruag;hG MLTERIALS: ~ Place in book‘droo~fo 4K aur- =-\ gay 12.9 can: LIBRARIES remove this checkout from 4—5.“. your record. fjjflzfi wiH ~—— be charged if book is returned after the date _ stamped be1ow. W - :v.17—<3¢7""“ ABSTRACT THE EFFECTS OF TASK AND SOURCE CREDIBILITY 0N EVIDENCE USAGE By Sandra L. Filion This study sought to assess the effects of high and low source credibility and of certain tasks on a descriptive set of evidence categories. Undergraduate students at Michigan State University completed a pretest questionnaire to: l) assess their initial attitudes on two issues considered to be salient and ego-involvingo-drugs and busing, 2) gain source credibility ratings for nine individuals. Two high credible and two low credible sources were chosen for the experiment. Experimental subjects read six messages of three types, truth, desirability and definition (three pro and three con) under one of the two propositions. One of four source conditions was included in each questionnaire: 1) no source, 2) role source, 3) high credible known source, and 4) low credible known source. In each source con- dition subjects completed one of three tasks: 1) Informative M, to rank order the six messages according to degree of helpfulness in understanding the proposition, 2) Belief‘IggE, to rank order the six messages according to degree of helpfulness in taking a stand on the proposition, and 3) Categorisation $33!, to categorize the six messages into three categories, truth, desirability, and definition. In the Sandra L. Filion no source condition subjects were asked to choose six sources from the list of nine and match them to the message each source most probably stated. Seven hypotheses were tested in this investigation: 1) Given the Informative $32!, a greater than chance frequency of subjects will rank order the messages in the following descending order of importance: a) truth, b) desirability, and c) definition. 2) Given the Belief 2235, a greater than chance frequency of subjects will rank order the messages in the following descending order of importance: a) desirability, b) truth, and c) definition. 3) Given the Categorisation 2335, a greater than chance frequency of subjects will rank order the messages in the following manner: a) the two truth messages in the truth cate- gory, b) the two desirability messages in the desirability category, and c) the two definition messages in the definition category. 4) In the next three hypotheses it was predicted that a greater proportion of the subjects in the no source condition than in any other source con- dition would perform each respective task as expected. 5) Finally, given the matching task, subjects will have a greater tendency to match high credible sources than low credible sources to messages supporting their initial beliefs. In order to test hypotheses I, II, and III, an expected frequency distribution of possible message orderings was generated. A criterion ranking was chosen. 0f the 78 possible combinations, 13 would include one or less errors (a correct ranking). One would expect the propor- tion of correct rankings to equal 13/78's within a population of sub- jects by chance. The proportion of correct rankings was compared against Sandra L. Filion the expected population proportion. For all three tasks, the results confirmed the hypotheses. For hypotheses IV, V, and VI, a 2' test for proportions was computed between the proportion of subjects correctly ranking the messages in each of the three conditions involving sources. The results failed to confirm the hypotheses. Attribution of source does not seem to effect the frequency of subjects ranking messages in the predicted orders. For hypothesis VII the percentage of people matching each of the sources with each message was calculated. The results failed to confirm the hypothesis. Further investigation is needed exploring the direct link between high, neutral, and low credible sources and high, neutral, and low evaluated propositions. Then, we might examine the amount of variance carried by the evidence and by the source. THE EFFECTS OF TASK AND SOURCE CREDIBILITY ON EVIDENCE USAGE By Sandra L. Filion A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Communication 1972 Accepted by the faculty of the Department of Commu- nication, College of Communication Arts, Michigan State University, in partial fulfillment of the requirements for the Master of Arts degree. Guidance Committee: ACKNOWLEDGEMENTS The author wishes to express appreciation to Prof. Donald P. Cushman for his assistance in all phases of this study. His concern and guidance has made graduate study a challenge well met. Dr. Gerald R. Miller and Dr. David C. Ralph as members of my guidance committee provided the theoretic and methodological balance necessary for a complete approach to the study of human communication. Dr. G. Edward Notting deserves special recognition for his invaluable advice in the formulation of the methodology employed in the study. The author wishes to acknowledge Dena Brunsting for the pro- fessional service of typing and final preparation of the manuscript. A general noteof appreciation is extended to the support of family, friends and community of scholars necessary to the completion of such a task. 11 ACKNOWLEDGEMENTS . . . LIST OF TABLES . . . . LIST OF APPENDICES . . Chapter I II III IV INTRODUCTION . TABLE OF CONTENTS Purposes of the Study . Divisions of the Thesis . . Problems Which Gave Rise to Theoretic and Practical Solutions our Inquiry Rationale . . . . . . . . . . . . Rationale for Hypotheses. METHODS AND PROCEDURES Overview. . . . Pretest . . . . The Sample. a o Plesuge.o o o 0 Independent Variab RESULTS. . . 1es and Dependent Variables and Operationalizations . PfOCBdUtESooooaosooossoooooo O pera tionali zat ons Effects of Task on Message Rankings . . . . . . . Effects of Source Manipulation on Message Ranking Matching of Source to Messages. . . . . . . . . . CWCLUSIWS. O C C C O O O C O O C 0 Effects of Task on Evidence Usage . . . . . . . Effects of Source Attribution on Evidence Usage thhingT.ekooooooooooococooo Research Extensions . . . . . . . . . . . . . . iii Page ii vi war- hawi¢>hah>ro 16 16 16 17 18 18 20 21 23 23 26 29 31 31 32 33 Page LIST OF REFERENCES 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 35 APPENDICES O O O O O O O O O O O O O O O O O O O O O O O O O O O O O 38 iv LIST OF TABLES Design including dependent measures for each treatment condition and number of Sample, correct rankings, of significance for H1. Sample, correct rankings, of significance for H2. Sample, correct rankings, of significance for H3. Sample, correct rankings, of significance for H4. Sample, correct rankings, of significance for H5. Sample, correct rankings, of significance for H6. subjects in each cell . . . . . . prOportions, 2 value, and level proportions, 2 value, and level proportions, 2 value, and level proportions, 8 values, and levels proportions, 2 values, and levels 0 O O O O O O O O O O O O O O O O proportions, 8 values, and level 0 O O C O O O O O O O O O O O O O Percentages of subjects matching high and low credible sources with each message type for H7 . . . . . . . . . . Page 22 24 25 26 27 28 29 30 LIST OF APPENDICES Appendix Page I Proportions, 2 values, and Level of Significance . . . . . 38 II Instrument o o o s o s o o o o o o o o o o o o o o o o o o 39 vi CHAPTER I INTRODUCTION Evidence has been defined traditionally in the following manner: Evidence may be defined as facts and opinions used as the basis of reasoning. Facts are cases, sta- tistics, and physical objects--phenomena that are observed, described, classified, reported and pre- sented. 92inions are points of view--interpreta- tions and evaluations of facts-obeld by persons other than the person doing the reasoning. (Rance, Ralph, and Wiksell, 1969) Many studies in the area of evidence usage have approached the problem on a prescriptive basis. Evidence, defined in the above manner, is evaluated against the formal standard of adequate rational proof. Several researchers have broadened the base of such analyses by descriptively measuring evidence usage. This study will, it is hoped, be another step in that direction. Purposes of the Study The purposes of this study are three: 1) to determine the effect of high and low source credibility upon a descriptive set of evidence categories; 2) to determine the effect certain tasks have upon a descriptive set of evidence categories; and 3) to determine the effett of both source credibility and task upon a descriptive set of evidence categories. Divisions of the Thesis The divisions of the thesis will be as follows: 1) The remainder of this chapter will concern itself with a) an examination of the problem which gave rise to this study, b) a theoretic and practical rationale for a solution to this problem, and c) a state- ment of the hypotheses involved in identifying a solution. 2) Chapter II will outline the methodology employed in our inquiry. 3) Chapter III will present the results of our inquiry. 4) Chapter IV will draw some conclusions regarding our inquiry and make suggestions for further research. Problems Which Gave Rise to our Inggiry Rhetorical Theory suggests that evidence should be important in persuasion. To build a rational proof, speakers need evidence to establish premises and to lay down foundations for statements from which inferences can be made. In this context, theories of evidence are largely prescriptive, in that statements about evidence are cast in terms of the validity of arguments. Traditionally, theories of rhetoric hold that the proper use of evidence is central to establishing belief in the validity or probable truth of a proposition. In order to empiri- cally establish this relationship numerous studies have been conducted. However, these studies into the effects of evidence present a confusing array of results. First, several studies investigated the effect of evidence-plus-assertion and assertion-only speeches on attitude change. Cathcart (1955), Bostrom and Tucker (1969) and Kline (1969) found that a speech containing evidence and assertion is more effective in chang- ing attitudes than a speech containing only generalization and assertions. On the other hand, Costley (1958) and Wagner (1958) found no significant difference in attitude change between the same two types of speeches. Second,several studies investigated the effect of evidence-plus- assertion-plus-qualified source and assertion-only speeches on atti- tude change. Cathcart (1955) and Bostrom and Tucker (1969) found that speeches containing evidence attributed to qualified sources were sig- nificantly more effective in changing attitudes than speeches contain- ing only unattributed assertions. Ostermeier (1967) and Whitehead (1971), however, found no significant differences in attitude change between the same two types of speeches. McCroskey (1967) and McCroskey (1970) reported that evidence-plus-assertion-plus-qualified source speeches were significantly more effective in changing attitudes than assertion-only speeches when the speaker was originally perceived as low or moderately credible, but were not significantly more effective when the speaker was perceived as highly credible. One researcher, Bettinghaus (1953), discovered that speeches containing evidence-plus- assertion-plus-qualified source produced significantly more attitude change than speeches containing only evidence-plus-assertion, but his findings have never been replicated. Cathcart (1955), Gilkinson, Paulson and Sikkink (1954), Sikkink (1956) and Bostrom and Tucker (1969) all found no significant differences in attitude change between the two types of speeches. Furtheg Bostrom and Tucker (1969) reported that a speech containing evidence-plus-assertion-plus-source was significantly 1233 effective in changing attitudes than speeches containing evidence- plus-assertion-plus-source and qualifications (i.e., the ”qualified source” treatments noted above) or evidence-plus-assertion alone. This confirmed a similar, but non-significant, trend in the 1955 Cathcart study. Third, several studies investigated the effects of speeches 4 containing high and low quality evidence on attitude change. Warren (1971) found that speeches containing testimony from highly credible sources produced significantly more attitude change than speeches in which the same testimony was attributed to sources of low credibility. Dresser (1963) and Gardner (1966) found no significant differences in attitude change between messages containing high and low quality evidence. These latter results are consistent with the findings of a study by Harte (1971) in which he discovered that "...audiences are not notably successful at applying the appropriate tests of evidence to material offered as proof of an assertion.” Fourth, Wagner (1958) and Ostermeier (1967) manipulated the amount of evidence in experimental messages but found no significant differences in resultant attitude change. Finally, several studies investigated the effects of evidence on variables other than immediate attitude change. McCroskey (1967) reported that in a number of studies a source who was initially per- ceived as low to moderate in credibility was rated significantly higher in credibility if he included evidence in his speech than if he did not. This finding has been at least partially supported by the studies of Ostermeier (1967) and Whitehead (1971). McCroskey also reports that the inclusion of evidence in a persuasive message resulted in signifi- cantly greater delayed attitude change, regardless of the initial credi- bility of the source. A final discovery by McCroskey was that evidence served as an effective inhibitor to immediate counterpersuasion attempts. These conflicting and negative results demand a re-examination of the relationship of evidence and belief. An analysis of the previOusly cited research suggest four reasons for the conflicting results: 1) a prescriptive rather than a descriptive approach, 2) inadequate 5 treatment of task, 3) inadequate treatment of source, and 4) inadequate control of other intervening variables. One reason for the conflicting results of many evidence studies is the prescriptive approach mentioned in the introduction. Evidence is often evaluated as "satisfactory” or ”unsatisfactory" according to tests of logical adequacy. An example of such a prescriptive study is Cathcart (1955). This is an experimental study of the relative effect- iveness of four methods of presenting evidence. Evidence is never actually defined but is evaluated against prescriptive standards: 1) evidence is the basis from which 10gical argument is developed, 2) usually, the broader this basis, i.e., the more evidence presented, the more likely it is that proof will be generated, 3) evidence which has been evaluated by the so-called ”tests of evidence” is more likely to be valid, and 4) evidence which has been care- fully documented is generally more acceptable than un- documented evidence. This "evidence" is manipulated only in amount, documentation, and qualifications of source. The study results showed that evidence had an effect on changing attitudes, but the relationship was unclear. What factors operate in the relationship between evidence and attitude change cannot be isolated if evidence is arbitrarily defined and held constant in type and usage. Hance, Ralph, and Wiksell (1969) provide a step forward toward a descriptive analysis of evidence. In the rhetorical tradition they suggest that evidence be evaluated by tests of logical adequagy which are: A. Is the evidence clear? To be of value, evidence must be free from ambiguity and from the danger of being misinterpreted. B. Is the evidence consistent internally? Evidence will be worthless if it has contradictions within itself. C. D. E. F. Is the evidence consistent with other known evi- dence? If it is, the speaker can find other evidence to strengthen his point; if not, the new evidence must be stronger than the known evidence. Is the evidence relevant to the matter at hand? Sometimes we can cite evidence that may seem to prove our point but is actually concerned with a different subject. Is the source of the evidence competent? Does the person who presents the testimony possess the physical faculties and mental capacity to make him a competent source of facts and opinions? Is the source of the evidence free from prejudice? Is this person sufficiently objective or disinter- ested to make him a source we can regard as fair and unbiased? Is the source of the evidence reliable? Is he free from habits of superficial observation, irresponsible assertion, and inconsistent behavior? (p. 99) However, they also suggest that such tests are not sufficient. How the evidence is received by audiences must also be considered by apply- ing tests of pgychological adequacy: A. Is the evidence in harmony with the beliefs of the listeners? This test does not imply that the speaker should restrict his evidence to what will please his audience, or confirm what may be erro- neous beliefs; rather it suggests that the speaker should expect some resistance to evidence that does not coincide with the attitudes, values, and per- sonal beliefs of the audience, and he should take this resistance into consideration when he selects and presents his evidence. Is the source of the evidence a person whom the listeners are willing to accept? While the source may measure up in competence, freedom from preju- dice, and reliability, some resistance may be encountered if he does not measure up in social position, party affiliation, profession, and the like. (p. 99) These tests of psycholOgical adequacy bring us closer to a descriptive approach to evidence. However, evidence is still judged to be "satis- factory” or "unsatisfactory" by the speaker himself. Miller (1966) provides the best example of a descriptive approach to evidence. Evidence is defined as "those data that are intended to induce a sense of belief in the proposition which the data purportedly support." (p. 25). As this definition shows evidence is conceptualized as a diverse range of materials characterised most pre- cisely by the function they perform. This emphasized the manner in which evidence 9253 affect people rather than the manner in which it ‘ggghg to affect them. To this point Miller's approach is acceptable. Evidence must induce a sense of belief in a proposition to be satisfactory in its function. The evidence-assertion link, in other words, should be strong. However, the successfulness of such communication can be more precisely measured by a consideration of task superimposed on proposition and evidence. The importance of task in influencing behavior has been explored and established extensively. Collins and Guetskow's model (1964) talks of both interpersonal and task obstacles as determinants of behavior. Thibaut and Kelly (1959) define task as a "problem, assignment, or stimulus complex to which the individual or group responds by performing various overt and covert operations which lead to various outcomes." Shaw's current summary (1971) also isolate the task environment as one of the several environments determinative of behavior. Thus, task has been defined as an important factor in explaining variance in behavior. Two studies which have measured the effects of varying task are Morris (1966) and Hackman (1968). In Morris' study the question investigated was whether tasks perceived as different by members of groups would lead to different interaction patterns and different levels of perf pres 150‘ [651 Va ad performance. 3 tasks types, 3 levels of difficulty, and 4 orders of presentation were varied. The 3 tasks were: 1) problem solving tasks involving the implementation of solutions. 2) production tasks involving the generation of highly original solutions. 3) discussion tasks requiring consensus on some socially relevant issue. Though there were no overall significant differences in activity as a result of varying task type, significant differences were found for specific types of interaction. The conclusion was that tasks may 22; be considered equivalent, (Cappella, 1971). Hackman (1968) did a follow-up to Morris' study. In three of the seven measures of perfor- mance, task type explained more than 40% of the variance. Task diffi- culty, however, explained less than 8%. R.M. Gagne (1966) also analyzes the external environment in problem-solving situations. Gagne's analysis isolates the stimuli, the verbal directions, and the instructions as the key dimensions of the external situation. Hackman (1969) continued Gagne's line of study. He defined physical stimuli as physical elements mediated into symbolic form. Verbal direction serves to focus attention on important elements of the situation but not the operations. Instructions provide: 1) the goal, 2) the relevant concepts or 52133, 3) the conceptions of the relevant dimensions, and 4) a guide to thinking by reducing the number of alternatives. From these studies it is shown that varying the type and the structural nature of tasks is very important in explaining variance in human behavior. However, task has not been considered adequately in its effects on evidence categorization and usage. 9 The relationship between source credibility and evidence usage has been shown to be unclear in previously cited studies (p. 3). Some of the most recent writing on the subject done by McCroskey (1969) argues that the effects of source interacting with evidence resulted in previous conflicting research findings in evidence usage. A major theoretic problem in evidence usage is to determine the relationship between evidence and source credibility. Whether source primarily effects evidence, or evidence primarily effects source credibility, or, indeed, if the relationship is interdependent has not been clearly established. Therefore source must be carefully manipulated or con- trolled in any study of evidence usage. Several other factors need to be carefully considered in evi- dence research. McCroskey (1967) indicates that previous exposure to evidence by the receiver may reduce the impact of the evidence on the receiver. Familiarity with evidence should, therefore, be controlled. Also, in several of the evidence studies assertions from the evidence treatments were repeated in no evidence treatments with authority refer- ences removed. This resulted in differences in messages length and a redundancy effect. Troldahl, Costello and Robeck (1969) found that redundant messages produce more attitude change than non-redundant messages. Since redundancy results in differential effects it must be controlled. Theoretic and Practical Solutions It is necessary to descriptively categorize and evaluate evi- dence according to the actual patterns employed by subjects. Once these patterns of usage are isolated, evaluation of evidence usage can be done prescriptively, but against a realistic standard. McGuire 10 (1960), exploring the persuasive effect of dissonance producing messages, accounted for 90% of the variance in attitude change by measuring just two variables: 1) the probability that a set of propositions were 5525, and 2) the desirability of the consequences of a set of propositions. Infante (1970) and Clark and Hynes (1970) conducted research in which desirability was found to be an important type of evidence. Thus, at least two categories of evidence can be distinguished. In order to determine whether either evidence category is of importance, a no evidence alternative must be offered. No evidence treatments have been used as controls in research previously, but they have been imprecisely defined. The no evidence treatments usually differed from factual messages only in the deletion of authority refer- ences. A means of descriptively isolating patterns normally used by individuals in evidence usage is to have subjects rank a number of messages of different evidence types according to their utility in helping subjects perform a specific task. In this way evidence usage can be descriptively determined, given certain tasks. Other key factors, such as source, can be manipulated to determine their interrelationship with task in evidence usage. By observation of the results of such an interaction, a more descriptive theory which accounts for key factors effecting evidence usage can be built. Conceptual and Operational problems with distinguishing among evidence categories can be solved by categorizing evidence types and by concretely defining and operationalizing a no evidence treatment. It can then be determined whether subjects can distinguish among these types of messages and if subjects use these types differently for dif- ferent kinds of tasks. 11 It was pointed out that several key intervening factors need to be controlled or manipulated simultaneously. These factors can be easily controlled or manipulated through instructions, message wording, message tOpic, and attribution of various sources to the messages. Rationale As previously stated, the purposes of this study are to deter- mine the effect of high and low source credibility upon a descriptive set of evidence categories, to determine the effect certain tasks have upon a descriptive set of evidence categories, and to determine the effect of both source credibility and task upon a descriptive set of evidence categories. Miller's previously cited definition of evidence which refers to a diverse range of materials (data) which may be considered evidence depending upon the function they perform (inducing belief in a propo-I sition) suggests that evidence be broadly categorised. Three evidence categories are used in this study: truth, desirability, and no evi- dence. Truth evidence contains empirically verifiable data. Desira- bility evidence poses either positive or negative consequences resulting from the adoption of the proposition. The no evidence category is defined in this study as a message containing only definitions of terms within the proposition. The "categories" of truth and desirability suggested by McGuire's work refer to criteria for evaluation applied by the receiver of the message. Previous work has shown that subjects are able to differentiate evidence in at least these two ways. This study will determine whether these two categories are used differently under different task and source conditions. Subjects will be asked to use messages containing the evidence types to perform three tasks: 12 1) Categorization Task, differentiation and categorisation of truth, desirability, and definition messages, 2) Belief Task, rank ordering of messages which the subjects find most helpful in taking a stand on the proposition, and 3) Informative 2525, rank ordering of messages which the subjects find most helpful in understanding a proposition. Rationale for Hypotheses Previous research would suggest that truth evidence should be more highly ranked than desirability evidence in the understanding or informative task. Dresser (1963) found that "sound" or truth evidence does not significantly effect persuasion, but it is a significant factor in effecting understanding. On this basis the following hypothesis is offered: 1: Given the informative task of ranking messages in order of importance in understanding a preposition, a greater than chance frequency of subjects will rank order the messages in the following descending order of importance: 1) truth, 2) desirability, and 3) definition. Desirability messages should be more highly rated than truth evidence in the belief task. Clark and Hynes (1970) and McGuire (1960) report that positive attitudes are fostered by positing desirable and likely consequences. Infante (1970) conducted a study in which atti- tudes toward a propositional change in policy can be predicted from a knowledge of the desirability and likelihood ratings assigned to the consequences of the prOposed action. Given this reasoning, the follow- ing hypothesis is suggested: H2: Given the belief task of ranking messages in order of importance in taking a stand on a proposition, a greater than chance frequency of subjects will rank order the messages in the following descending order of importance: 1) desirability, 2) truth, and 3) definition. 13 Though truth and desirability have been determined to be bmpor- tant evidence categories, there is no direct evidence relating to subjects' ability to effectively categorise truth, desirability, and definition messages. However, Kline (1968) reported that subjects were able to effectively categorize evidence when provided with a cate- gory scheme, such as: 1) specific evidence, 2) non-specific evidence, and 3) no evidence. This would suggest the following hypothesis: H3: Given the task of categorizing messages into three evidence types, a greater than chance frequency of subjects will categorize the messages in the follow- ing manner: 1) the two truth messages in the truth category, 2) the two desirability messages in the desirability category, and 3) the two definition messages in the definition category. The second stated purpose of this study was to examine the effect of the manipulation of source on evidence categorisation and usage, given the above mentioned tasks. Four source conditions are used: 1) No source, 2) Role source, 3) High credible source, and 4) Low credible source. In the no source condition subjects read messages in which no sources are mentioned. In the role source condition subjects read messages attributed to unknown persons in well-known role positions. In the high credible source condition, subjects read messages attributed to sources perceived as highly credible by subjects. Finally, in the low credible source condition subjects read messages attributed to sources perceived as low credible by subjects. A distinction is made between high or low credible sources and role sources because research has indicated a differential effect between known and unknown sources. Kline (1970) isolated different types of encoding behavior which shows that differential effects occur in the presence of source credibility treatments and manifest source l4 treatments. Manifest source refers to the simple existence of a source without definite positive or negative evaluation. Greenberg and Miller (1966) provide another step in this line of analysis in a reported study in which the absence of personal experience with the source results in a somewhat positive evaluation of the source by subjects. These studies would indicate that a personally unknown source in a favorable role position would be evaluated somewhat positively. As has been established earlier, the relationship between source credibility and evidence is unclear. Several researchers, includ- ing Gilkinson, Paulson, and Sikkink (1954) and Sikkink (1956) have inves- tigated the effect sources have on evidence with few significant results. James McCroskey (1970) has done extensive research into how evidence effects source. One of the more intriguing findings reported by McCroskey is that a receiver mentally refutes a source on the basis of his past beliefs. This suggests the following hypotheses: H4: Given the informative task of ranking messages in order of importance in understanding a proposition, a greater prOportion of subjects in the no source condition than in any other source condition will rank order the messages in the following descend- ing order of importance: 1) truth, 2) desirability, and 3) definition. H5: Given the belief task of ranking messages in order of importance in taking a stand on a proposition, a greater proportion of subjects in the no source condition than in any other source condition will rank order the messages in the following descend- ing order of importance: 1) desirability, 2) truth, and 3) definition. H6: Given the task of categorizing messages into three evidence types, a greater prOportion of subjects in the no source condition than in any other source condition will categorize the messages in this manner: 1) the two truth messages in the truth category, 2) the two desirability messages in the desirability category, and 3) the two definition messages in the definition category. 15 The third purpose of the study is to determine the relationship between known high and low credible sources and evidence usage. This will be explored by having subjects in the no source condition match the pretested known sources with the messages they have read. Balance theory and McCroskey's work would suggest that high credible sources would be matched a higher percentage of the time with messages agreeing with the subjects' beliefs than with messages opposing those beliefs. McGuire's work on resistance to attitude change would also support this position. Subjects will bolster their initial attitudes by attributing messages supporting these attitudes to high credible sources. On this basis the following hypothesis is offered: H7: Given the task of matching sources to messages, subjects will have a greater tendency to match high credible sources than low credible sources to messages supporting their initial beliefs. CHAPTER II METHODS AND PROCEDURES Overview On the basis of pretest results, high and low credible sources were chosen. Subjects completed one of a set of randomly order ques- tionnaires. The questionnaires consisted of a general instruction page, six randomly ordered messages (three pro and three con) on one of the two prepositions. Each questionnaire contained messages in one of four source conditions: 1) no source plus matching task, 2) role source, 3) high credible known source, or 4) low credible known source. After reading the messages, subjects were given written instructions to per- form one of the three tasks: 1) Informative Task, 2) Belief Task, or 3) Categorization Task. The tasks consisted of ranking the six messages. Pretest Students in undergraduate courses were administered a pretest questionnaire, ostensibly to solicit student opinion on the two topics of the propositions, busing and drugs. These issues were believed to be salient and somewhat ego-involv- ing for undergraduate students. Each proposition was followed by a ten-point truth scale ranging from 0-1001 true. Subjects were asked to evaluate each proposition on this truth scale. They were then asked to evaluate the desirability of the proposition on a four-point scale from Very Desirable to Very Undesirable. Finally, subjects were asked to evaluate the favorability of each proposition by checking a 16 17 seven-interval, semantic differential scale from Very Favorable to Very Unfavorable. On the basis of this scale negative attitudes were established in the population from which the sample was drawn on both propositions. The sample rated the proposition advocating busing 2.8431 and a proposition advocating the searching of dormitory rooms for drugs 2.2941 on a scale of 1 to 7. The lower the mean, the less favorable the proposition. Clark and Hynes (1970), McGuire (1960) and Infante (1970) indicate a concern for the possible topic bound nature of their obtained results, limiting their generalizability. For this reason, two propo- sitions were used to control for topic. In order to locate high and low credible sources, subjects were asked to evaluate nine individuals on six, seven-interval, semantic differential scales bounded by the adjectives Safe-Unsafe, Untrained- Trained, Frank-Reserved, Closedminded-Openminded, Experienced—Inexperi- enced, Introverted-Extroverted (Berlo, Lemert and Mertz, 1966). Indi- viduals were scored by summing across the attitude scales on a one low, seven high basis. Thus, with six scales per individual, the possible range was six to forty-two. The nine individuals pretested were Edward Kennedy, George McGovern, Richard Daley, Ronald Reagan, Richard Nixon, Ms. Irene McCabe, Julian Bond, George Wallace, and B. F. Skinner. On the basis of this procedure two high credible sources, George McGovern and Edward Kennedy, were chosen, and two low credible sources, Ronald Reagan and Richard Daley, were chosen. The Sample Data were collected from 225 undergraduate students during July, 1972. All questionnaires were usable. Both males and females 18 were included in the sample. Subjects were sampled from the summer school undergraduate classes at Michigan State University. Messages Three basic message types were used in this study. a. Truth b. Desirability c. No Evidence (Definition) Two of each type were used-~one supporting the proposition, one attacking the pr0position. Thus each subject received six basic messages, two truth messages (pro and con), two desirability messages (pro and con), and two no evi- dence messages (pro and con). Half the subjects received six messages concerning the busing proposition and half received six messages con- cerning the drug proposition. Familiarity with the evidence was controlled by creating all evidence used in the messages. Thus, all subjects received equally new and thus unfamiliar evidence. Given that the no-evidence messages contained definitions of terms rather than a restatement of evidence used in other messages, there was no redundancy within the messages, thus controlling this key factor. Indgpendent Variables and Operationalizations Source Manipulation 1. No Source Condition. One-fourth of the overall sample received the six basic messages without sources attributed to the messages. l9 2. Role Source Condition. Another fourth of the subjects received the six basic messages attributed to unknown sources with known roles, e.g., professor, doctor, research director. 3. High Credible Known Source Condition. Another quarter of the subjects received the six basic messages attributed to the two high credible sources. For each subject in this condition, the three pro messages were attributed to one of these two sources, and the three con messages were attributed to the other credible source. This was done for both prepositions. 4. Low Credible Known Source Condition. The two sources judged by the pre-test sample to be lowest on the previously cited dimensions, Ronald Reagan and Richard Daley, were used in this condition. Messages were attributed to these two sources in the same manner as was done with the high credible sources. The remaining quarter of the overall sample received the basic messages attributed to these low credible known sources. Task Manipulation Within each of the source conditions, separate groups of sub- jects performed three tasks. 1. Informative Task. Approximately one-third of the subjects within each of the source conditions were asked to rank all six messages on the criterion of how helpful each was to the subject in gaining an understanding of the proposition. 2. Belief Task. Another third of the subjects in each source condition were asked to rank all six messages on the criterion of how helpful each was to the subject in taking a stand on the proposition. 20 3. Message Categprization Task. The remaining third of the subjects in each source condition were asked to categorize all six messages into three categories of evidence: 1) Truth, 2) Desirability, and 3) Definition. Depgndent Variables and Operationalizations Three measures of task performance were employed, one for each of the task manipulations. l) Informative Task Performance. The rank orderings of messages by the subjects performing the informative task were compared against a criterion rank order earlier predicted in the hypotheses, and the number of errors in each subject's rank ordering was determined. If a subject's ranking contained one or zero errors, he was categorized as being correct. Subjects who had more than one error were categorized as being incorrect. 2) Belief Task Performance. As in the measure of informative task performance, subjects' rank orderings were compared against a cri- terion rank order predicted in the hypotheses, and subjects were cate- gorized as being correct or incorrect. Again, zero or one error were considered correct. 3) Message Categorization Task Performance. Subjects' cate- gorizations of the six messages were compared against a criterion cate- gorization based on the type of evidence used in the construction of each message. As before, zero or one error was considered correct, and subjects were categorized as being correct or incorrect on this basis. MatchinggTask In order to explore source credibility as it effects subjects' attribution of sources to types of messages, a matching task was included. All subjects within the no source condition were asked to 21 choose six sources from the list nine pretested and match them to the message each source most probably stated. For each message the percent- age of the subjects in the no source condition matching each source to that message was determined. Procedures Instrument Construction. A cover sheet on the front of each booklet presented general instructions. The six messages followed and were randomized for each booklet. Next a page of instructions specific to the required task was presented, followed by that task (Informative, Belief, or Message Categorization). For the subjects in the no source condition, the matching task was presented next. The final page for all subjects asked a series of demographic questions (age, class in school, sex, grade point average, and previous course work in persuasion or argumentation). Randomization of Treatments to Subjects. The booklets were placed in random order, to assure that when they were distributed to a class, all source conditions and all tasks were represented. Classes were chosen to prevent any bias caused by a particular time of day or day of week. Procedures for Data Collection. The experimenter went to class- rooms and distributed the booklets. A short introduction was given, and the instruction page was read aloud. Then the subjects completed the booklets. Near the end of the task, the experimenter asked the subjects to recheck their booklets to make sure the task was done as instructed. Booklets were then collected and a short debriefing was given. The entire procedure took approximately twenty minutes. 22 Table 1. Design. The design is summarized below, including dependent measures for each treatment condition and number of subjects in each cell: Known High Known Low No Role Credible Credible Source Source Source Source Condition Condition Condition Condition Inf. Rank- Inf. Rankings Inf. Rank- Inf. Rank- Informative ings and ings ings Task Matching (N-ZO) (N-ZO) (N-ZO) (N-18) Belief Rank- Belief Rank? Belief Rank- Belief Rank- . ings and ings ings ings Belief Task Matching (N=l9) (N-20) (N-l6) (N220) Categori- Categori- Categori- Categori- Categorization zations and zations zations zations Task Matching (N=20) (N215) (N=20) (N-l7) CHAPTER III RESULTS Effects of Task on Messagg Rankings Informative Task It was predicted by H that subjects would rank messages in the l informative task in the order of truth, desirability, and definition with a greater than chance frequency. In order to test this hypothesis an expected frequency distributed was generated by determining all possible orderings of six messages: two truth messages, two desira- bility messages, and two definition messages. It was found that there are 78 orderings of three message types, two of each type. The six messages were labeled by the following letters: A - Truth pro message B - Truth con message C - Desirability pro message D - Desirability con message E - Definition pro message F - Definition con message In each hypothesis a criterion ranking was predicted. As described in Chapter II the criterion ranking or one error (two out of place messages) would be considered a correct ranking. An example of this procedure would be as follows: in the Informative Task a com- pletely correct ordering of the messages according to the hypothesis 23 24 would be: 1) A, B or B, A (truth messages), 2) C, D or D, C (desira- bility messages), and 3) E, F or F, E (definition messages). Also considered correct would be one error which would necessitate two mis- placed letters. An example of one error would be C, A, D, B, E, F. Letters B and C are misplaced. From the expected frequency distribution there are only thirteen of the 78 possible combinations which include one or less errors. Thus, by chance alone, one would expect the propo- sition of correct rankings to equal l3/78's within a population of subjects ranking the six messages. To test H1 the preportion of subjects ranking the messages cor- rectly was compared against the expected population proportion. A 2 test for proportions was computed to determine if the sample correctly ranked messages with a higher frequency than expected by chance. Out of the total of 20 subjects in the no source condition performing the task, 9 ranked the messages correctly. This proportion (0.45) signifi- cantly exceeded the expected proportion of 0.167 (3 - 3.41; p<:.001). Therefore, given the informative task, subjects ranked the messages in the order predicted (truth, desirability, and definition) beyond chance expectations. Table 2. Sample, correct rankings, prOportions, 8 value, and level of significance for H1. Informative Task (H1) Actual Expected Correct by S's Rankings Chance p (Sample) P (Expected) 2 ‘2 No Sourc Conditio 20 9 13/78 .450 .167 3.41 '(.001 25 Belief Task To test Hz the same procedure as used to test H was implemented. 1 From the expected frequency distribution the predicted ranking of mes- sages (desirability, truth, definition) would occur 13/78's of the time within a population of subjects ranking the six messages. Again one error, i.e., two messages out of place, was considered a correct ranking. The proportion of subjects completing the belief task who ranked the messages correctly was 0.368. In absolute numbers, out of 19 subjects in the no source condition performing the task, 7 correctly ordered the messages. This significantly exceeded the proportion expected by chance, 0.167 (_3_ a 2.42; p<.01). Therefore, given the belief task, subjects ranked the messages in the order predicted (desirability, truth, defini- tion) beyond chance expectations. Table 3. Sample, correct rankings, proportions, 2 value, and level of significance for H2. Belief Task (H2) Actual Expected Correct by S's Rankings Chance p (Sample) P (Expected) 2 .2 No Source Condition 19 7 13/78 0.368 0.167 2.42 ‘<.01 Catggprization Task From the expected frequency distribution used to test H1 and H2, it was determined that by chance, 13/78's of the subjects would correctly categorize the six messages with one or less errors. It was found that the proportion of subjects in the categorization task who correctly categorized the messages was .70, i.e., 14 out of 20 subjects ordered the six messages correctly in the no source condition. This greatly exceeded the proportion expected by chance, 0.167 (3 a 6.4; p‘-' \ ' ; A" ‘. ,. J‘ C. Proposition - University administrators must be given the right to search student dormitory rooms for drugs. Edward Kennedy recently argued that one consequence of the early identification and counseling of drug users is that they are less likely to become serious drug abusers. In-depth interviews with twenty—two students who had been treated for drug addiction revealed that student dormitory rooms were the most frequently used place for, storing drugs. If students can be identified in the early stages of drug usage, ed- ucational and psychological counseling can frequently prevent drug ad- diction. Thus, university administrators must be given the right to search student dormitory rooms. a ‘( l. _\ .‘i ,Lui. .1n'. . I. . L I .nl‘ .' l _ ”I. q. .l. 1. “.IE . "V- . l i D. Proposition - University administrators must be given the right to search student dormitory rooms for drugs. Some people, however, believe that university administrators must not be given the right to search dormitory rooms because this would be a violation of individual rights. According to Dr. William Cartwright, Professor of Law at Columbia University, the right to search a student's dormitory room would violate the student's constitutional right of free~ dom from unlawful search and seizure. James Frederick, Professor of Sociology at American University, argues that students might come to fear, rather than respect, the authority of the administration if it were allowed to search student dormitory rooms. Thus, university adminis- trators must not be given the right to search student dormitory rooms. ff“. . .4 ~‘- l... ,- 'I‘” m—M- C) I; II in .\ . ”:1! . t “)I- E. Proposition - University administrators must be given the rightto search student dormitory rooms for drugs. The reason that some people believe this is that narcotics are drugs. According to Edward Wilson, research director of the Justice Department, narcotics are legally defined as addictive drugs. Addictive drugs may appear in many forms—-pills, liquids, powders, etc. According to Dr. James Greene, Professor of Sociology at the University of Illinois, stu- dent dormitory rooms are temporary domiciles in which students keep books, clothes and other personal belongings. Thus, university adminis- trators must be given the right to search student dormitory rooms. i .' ifn r. J :73 V r 3 . VJ. Ii 5‘ I Proposition - University administrators must be given the right to search student dormitory rooms for drugs. Some people, however, believe that university administrators must not be givencthe right to search student dormitory rooms because admin- istrators are systems analysts. According to Dr. Richard Bartlett, Pro- fessor of Education at the University of Southern California, university administrators are normally defined as the president, provost, registrar and deans of the various colleges. Martin Pierce, regional director of the National Education Association (N.E.A.), argues that an educator func- tioning as a systems analyst must have expertise in the social, psycholog— ical, and maintenance system of universities. Thus, university adminis- trators must not be given the right to search student dormitory roon‘ . - "l (Y1. r {‘1'- . ‘ ~ q ~\ - - ~o- (.L, - 4 v ~l -1 - - - I v .L - fl . .L‘t- J I V > . .- 1 I A. .‘. .. . D I ' L‘ 1 J ‘ ( i! 'l!' 1 f . ' .- ,r.” ““1 “J""~"’J’€: I11 33:1...) ..‘ “‘ ";-r'.’v"’-.f:f§ F3“ u - L ‘V v.. s . t. . . ' i '- =~‘ O \ -J Proposition - Cross-district busing is a necessary tool in achieving racial integration in our nation's schools. According to Ronald Reagan: "Busing to another school district is not the cause of improved academic performance. A study of three pre- dominately black schools districts in Illinois revealed that with in- creased funds providing more highly skilled teachers and more adequate teaching facilities, students improved their academic performance sig- nificantly. A nationwide survey found that none of the 800 students studied.were any more successful academically in a predominately white suburb school than they had in their own neighborhood school. Thus, cross-district busing to achieve racial integration in our nation's schools in not necessary." . . . i I ‘ _ u,’ . . _ . I .. ‘ ’ '3 1.4 J .lJ ' , K A. ' ’ K 1- .v. ,___ u- I . Q ‘ r ..t .. r r. ‘r' 1 3 ' .‘f ' ' 5 iv ‘ . ’4: _ y , ,' f"- t l O H i{“W’c{“-' ”fw' 3‘ J ' ‘1?" ‘ ' ‘ ‘. y"! r“ ' ‘ .LL -.-..‘ - ' . . ‘ ‘ . . . )-..1 , I . . .p ‘— - 1‘ v - .. - . l- -. — . ,«rH, ~ 3:337} ._ 3K - .E'. ,1: ‘ - {I - "'-_ ‘ " ' ?: .1: 4.‘ -r.' . ' ' ' '- -.. . ‘ J . J" .L ‘ 1 a .i ‘ .P' i- .1- 4“ ‘1' I 'i1-2j'fwfl.'.. 9.416;: 1:31,: '3 , fr'::';7 WI: Kerr: :3! 713.3; .5965" , J.,. . . . _'.."'..—_L -.",... ,~ :. ”Th/J is... w s -.-| , ‘LL .- 3, . '1 L33 :1. J - "wins-~39 - I. ,.L ,, , "\ 3 . f. .'. ., ,. . ,. ‘ . _,... U . . ~-' , .. ‘ _-, . '. ')LI.J'.‘ Jo} N: 7 ‘ .I ' 'l ; 1U -’.”.‘.’I"- ' - '/.'..-.':.‘_- f1 ‘ . .' ' ‘ v . ' ‘ ‘ - ’\ a” ‘- . r ., 'L31‘Tl"’31.-:“!‘7 13,-. m E- .4 ' . -- :m A” u,._., .. A, . 3. ’ ., .'. - ' - 'A ' '. - “1.1.- b.’ in" . . 7“ ‘2 .;- .3; “3 J. ”MU .[Ofifi .’ (1 ”In-" ' '.- . ' '4 . — -. ..--'. 1'. ' ., ,,- - ' > ‘ — ,, ,. '. _ ' "‘u : - ‘: .‘l .T‘T.‘ £11: "11:- .‘Zil. .L.‘_’ ~-“ ~ ‘. ‘13.;4‘ ‘1..E " - 3.1 'J’.“ C. Proposition - Cross-district busing is a necessary tool in achieving racial integration in our nation's schools. Edward Kennedy recently argued that one consequence of cross- distric busing is that daily interpersonal contact between minority and white students results in more awareness of each other and a greater likelihood of building cooperative social systems. In—depth interviews with fOrty black and white students in racially mixed schools revealed that they had more pleasant interpersonal friendships with students of another race than a similar group of students in schools of predominately one race. If students are exposed at early ages to a racially mixed atmosphere, they are less likely to develop negative stereotypes of different ethnic groups. Thus, cross-district busing is a necessary tool in achieving racial integration in our nation's schools. 7121'" 4"." ' “II - l o .V. ~ 0 n . a. II t... . IA r) a). ‘1. . .- 4 . an I. .J . . . ul . I! ~. .1. 7 1. vi .‘4 .. .i Pr0position - Cross-district busing is a necessary tool in adhieving racial integration in our nation's schools. Ronald Reagan contends that cross-district busing to achieve racial integration in our nation's schools is undesirable because it is a violation of individual rights. The busing of school children out of their own school districts is a violation of individual rights. If the parents pay property taxes to support a particular school system, it is a violation of their rights to have their children bused into another, possibly more inadequate, district. Resentments caused by forced cross- district busing in violation of individual rights will result in increased racial tension in our schools. Thus, cross-district busing to achieve racial integration in our nation's schools is unnecessary. ‘ 'l'i'x-f‘ 50"". 'J ' ( vi f .n. “a '- -.. . . 0L: 113.0 3‘ t JV- #4 .4 _. Tr til :32".- r V. -\ ”I. r! a-“ . ..t :- Ila f.- .. . 1.17%! L ll .1 a I.. a 0'.— ,1 .9; .. .| . to” .x. .- V. .I. .1 n0. ... a d n r or. .‘ . .a. \. . : a I. c. . {I u: +J II. . . .. . I: ... .,. w: .rlw . aw ,- I “ 3' y.- BO Proposition - Cross-district busing is a necessary tool in achieving racial integration in our nation's schools. Edward Kennedy believes that cross-district busing involves the movement of children from one school district to another in order to achieve an equitable racial balance within schools. Racial integration is the process by which all public institutions must be racially balanced if possible. Thus, cross-district busing is a necessary tool in achieving racial integration in our nation's schools. I . ,. . . ..C r. w rr . .la. n I. a . 1. . . .- 1 . . . .0 . I-i. to ,r a , .: J l. f”. F? '2" l" a.) I. a; 1 “a. . _. 8. PJ 1 . . . v . ¢ . i .. v, V\ .'. _ . ll. _ .. \a4 .. ,(Kr “ l 5 31 _ v D ' . I. .u ..a /\ \l vl‘; . i r 1. n‘ 'll. :. r. q. . u U .- i. ..A o r .. . r." . . . . x ‘v ‘. -'r\ \. J r‘.” F. Propggition - Cross-district busing is a necessary tool in achieving racial integration in our nation's schools. Ronald Reagan believes that cross-district busing is not a necessary tool in achieving racial integration in our nation's schools because school is the institution in which children are both educated and social- ized into society. Busing is the means of transporting students from outlying areas of a district to and from school. Thus, cross-district busing to achieve racial integration in our nation's schools is unnecessary. A. Proposition - University administrators must be given the right to search student dormitory rooms for drugs. Richard Daley stated in a recent speech: "Many students have been found to store drugs in their dormitory rooms. In-depth interviews with twenty-two students who were treated for drug addiction revealed that student rooms were among the most frequently used places for storing drugs. A nationwide survey produced similar results. This survey of 1,200 college drug users revealed that 68.5% of the students listed student rooms as one of the most commonly used places for storing drugs. Thus, university administrators must be given the right to search student dormitory rooms." Proposition - University administrators must be given the right to search student dormitory rooms for drugs. According to Ronald Reagan: "University administrators must not be given the right to search student rooms because most students do not store drugs in their rooms. A study of three Illinois colleges revealed that none of the students interviewed had used their rooms to store drugs. A nationwide study found that none of the 800 students he interviewed had stored drugs in their dormitory rooms. Thus, university administrators must not be given the right to search student lockers." . . an, ru- 1 4;:1‘; .. -‘-I 5T ‘0 3;. ”.1 3.22.1!" o {1 :1 ‘3 :1? .s5 (1513“ C .. '11- C. Proposition - University administrators must be given the right to search student dormitory rooms for drugs. Richard Daley recently argued that one consequence of the early identification and counseling of drug users is that they are less likely to become serious drug abusers. In-depth interviews with twenty-two students who had been treated for drug addiction revealed that student dormitory rooms were the most frequently used place for storing drugs. If students can be identified in the early stages of drug usage, educa- tional and psychological counseling can frequently prevent drug addic- tion. Thus, university administrators must be given the right to search student dormitory rooms. .l f . w...“ .. ._; 3' "I J _ “man—'w. . | 1 ' 7“".{0 v '1' ‘s o .31, CW 1-. “L s | \ n. . .T. 1V y . - l . s l 4. . ..J .. ._ .5. . Ia. d . N I v . 5. fl. - .. co; ‘ rr . I .. .I. f. r .v .. . I; a. . p I .r. . < _| . . -.. ,l . T q. .l \ u I. L. .1. I a.. .\ . ... . 1. . . .. .1, I . . . . . PrOposition - University administrators must be given the right to search student' dormitory rooms for drugs. Ronald Reagan contends that university administrators must not be given the right to search dormitory rooms because this would be a violation of individual rights. The right to search a student's dor- mitory room would violate the student's constitutional right of freedom from unlawful search and seizure. Students might come to fear, rather than re5pect, the authority of the administration if it were allowed to search student dormitory rooms. Thus, university administrators must not be given the right to search student dormitory rooms. .. * 3?? Iativ ' .jJ.'..5.'1.J’/:‘,’T'J‘ - at”; ' - 7L: E L' T 1 r‘ . f. - -ndtn :d! . ’ ' xv: -. M H . _ .r. “p. . .': d ' -2 \ . t3. «1",! ."" l ” \ ‘ ,,_ .. _. ‘ a " ".‘}|'“‘N . a id - \ {\.):T ‘\"':|-‘ rpl' A :'.' ‘l \ . i I .- ., .. . - ’ ‘1. E. Proposition - University administrators must be given the right to search student dormitory rooms for drugs. Richard Daley believes that narcotics are drugs. Narcotics are legally defined as addictive drugs. Addictive drugs may appear in many forms--pills, liquids, powders, etc. Student dormitory rooms are tem- porary domiciles in which students keep books, clothes and other per- sonal belongings. Thus, university administrators must be given the right to search student dormitory rooms. a .- a. _ .7.” .3) 7 . . .. . ~ . _ l x i .0 . .r . . . ... . ... \ , ‘ ,> Dr... . .. _ .. ., .. 1. v. ..4 . v . .. . . I L v. Proposition - University administrators must be given the right to search student dormitory rooms fer drugs. Ronald Reagan believes that university administrators must not be given the right to search student dormitory rooms because administrators are systems analysts. university administrators are normally defined as the president, provost, registrar and deans of the various colleges. An educator functioning as a systems analyst must have expertise in the social, psychological and maintenance system of universities. Thus, university administrators must not be given the right to search student dormitory rooms. Instructions : Now that you have read all six messages, we would like you to do another task . By each message is a letter A through F. We would like you to rank order the messages according to how much they helped you understand the proposition. You must rank all the messages; that is, you cannot me any letter more than once. Please place the letters corresponding to the blanks below: 1. The message 2. The message 3. The message u. The message 5. The message 6 . The message that helped me that helped me that helped me that helped me that helped me that helped me understand the understand the understand the understand the understand the understand the messages in the most is second most is third most is fourth most is . fifth most is sixth most is Please check j:_o_ make sure you have used all the letters. "fl! ”~- 'Is‘. .. Ins tructions : Now that you have read all six messages, we would like you to do another task . By each message is a letter, A through F. We would like you to rank order the messages according to how much they helped you take a stand a! the proposition . use any letter more than once. You must rank all the messages; that is, you cannot Please place the letters corresponding to the messages in the blanks below: 1. 2 . 3. message ness age message message message message Please check to make that helped me take _a_ stand the most is that helped me take a stand the second most is that helped me take a stand the third most is that helped me take a stand the fourth most is . that helped me take a stand the fifth most is that helped me take a stand the sixth most is sure you have used all the letters. . a .. .- -\ til I . s - Instructions : Now that you have read all six messages, we would like you to do another task. By each message is a letter, A through P. we would like you to choose messages that helped you in certain situations, and indicate- which ones you chose by placing the letters of those messages in the blanks below. You cannot 23 3 letter more than once, so be sure you choose different usage: for each question. 1. Which two messages provided you with the most evidence pertaining to the truth of tits-proposition? and . 2. which two messages provided you with the most evidence pertaining to the desirebmty of the proposition? and . 3. flhich _t_w_o_ messages provided you with the met definition _o_._‘._ term within the preposition? and . Please 13: sure 12“. have 99; used 3 letter more than once is; answering the W's—guesaons. Instructions: Now that you have read all six messages, we would like you to do another task. By each message is a letter, A through P. For each message we would like you to decide which person, from the nine listed below, is the one m likely to have written each message. Place the number of that person in the blank following each question. You cannot use any person more than once. 1. Ted Kennedy a. Richard Daley 7. Irene McCabe 2. Richard Nixon 5. Julian Bond 8. George Wall' 3. B. F. Skinner 6. Ronald Reagan 9. George McGovern 1. Which person do you feel was most likely to have written message A. 2. Which person do you feel was most likely to have written message B. 3. Which person do you feel was most likely to have written message C. ' - u. Which person do you feel was most likely to have written message D. 5. Which person do you feel was most likely to have written message E. (D e Which person do you feel was most likely to have written message P. Please check _t_o_ make sure you have not used _a_person more than once. -.... on ..,, l' -.... '>:- .- ' Y! \ - s "I‘. y . C .. _ 1| , 5’ - ...—.. ‘-_. . . .“I u' . . 'O ' .n: . A o '.,'.r.. ’ cm .‘,,'I .. . ’ '-_l.' .' -: | .- ‘.. U' . - . 4| e‘» '~n‘- " .x .. , . _ .. ‘— -- I.’ I .- ‘ i \. f. '-"' es. 2-. 3'1 I :k I 3 u Finally we would like to ask some questions d>out you. 1. what is your age? years. 2. Are you (1) male (2) female ? 3. What is your college class? (1) freshman (2) sophomore (3) junior (It) senior u. “bet is your cumulative grade point average (GA)? ( ) 3.51 - u.oo AI.“ ) 2.01 - 2.50 ( ) 2.00 or less 5. Have you ever taken a course in persuasion or logic? ( ) Yes ( ) No If yes, what course(s)? HIGH 6 B QRIES uuuun m 5