ff}. / / ., . dew-72.. :\i 3.1. AU rifle- ABSTRACT DOGMATISM AS A PREDICTOR OF COMMUNICATION BEHAVIOR IN THE DIFFUSION OF CONSUMER INNOVATIONS by John H. Holmes The present study proposed the thesis that the socio—phychological theory of beliefs as formulated by Rokeach1 would generate new insights into some of the communication behavior involved in the diffusion and adoption of a consumer innovation, the Ford Mustang automobile. The six distinct areas_ofiggmmnninaLiQQ.hfihanior investigated were (1) communica- tion sources, (2) mass media communication channels, (3) innovativeness, (4) product loyalty, (5) Opinion leadership, and (6) homophily. The basic objective was to identify and measure the relationship which .3; ~w.. 7‘, existed between dogmatism and each of these six concepts. Dogmatism was defined as a personality variable which governs a person's receptivity or lack of receptivity to new ideas and further includes how a person perceives, evaluates, acts, and reacts to such ideas. A review of the literature combined with an assessment of the theoretical position of belief Systems led to the formulation of six general hypotheses, which were investigated. The hypotheses were as follows: GH 1: HIGH DOGMATICS UTILIZE COMMUNICATION SOURCES I‘DRE CONSISTENTLY THAN LOW DOGMATICS. GH 2: LOW DOGMATICS EXPOSE THEMSELVES MORE TO MASS MEDIA COMMUNICATION CHANNELS THAN HIGH DOGMATICS. GH 3: LOW DOGMATICS ARE MORE INNOVATIVE THAN HIGH DOGMATICS. GH 4: HIGH DOGMATICS EXHIBIT GREATER LOYALTY TOWARD PRODUCTS THAN LOW DOGMATICS. GH 5: LOW DOGMATICS EXHIBIT MORE OPINION LEADERSHIP THAN HIGH DOGMATICS . GH 6: HIGH DOGMATICS EXHIBIT GREATER HOMOPHILY THAN LOW DOGMATICS. Twenty-three empirical hypotheses which were derived from the six general hypotheses were tested. A systematic ordered sample of 150 Mustang owners residing in two adjacent Northwestern Ohio counties located in the interurbian strip extending between Toledo and Cincinnati were interviewed in person during a four week period which extended from November 17 through December 13, 1966. In addition to answering questions pertaining to the six aspects of Communication being investigated, each respondent complet- ed a short-form dogmatism test.2 The scores of these tests were correlated with the responses obtained from the other questions. A zero-order correlational analysis of the data led to the rejection of all but one of the twenty-three empirical hypotheses. The only correlation which was significantly different from zero in the pre- dicted direction was for EH 48: High dogmatics state 2 preference for replacigg‘ghgig original Mustagg with g 333 Mustang more often Ehgg‘lgw doggatics. Eight of the correlation coefficients were in the expected direction, but fifteen were in the direction opposite to that which was predicted. Two of the latter were significantly different from zero. ii Consequently, none of the six general hypotheses were accepted. The findings were not appreciably affected by a secondary analysis which consisted of the statistical elimination of control variables significantly correlated with either the independent variable and/or the dependent variable(s). The unexpected results could have been partially attributable to many factors which may have included (1) inherent weaknesses in measurements, (2) nature of the innovation, (3) representativeness of respondents, and (4) interviewer bias. Nevertheless, the results clearly revealed that dogmatism as measured in the present study is of little value in predicting communicative and adoptive behavior. Consideration of the findings indicates that individuals high in dogmatism do behave somewhat differently than those low in dogmatism and these slight differences become more noticeable in the areas of product loyalty and opinion leadership. Because of the importance of the concepts considered in the present study and their relevance to marketing theorists and practi- tioners, additional investigations in these areas should be undertaken. 1 Milton Rokeach, The Open and Closed Mind, Basic Books, New York, 1960. 2Verling C. Troldahl and Frederic A. Powell, "A Short-Form Dogmatism Scale for Use in Field Studies," Social Forces, vol. 44, no. 2, December 1965. iii DOGMATISM AS A PREDICTOR OF COMMUNICATION BEHAVIOR IN THE DIFFUSION OF CONSUMER INNOVATIONS by John Henry Holmes A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Communication 1967 @DCopyright by JOHN HENRY HOLMES 1968 ii Accepted by the faculty of the Department of Communication, College of Communication Arts, Michigan State University, in partial fulfillment of the requirements for the Doctor of PhiIOSOphy degree. Zia 74721,... Director of Thesis Guidance Committee: 77/ 6: :ZZ:::é214§égy/f'Chairman W7 ‘.'4 4 All .4] ' (Ll/ll * .7 1‘]: 1.4 .1' A ' AV ’ ‘gr* ‘ a ’ / r 15%! z? .;LQ7ZI.ZJ%/r iii ACKNOWLEDGEMENTS I would like to acknowledge Dr. Everett M. Rogers, director of this disseratation, who generously gave of his time to assist me in its preparation. His understanding and kindness were a constant source of inspiration for which I will always be most appreciative. Grati- tude is owed to my doctoral guidance committee composed of Dr. Verling C. Troldahl, Chairman, Dr. Kénward L. Atkin, Professor John W. Craw- ford, Dr. William J. E. Crissy, and Dr. Gerald R. Miller. It was a pleasure to work under the direction of these men, and I know that my completion of the doctoral program would not have been possible without their guidance and counsel. I wish to thank the Marathon Oil Company and the Cooper Tire and Rubber Company for their financial support. A special note of thanks is given to my colleagues in the Marketing Department at Bowling Green State University for their encouragement and to the Computer Center for its assistance. Finally, I would like to acknowledge my wife who assisted in the collection of data and also patiently bore the trials and tribulations of graduate education. iv TABLE OF CONTENTS CHAPTER I. INTRODUCTION . . . . . Diffusion and Adoption of Innovations. . . . The Communication Process. . . . . . . . . Characteristics of the Innovation. . . . . Characteristics of Adopters. . . . . . . . ,Personality Variables in The Innovation Studied: Description of the Study Implications . . . Limitations . . . Objectives . . . II. CONCEPTUAL FRAMEWORK . Theory . . . . . . . The Concept of Dogmatism . Dependent Variables. . . Communication Sources. Communication Channels Innovativeness . . Product Loyalty. . Opinion Leadership Homophily. . . . . Control Variables. Diffusion Research. The Mustang . . . . PAGE 11 12 13 13 15 15 17 19 19 23' 27 31 35 38 43 III. METHODOLOGY. Operationalization. Empirical Hypotheses. Measures. Instrument Development. Pretest . Interview Schedule Form . Setting . Data Collection. Sample Selection. Interview Procedure . Interview Summary . Sample Description. Description of the Variables. v“ Source Utilization. ‘f' Media Exposure. Innovativeness. Product Loyalty . Opinion Leadership. Homophily . Dogmatism . Analysis of the Data. Indeterminate Responses . Sample Size Variation . Primary Analysis. Secondary Analysis. vi 44 44 44 46 54 54 55 56 60 60 61 62 63 64 64 65 66 67 68 69 7O 70 7O 71 71 72 IV. RESULTS. . . . . . . . . . . . . . . . . . . . . . . . . . . 73 General Hypothesis 1. . . . . . . . . . . . . . . . . . . 73 General Hypothesis 2. . . . . . . . . . . . . . . . . . . 73 General Hypothesis 3. . . . . . . . . . . . . . . . . . . 77 General Hypothesis 4. . . . . . . . . . . . . . . . . . . 78 General Hypothesis 5. . . . . . . . . . . . . . . . . . . 81 General Hypothesis 6. . . . . . . . . . . . . . . . . . . 82 V. SUMMARY AND DISCUSSION . . . . . . . . . . . . . . . . . . . 87 Summary . . . . . . . . . . . . . . . . . . . . . . . . . 87 Explanations for the Results. . . . . . . . . . . . . . . 89 Relationships Between the Variables . . . . . . . . . . 90 Representativeness of the Subjects. . . . . . . . . . . 93 Interpretation of the Results . . . . . . . . . . . . . . 94 Communication Sources . . . . . . . . . . . . . . . . . 94 Communication Channels. . . . . . . . . . . . . . . . . 94 Innovativeness. . . . . . . . . . . . . . . . . . . . . 96 Product Loyalty . . . . . . . . . . . . . . . . . . . . 98 Opinion Leadership. . . . . . . . . . . . . . . . . . . 100 Homophily . . . . . . . . . . . . . . . . . . . . . . . 102 Commentary. . . . . . . . . . . . . . . . . . . . . . . 104 Comparing Laboratory Research with Field Studies. . . . . 104 Personality Variables in Perspective. . . . . . . . . . . 106 Implications for Future Research. . . . . . . . . . . . . 108 BIBLIOGRAPHY . . . . . . . . . . . . . . . . . . . . . . . . . . . 110 APPENDICES . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 vii TABLE 10 11 12 l3 14 15 LIST OF TABLES Perceived Characteristics of the Innovation Related to Its Rate of Adoption. Social Characteristics of Adopters Related to Innovativeness. . . . . . . . . Social-Psychological Variables Related to Innovativeness . . . . Data Collection Summary. Obtained Dogmatism Scores. Zero-Order Correlations and Partial Correlations Existing Between the Independent Variable and the Dependent Variables . Zero-Order Correlations, Eta Coefficients, F of Eta. Age Distribution of the Respondents. Occupational Ranks Held by Respondents Geographic Mobility Scores . Number of Hours of Exposure to the Mass Media in An Average Week . Number of Mass Media Communication Channels Seen and/or Heard in an Average Week . Number of Mustang Commercials Recalled During The Week Preceding the Interviews. Adoption Dates for 150 Subjects. Number and Frequency of Conversations about The Mustang During the Week Preceding the Interviews . . . . . . . . . . . . . . . viii PAGE 62 70 74 92 150 151 152 153 154 155 156 157 l6 17 18 19 Number of Conversations about the Mustang with People Living in Similar Neighborhoods, Having Similar Occupations, and Having Similar Educations as a Percent of Total Conversations . Number of Others Who Drove the Respondents' Mustangs . Number of Persons Allegedly Influenced by the Respondents. . . . . . . Number of Other Mustang Owners Known by the Respondents. ix 158 159 160 161 LIST OF FIGURES Figure l Interurbian Areas in the United States. 58 Appendix A Appendix B Appendix C Appendi x D Appendix E LIST OF APPENDICES Telephone Instruction Form. Interview Schedule Form . Letters of Introduction . Summary of Data Collection. Tabular Description of Sample . xi 122 124 144 147 149 Chapter I INTRODUCTION Some of the most serious problems confronting marketers in the post-war years are in the areas of new product innovation, diffusion, and acceptance. The severity of these problems has increased during the 1960's because the affluence and mobility which characterize today's consumers have produced significant changes in their needs and wants and these in turn have brought about significant changes in buyer behav- ior. The problems have been further complicated by aggressive competi- tors who are continuously striving to obsolete one another's product offerings. In order to survive, firms must innovate. Surprisingly little research has been published about this topic where success frequently brings profits beyond expectations and failure often brings a multiplicity of financial woes and non-financial embar— rassment. The bulk of the research which has been published is in the field of rural sociology and, to a lesser extent, anthropology, education, general sociology, and medical sociology. The majority of the studies conducted and reported in these fields focused on the diffusion and adoption of innovations. Diffusion and Adoption of Innovations An investigation of the diffusion and adoption of an innovation involves (1) the new.iggg (2) which is communicated through various channels, (3) among the individuals comprising a social system, (4) over a period of time. Time is the crucial element (1) in the rate f 2, adoption of the new idea in the social system, and (2) in the adogtion period through which a person moves from an initial awareness of the innovation, to a positive attitude of its usefulness, to its adoption, and finally to subsequent purchase.1 Time also is involved in the igggr vativeness dimension, which is a measure of when a person, relative to the other members in the social system, adopts the innovation. .323 Communication Process The diffusion and adaption process just described above is directly related to the four conceptually distinct elements in the communication process: source, message, channel, and receiver.2 Each of the four variables and the relationships existing among them can similarly be identified in the diffusion and adoption of new products. In this process, the source is a firm which has invented and/or sponsored the new product or service. In order to capitalize on the innovation, the firm will purposely.attempt to influence the receivers (consumers) to purchase its product. After the innovation has been perceived, it is possible for the original receiver(s), in the role of an opinion leader, to act as a source conveying additional information about the product to other receivers. Messages refer to all statements made about the innovation regardless of source. Channels are the ‘means or vehicles used to convey the messages to the 1Everett M. Rogers, Diffusion'gf'Innovgtions, Free Press of Glencoe, New York, 1962. 2Wilbur Schramm, "How Communication Works," in the Process and Effects gf_Mass Communication, (Wilbur Schramm, ed.) University of Illinois Press, 1954, pp. 3-26. 3 intended receivers. They are either mediated or interpersonal. The former are exemplified by newspapers, magazines, radio, or television which may carry advertising messages, whereas the latter refer to individu- alized attempts at persuasion made either by the firm's personal repre- sentative or other individuals communicating directly with one another. Finally, the receiver(s) is the potential consumer who either is aware or unaware of the innovation or who has or has not adopted it. Re- ceivers in certain instances may initiate the communication process by actively seeking information about new products. Because the analysis of the diffusion and adoption process is intimately related to the dynamics of communication behavior, an investigation of the diffusion and adoption process demands an inter- disciplinary approach involving concepts from social-psychology, rural sociology, marketing, and other behavioral disciplines. Consideration should also be given to the characteristics of the innovation and to the personal, social, and enabling conditions affecting its rate of adoption.3 Characteristiggwgf the Inngggtion An innovation is defined as an idea which is perceived as new by the individual. The idea can include either a product or a ser- vice. "It is important to remember that the distinctive aspect of an innovation, as compared to other kinds of ideas, is that it is con— sidered new by the individual. He lacks experience with the idea."4 3Thomas A. Staudt and Donald A. Taylor, A Managerial Introduction £2.Marketi , Prentice-Hall, Englewood Cliffs, New Jersey, 1965, ch. 4. 4Everett M. Rogers and J. David Stanfield, "Adoption and Dif— fusion of New Products: Emerging Generalizagions and Hypotheses," paper presented at the Conference on the Application of Sciences to Marketing Management, Purdue University, July 12-15, 1966. 4 Several diffusion researchers have considered the characteris- tics of a new product and how these affect its rate of adoption. Table 1 presents a recapitulation of 206 studies which related the characteristics of an innovation with its rate of adoption. It is readily seen that those innovations which are perceived as having relative advantage over existing ideas, those which are compa- tible with existing behavior, those which are visible, those which are readily available, and finally those which satisfy the consumers' needs, tend to have a relatively rapid rate of adoption. Other characteris- tics such as purchase cycles and trade-in allowances which are ulti- mately included in the individual's perception of the innovation have not as yet been investigated. Innovations can be categorized in many different ways.5 For example, an innovation can take the form of a new generic product or service. As far as the consumer is concerned, this form of an innova- tion represents a totally new idea. In such a situation, the marketer must communicate the idea to the potential consumer. Recent examples of new generic products include the television receiver and the room air conditioner. A second form of consumer innovation is development of a new brand and results from competitive interaction. It occurs only in those instances where the generic innovation is perceived to have attained a certain degree of market acceptance. The consumer dependent SA fourteen item delineation was presented by Wasson. Chester R. Wesson, "What is 'New' About a New Product?", Journal g§_Marketin , Vol. 25, July, 1960, pp. 52-56. According to Robertson, "Innovations may be classified as (1) continuous innovations, (2) dynamically con- tinuous innovations and (3) discontinuous innovations." Thomas S. Robertson, "The Process of Innovation and the Diffusion of Innovation," Journal 2; Marketin , Vol. 31, January, 1967, p. 15. Table l «Perceived Characteristics of the Innovation Related to Its Rate of Adoption** Percentage Relationship_to Innovativeness Perceived charac- Condi- Total no. teristics on the Positive None Negative tiona1* Total of publi- innovation (X) (Z)_ 4(2) (Z) (Z) cations Relative advantage 78.8 15.2 3.0 3.0 100 66 Compatibility 86.0 14.0 0.0 0.0 100 50 Fulfillment of felt needs 92.6 3.7 3.7 0.0 100 27 Complexity 18.8 37.5 43.7 0.0 100 16 Divisibility 42.9 42.9 14.3 0.0 100 14 Communicability 75.0 25.0 0.0 0.0 100 8 Availability 55.6 22.2 16.7 5.6 100 18 Immediacy of benefit 57.1 28.6 14.3 0.0 100 7 *A relationship that may be positive or negative depending upon (or conditioned by) other variables. **Data reported in Tables 1, 2, and 3 were obtained from the Dif- fusion Documents Center at Michigan State University. tains more than 1,200 publications pertaining to the communication of .ggg_ideas among members of a social system over time. study catalogued in the Center has been content analyzed and informa- tion pertaining to both the independent variables and the dependent variables and the relationships between them has been placed on IBM cards. Source: East Lansing, Michigan, July, 1966. The Center con- Each empirical Using IBM sorting procedures, a summary bibliography of rele- vant publications is readily obtained. Diffusion Documents Center, Michigan State University, 6 upon his familiarity with existing market offerings may or may not per- ceive the new brand as a totally new idea but instead may consider it as a substitute for the previously introduced generic product. Exam- ples of brand innovations would be the several brands of fluoridated tooth paste introduced subsequent to the introduction of Procter and Gamble's Crest. CharacteristicsIg£,Ad02ters In addition to considering the characteristics of the innova- tion itself, diffusion researchers investigated the characteristics of the individuals who have adopted and rejected new ideas and products. More than 835 published empirical studies investigated the sociological and psychological attributes of these persons.6 Table 2 presents a summary of the social characteristics of adopters related to innovativeness. As can be seen from Table 2, several socio-economic factors are consistently related to innovative- ness. Several of the negative relationships can in part be explained because the particular innovations which were studied were designed to penetrate a loweincome market or perhaps one primarily composed of senior citizens. More than 300 studies focused on the socio:p§ychological characteristics of the adopter.7 These research findings are summar- ized in Table 3. With regard to the relationships enumerated in 6Diffusion Documents Center, Michigan State University, East Lansing, Michigan, July, 1966, and Everett M. Rogers, Bibliography gf Research gguthg Diffusion‘gf Innovgtions, Department of Communication, Michigan State University, East Lansing, Michigan, July, 1966. 7Everett M. Rogers and J. David Stanfield, 92. cit. Table 2 - ~Social Characteristics of Adopters Related to Innovativeness Percentage Relationship_to Innovativeness Social charac- Condi- Total no. teristics of the Positive None Negative tional* Total of publi-; unit of adoption (1) (Z) (Z) (X) (Z) cations Education 74.6 16.1 5.2 4.1 100 193 Literacy 70.4 22.2 3.7 3.7 100 27 Income 80.3 10.7 6.3 2.7 100 112 Level of Living 82.5 10.0 2.5 5.0 100 40 Age 32.3 40.5 17.7 9.5 100 158 *A relationship that may be positive or negative depending upon (or conditioned by) other variables. Source: Diffusion Documents Center, Michigan State University, East Lansing, Michigan, July, 1966. Table 3 é'Social-Psychological Variables Related to Innovativeness Percentage Relationship to Innovativeness Social-psychological Condi- Total no. characteristics of Positive None Negative tiona1* Total of publi- the unit of adgption (Z) (Z) (Z) (2) (Z) cations Intelligence Knowledgeability 78.8 16.7 ‘ 1.5 3.0 100 66 Attitude toward change 73.6 14.5 8.2 3.8 100 159 Achievement Motivation 64.7 23.5 0.0 11-8 100 17 Aspirations for children 82.6 8.7 4.3 4.3 100 723 Business orientation 60.0 20.0 20.0 0.0 100 5 Satisfaction with life 28.6 28.6 42.8 0.0 100 7 Empathy 75.0 0.0 25.0 0.0 100 4 Mental rigidity 20.8 25.0 50.0 4.2 100 24 .L *A relationship that may be positive or negative depending upon (or conditioned by) other variables. Source: Diffusion Documents Center, Michigan State University, East Lansing, Michigan, July, 1966. 9 Table 3, Rogers implied that socio-psychological attributes are some- what difficult to measure and conclusions drawn from them may be tenuous. Personality Variables in Diffusion Research Personality variables have been relatively neglected by diffusion researchers. Both Lionberger and Rogers recognized the need for and deplored the lack of empirical research directed toward personality variables related to diffusion.9 Because of the paucity of past research, one of the key questions according to Barnett is: Just which individuals in a given group are more likely than others to accept a novelty? The problem now is to find out, not why a novelty or its auspices has an appeal, but why it appeals more to one person than to another, presuming that everything else is as constant as it can be. The present study, therefore, assesses the utility a person- ality variable in explaining communication behavior by focusing on the diffusion and adOption of a consumer innovation, the Mustang automo- bile. The Innovation Studied: The Mustang The Ford Mustang is the generic automotive innovation of the decade. The automobile comes complete with a long leping hood, short 8Everett M. Rogers, Diffusion g; Innovations, _p. cit., pp. 177- 178. Iowa State University Press, Ames, 1960 and Everett M. Rogers, Diffu- sion‘gf Innovations, gp. cit. 10H. G. Barnett, Innovation, McGraw-Hill Book Company, New York, 1953, p. 378. 10 rear deck, bucket seats, and a fast, sporty appearance. Its low price tag (under $2,400) opens the door to a large segment of the market. The car additionally offers a wide range of extra cost options which enable the prospective buyer to personalize the car to his own satis- faction. The Mustang's success was probably attributable to several factors. First, credit must be given to the market researchers of the Ford Motor Company, who in this instance accurately diagnosed the consumers' transportation needs and wants. Second, the innovation possessed several characteristics which favorably effect adoption. More specifically, the Mustang had a relative price advantage over its nearest competitors. The hardtOp and convertible body styles, while definitely distinctive, were compatible with accepted automo- tive design. Finally, the Ford dealers throughout the country aggres- sively merchandised the innovation. The rapid acceptance of the Mus- tang prompted similar brand innovations by the Chrysler Corporation and American Motors. A year and a half later, General Motors entered the field, and the Ford Motor Company's Lincoln-Mercury Division also introduced a model similar to the restyled 1967 Mustang.‘ This particular innovation was selected for three reasons. c First, an analysis of this product would enable a comparison to be made between early and late adopters. Second, because of the magni- tude of the purchase, it was expected that people would be inclined to remember those things which influenced their purchase decision, and third, a representative sample of Mustang owners was readily obtainable. 11 Description of the Present Study The present study proposes the thesis that the socio- psychological theory of belief systems as formulated by Rokeach11 will generate new insights into some of the communication behavior involved in the diffusion and adoption of a new generic product. It investi— gates six conceptually distinct areas of communication behavior: (1) communication sources, (2) communication channels, (3) innovative- ness, (4) product loyalty, (5) opinion leadership, and (6) homophily. The basic objective is to identify and measure the relationship exist- ing between the personality variable of dogmatism and the six areas of communication behavior. A review of the literature combined with an assessment of the theoretical position of belief-disbelief systems, both of which are considered in Chapter 11, led to the formulation of six general hypo- theses which are considered in the present study. The hypotheses are as follows: GH 1: HIGH DOGMATICS UTILIZE FEWER COMMUNICATION SOURCES THAN LOW DOGMATICS. GH 2: LOW DOGMATICS EXPOSE THEMSELVES MORE TO MASS MEDIA COMMUNI- CATION CHANNELS THAN HIGH DOGMATICS. GH 3: LOW DOGMATICS ARE MORE INNOVATIVE THAN HIGH DOGMATICS. GH 4: HIGH DOGMATICS EXHIBIT GREATER LOYALTY TOWARD PRODUCTS THAN LOW DOGMATICS. GH 5: LOW DOGMATICS EXHIBIT MORE OPINION LEADERSHIP THAN HIGH DOGMATICS. GH 6: HIGH DOGMATICS EXHIBIT GREATER HOMOPHILY THAN LOW DOGMATICS. 11Milton Rokeach, The Open and Closed Mind, Basic Books, New York, 1960. 12 The general methodological approach used in the investigation is survey research.12 The data were collected by trained undergraduate student interviewers using a pre-coded and pre-tested interview schedule. The principal analytical technique is correlation.13 The obtained correlations indicate the extent to which the communication behavior of relatively high dogmatic individuals differed from that of low dogmatic individuals for each of the six communication concepts being investigated and whether the hypothesized relationships are confirmed or rejected. Implications It is expected that the present research will make a contribu- tion to the existing although limited fund of knowledge about the diffusion of consumer innovations. It is expected further that the interdisciplinary orientation cultivated in the present investigation will provide further impetus for a merger between the behavioral disci- plines and the field of marketing. Finally, the study anticipates the generation of practical conclusions concerning which individuals within a given market are most likely to (l) spend more time with the mass media, (2) be among the first to adopt an innovation, (3) be most loyal toward a product, and (4) be most effective as Opinion leaders. It is expected that the results can be directly beneficial for marketing practitioners in pre-testing the market acceptance of new products and 12Charles H. Backstrom and Gerald D. Hursh, Survey Research, University of Illinois Press, Chicago, 111., 1962. 13Quinn P. McNemar, Psychological Statistics, third edition, John Wiley and Sons, Inc., New York, 1963. l3 indirectly for developing promotional strategies after the products have attained initial acceptance. Limitations The present study contains two factors which may affecttthe validity of the findings. The first is concerned with the nature of the innovation being studied. An automobile is a high-priced consumer product and in most instances new car buyers keep the car from two to three years. As a result certain individuals who may have been favorably disposed toward the car when it first came on the market postponed their purchases because their present automobile was relatively new. The second factor pertains to the study's methodological approach. More specifically, the answers to several questions are based upon the res- pondents' ability to recall and accurately relate their past experi- ences. The rather modest sample size (150 subjects) coupled with.the circumstances surrounding the collection of the data may influence the reliability of the findings. Finally, generalization from the present findings to other innovations may be somewhat limited because the analysis was concerned only with the diffusion and adoption of one product. To date most of the evidence suggests that innovativeness, product loyalty, and opinion leadership vary in their relationship to both innovations and subjects. Objectives In summary, the present study proposes the thesis that the socio—psychological theory of beliefs will generate new insights into some of the communication behavior involved in the diffusion and adoption of~a new generic product. Six specific concepts are 14 investigated: communication sources communication channels innovativeness product loyalty . opinion leadership . homophily O‘kflJ-‘LQNH O The present study asks whether any or all of these variables are ssignificantly correlated with dogmatism. Thus, the main objectives (of the present research is to identify and measure the predictive power of the personality variable of dogmatism in the diffusion and adoption of a consumer innovation. Chapter II CONCEPTUAL FRAMEWORK This chapter elucidates the theoretic approach espoused in the present study. It further presents definitions of the concepts being studied, the hypothetical relationships, and a review of research which considered the separate concepts and/or the relationships exist- ing among them. Theory The majority of the more than seven hundred empirical studies catalogued to date in the diffusion field were directed toward the solution of practical problems; less attention was given to theoretic concerns than was the case in many other research traditions. As a result, it is sometimes very difficult to offer suitable explanations for the multitude of relationships uncovered by diffusion researchers. The inclusion of a theoretical position in the present investigation, therefore, should aid in explaining the hypothesized relationships and provide a measure of understanding of why and how these relationships occurred. The theoretic approach used in the present research is the theory of belief systems developed by Rokeach.1 Prompted by the work of Adorno and Frenkel-Brunswik on authoritarian personalities,2 Rokeach 1Milton Rokeach, The Open and Closed Mind, _23 cit., p.32. 2T. W. Adorno, Else Frenkel-Brunswik, Daniel J. Levinson, and R. Nevitt Sanford, The Authoritarian Personalit , Harper and Brothers, New York, 1950. 16 conceptualized a personality theory which ranged over a complete spectrum of beliefs incorporated in a person's cognitive system. An individual's belief was defined as follows: We have to infer what a person really believes from all the things he says and does. It is in this sense that we would use the term belief, and the total belief-disbelief system would then be an organization of verbal and nan-verbal, implicit and explicit beliefs, sets, or expectancies. Through the coneption of a belief-disbelief system, Rokeach uncovered a single set of concepts which would serve as a basis for understanding and explaining the interconnected problems of personality and cogni- tion.4 In effect, knowledge about the organization of an individual's beliefs and expectancies should enable predictions about the indivi- dual's behavior. Since emphasis is placed upon the structure of the beliefs rather than specific content, the theory provides a high degree of parsimony insofar as predictions can be made about the totality of the person's behavior. According to the theory, all individuals possess belief- disbelief systems. Those having relatively Open systems are referred to as open-minded or low dogmatic individuals; whereas, those having closed systems are defined as high dogmatic or closed-minded. The l extent to which a system is open or closed is determined by the struc- tural arrangement of its component parts. This structure measures 3 beliefs which range along a continuum from central beliefs, concerning 3Milton Rokeach, The Open and Closed Mindl gp. cit., p. 32. 4Milton Rokeach, The Qpen and Closed Mindelgp. cit., Chs. 1-3. the nature beliefs der with common structural beliefs the hclistic an Idea introduced assimilate: coupatibil the inter: ties which then be if. a D85 inCludes 14 ideas. fl 1 matism 1, 17 the nature of self and reality, to inconsequential or peripheral beliefs derived from authority. These latter beliefs are concerned with common everyday experiences and personal preferences. "It is the structural interconnections among central, intermediate and peripheral beliefs that gives the total belief-disbelief system its integrated holistic and systematic character." Ideas and information concerning new products generally are introduced at the inconsequential level. But before new ideas are assimilated into the individual's system, they must first be tested for compatibility with the person's central beliefs and then screened at; the intermediate level. If the ideas are compatible with the authori- ties which undergird the intermediate beliefs, the new material will i then be incorporated into the belief system. The Concept of Dogmatism Dogmatism is defined as a personality variable which governs the individual's receptivity or lack.of receptivity to new ideas and further includes how a person perceives, evaluates, acts and reacts to such ideas. This definition closely parallels Rokeach's definition: Dog- matism is a personality variable which governs the individual's recep- tivity to ideas, people, and places and further includes the person's ability to evaluate information pertaining to each of these areas on 7 its own merit."‘ 5 . Milton Rokeach, The Open and Closed Mind, pp, cit., p. 50. 6Milton Rokeach, The Open and Closed Mind, pp, cit., p. 48. 7As per telephone conversation with Dr. Milton Rokeach, August, 1966. This definition differs from an earlier definition, where dog- matism was defined as "(a) a relatively closed cognitive organization / . 18 According to Rokeach, dogmatism is related to the structure of the person's belief system and to the manner in which the various beliefs are interrelated within the total belief system. This premise led Shoemaker to the conclusion that "the structure of the system is immensely more important than the content of the beliefs within that 8 If such is actually the case, it is logical to assume that system." those high in dogmatism would hold beliefs in different ways than those low in dogmatism. More specifically, a high dogmatic person, as described by Rokeach, has a relatively undifferentiated belief-disbelief system and tends to isolate different clusters of beliefs which may logically appear to interact. High dogmatic individuals, because of their overly reliant dependence upon authority, tend to view new ideas as threatening or hostile at least until they have been validated by an accepted authority. High dogmatic persons, provided they behave according to the tenets of the theory stated in the preceding paragraph, will have ,‘. different perspectives than low dogmatic individuals. Therefore, it is expected that closed-minded persons will react in different ways from open-minded persons when exposed to new products. For this reason, of beliefs and disbeliefs about reality, (b) organized around a central set of beliefs about absolute authority which in turn, (c) provides a framework for patterns of intolerance and qualified tolerance toward others.". See Milton Rokeach, "The Nature and Meaning of Dogmatism" in E. P. Hollander and Raymond G. Hunt, (eds.), Current Perspectives i3 Social Psychology, Oxford University Press, New York, 1963, p. 1963. 8F. Floyd Shoemaker, "Personality Dimensions of Innovativeness," 'unmublished,for Psychology 936, Michigan State University, East Lansing, Michigan, March 17, 1966. l9 dogmatism is used as the independent or predictor variable in the present investigation. Dependent Variables The dependent variables analyzed in the present study are (1) communication sources, (2) communication channels, (3) innovative- ness, (4) product loyalty, (5) opinion leadership, and (6) homoPhily. Each is treated in the following way. First, there is a definition and brief discussion of each variable. Then a general hypothesis stating the dependent variable's conceptual relationship to the inde- pendent variable is presented. Literature reporting past investiga- tions of these two concepts is then reviewed. Finally, a specific rationale supporting the hypothesized relationship is given. Communication Sources Communication sources are defined as individua1(s), acting either as representatives of a commercial organization or indepen- dently, who behave in such a way that the attitudes and/or behavior of others are affected. This definition is similar to Aristotle's who considered the source as the person who speaks.9 Two other defini- tions frequently used include Shannon and Weaver who referred to the "information source" as the individual who "selects a desired message 10 out of possible messages," and Berlo who described the source as 9W. Rhys Roberts, "Rhetorica" in The Works‘p§,Aristotle, (W. D. Ross, ed). Oxford University Press, 1946, Volume XI, p. 14. 10Claude Shannon and Warren Weaver, The Mathematical Theory.g§ Communication, University of Illinois Press, Urbana, Illinois, 1949, p. 70 20 "some person or group of persons with a purpose, a reason for engaging incommunication."11 Because several communication sources are encountered in the present study, e.g., corporate entities, dealer representatives, and present owners, the investigation recognizes a distinction between commercial sources and non-commercial sources. Prior to adopting a new product, an individual generally acquires various kinds of information before reaching a decision. Therefore, it is readily apparent that adoptive behavior involves decision mak- ing.12 In making an adoption decision an individual may mentally become involved in one or all of four functional areas as he pro- gresses from initial awareness to acquisition.13 The four areas are described as follows: 1. Knowledge function--individual learns of the idea or practice and begins to make instrumental communicative responses seek- ing how-to-doeit types of information. 2. Persuasion function--attitude formation or change in which 1 David K. Berlo, The Process g§_Communication, Holt, Rinehart, and Winston, New York, 1960, p. 30. 12This point of view supersedes the report of the North Central Rural Sociology Subcommittee which in 1955 standardized the adoption process as consisting of five sequential mental stages of thinking and acting. According to their definition, the stages were (1) awareness, (2) interest-information, (3) evaluation, (4) trial, and (5) adoption. See North Central Rural Sociology Subcommittee for the Study of Dif- fusion of Farm Practices, H y Farm People Accept New Idegg, Agricul— tural Extension Service Special Report, Ames, Iowa, 1955. 13F. Floyd Shoemaker, "A Reconceptualization of a Process," unpublished, COM 470, Michigan State University, East Lansing, Michi- gan, Spring, 1966. 21 individual assumes a favorable or unfavorable stance toward practice or idea. 3. Decision~making function--overt behavioral change in which person adopts or rejects new idea or practicew 4. Confirmation-~supporting function--individual makes consumma- tory communicative responses seeking support for his decision. In certain instances a single source of information could satisfy the information needs of both high and low dogmatics for each of the functional areas mentioned previously. This situation is most likely to occur when the innovation is relatively inexpensive, repre- sents only a small percentage-of the person's disposable income, or is perceived to be lacking in terms of either social or monetary rewards. Because of the relatively high cost of the innovation stu- died in the present investigation, it is reasonable to believe that the low dogmatic individual will have sought information from several sources as he moved from awareness to adoption, whereas the high dog- matic will have used fewer sources in making his decision. Therefore, it is hyothesized: GH 1: HIGH DOGMATICS UTILIZE FEWER COMMUNICATION SOURCES THAN LOW DOGMATICS. The utilization _§.g communication source is defined as expo- sure to and perception of any of several messages emanating from a single source over a period of time. With regard to the acquisition of information, Rokeach stated: A closed-minded person may expose himself to only one point of view in the press, selectively choose his friends and associ- ates solely or primarily on the basis of compatibility with systems, selectively avoid social contact with those who adhere to 22. different systems and ostraCize renegades.14 According to the theory the high dogmatic person has difficulty distinguishing between the source and the message. If a person were completely able to evaluate information on its own merits, he would seek information about a particular dis- belief system. But the more closed his system, the more sensitive should he be to communications, reinforcements, warnings, prOhibi- tions, and more should he be dependent upon such positive authori- ties for information he accumulates about a particular belief sub- system. Information about this system if received at all, shoul come second hand, spoon-fed, by the person's positive authority. This point of view was confirmed by Powell in his analysis of Lansing area residents. Based upon the results, he concluded that: "Open-minded individuals are better able to distinguish between and evaluate independently the content of a message and the source Of a "16 message than are closed-minded individuals. With regard to source credibility, he reported: In the case Of closed receivers, the credibility of the source may be a largely influential factor in determining whether or not the receiver's Opinion will be changed as a result of persuasive communication. Unless the message strongly contradicts the - existing Opinions of the receiver, a high credibility source may be more effective in changing the opinions of the receiver than a low credibility source. Because the high dogmatic needs a certain amount of information before adapting, and since he has difficulty in distinguishing between sources and messages, it would seem logical that he would restrict his 1l'iMilton Rokeach, The Open and Closed Mind, pp, cit., p. 49. 15Milton Rokeach, The Open and Closed Mind,.pp. cit., p. 61. 16Frederic. A. Powell, "Open and Closed-Mindedness and the Ability to Differentiate Source and Message," Journal p£_Abnorma1 ppd Social Psychology,il962, 65:61-64. 17Frederic A. Powell, ibid., p. 63. 23 attention to sources perceived as being the most credible. Not only would he use this source for information in developing a favorable attitude, but also he would have recourse to this same source in mak- ing his actual purchase decision. On the other hand, the Open-minded person, "capable of evaluating the source and the message indepen- "18 and less effected by source credibility may seek informa- dently, tion from a variety Of sources prior to adopting the new product. Only one diffusion study considered the relationship between dogmatism and information sources. Jamias, studying 147 Michigan farmers, tested the hypotheses that low dogmatic farm Operators would use extended group members more than relatively high dogmatic indivi- duals (l) for initial information and (2) for validating information.19 Neither of the hypotheses was empirically supported. Communication Channels Communication Channels are defined as vehicles carrying mes- sages tO the receivers or potential adopters. This definition closely parallels the message-vehicle definition described by Berlo.20 Traditionally, communication channels have been dichotomized as mediated or interpersonal, but several other distinctions such as cos-1 mopolite vs. localite and active vs. passive have appeared in the 18Frederic A. Powell, ibid., p. 63. 19Juan F. Jamias, The Effects pf Belief System Styles'pp Egg Communication gpg Adoption.p£ Farm Practices, Unpublished Ph.D. Thesis, Michigan State University, Department of Communication, East Lansing, Michigan, 1964. 20 David K. Berlo, pp” cit., pp. 63-70. 24 literature. In the present investigation, channels are categorized further into (1) mass media and (2) interpersonal communication channels. In addition to limiting their exposure to sources of information, it is believed that high dogmatics have less exposure to mass media communication channels. GH 2: LOW DOGMATICS EXPOSE THEMSELVES MORE TO MASS MEDIA COMMUNI- CATION CHANNELS THAN HIGH DOGMATICS. Rokeach, with regard to channel exposure, stated: People Often selectively avoid contact with stimuli, people, events, books, etc., that threaten the validity of their ideology or proselyte for competing ideologies. Cognitive narrowing may b manifested at both the institutional and noninstitutional levels. As a concomitant effect, the open-minded person according to the theory has a tendency to acquaint himself with distant disbelief subsystems as well as other belief systems. Because the low dogmatic is prone to learn new information from many sources, he is likely to expose himself to a variety of channels, whereas the high dogmatic restrained by the postulated tendencies for isolation, low differentiation, and cogni- tive narrowing would restrict his attention to those channels conveying 21Milton Rokeach, The Open and Closed Mind, pp..gip., p. 48. It is interesting to note how Rokeach's ideas Of cognitive narrowing are related to certain aspects of Festinger's theory of cognitive dis— sonance. According to Festinger, dissonance occurs when two incompa- tible items are associated with each other. TO illustrate the situa- tion, suppose that a man drinks and at the same time realizes that alcohol is injurious to health. This dissonance can be reduced in many ways e1g., ceasing to drink, changing beliefs about alcohol's effect on health, or attending to messages minimizing the associative bond linking alcohol and sickness. One might speculate that a highly dogmatic individual would have less tolerance for dissonance within his belief system than a more Open- minded person and therefore take more decisive action to shore up his beliefs and in the process narrow his cognitive field. See Roger Brown, "Models of Attitude Change" in Eugene Galanter, Roger Brown, Eckhard H. Hess, and George Mandler, New Directions.ip Psychology, Holt, Rinehart and Winston, New York, 1962. 25 messages which he perceives have been approved by his authority figures. Very few studies attempted to investigate the relationship between dogmatism and exposure to communication channels. Ehrlick, using a sample of sociology students, found a negative relationship between dogmatism and the amount of information the students learned in the classroom situation.22 The finding suggested that closed- minded individuals would tend to screen out channels conveying mes- sages which appeared to contradict present beliefs. In assessing the dissemination of horticultural information among residents in a Boston suburb, Troldahl found, contrary to his predictions, that Open-minded persons were relatively low in their exposure to gardening and household magazines, and were also less likely to read the home, food, and gardening sections of newspapers. In his discussion of the findings, Troldahl suggested the possibility that newspapers and other specialized mass media channels served to reinforce the beliefs which were presently held by the more dogmatic individuals. The result may also have occurred if the closed-minded subjects had-g priori considered these mediated (mass media communica- tion)_channels as an accepted authority for this type of information. Although not concerned with dogmatism, diffusion studies con- ducted by Marsh and Coleman (1955a), Fliegel (1956), Copp (1956), 22Howard J. Ehrlick, "Dogmatism and Learning," Journal pf Abnor- mal and Social Psychology, January, 1961, 69:148-149. 23Verling C. Troldahl, The Communication pf Horticulture Infor- mation apd Influence $3.5 Suburban Communit , Communication Research Center Report No. 10, Boston University, Boston, Mass., March, 1963. 26 Emery and Oeser (1958), and Coughenour (1960b) suggested that inno- vators and early adopters expose themselves to more communication channels than later adopters.24 Bell similarly found that the innovators for colored television sets, stereophonic equipment, dishwashers, and air conditioners had substantial exposure to communication channels.25 King, in his study of fashion adoption, also reported that early buyers had moderately higher exposure to mass communication channels, and higher exposure to information from outside the immediate social environment.26 Additional evidence of the positive association between time Of adoption and media exposure was presented by Gedalecia in a study involving all media.27 The findings reported by these researchers indicate that those people having the most exposure to the media are the most likely to 24C. Paul Marsh and A. Lee Coleman, "Differential Communication Among Farmers in a Kentucky County," Rural Sociology 20:93—101, Fred- eric C. Fliegel, "A Multiple Correlation Analysis of FactOrs Associ- ated with Adoption of Farm Practices," Rural Sociolggy, 21:284-292, ————;;_ ____ Experiment Station Technical Bulletin 83, 1956, F. E. Emery and O. A. Oeser, Information, Decision, gpg Action: ,A,Study,p§ the ngchological Determinants.pf Chapges.ip Farming Technigues, Cambridge University Press, New York, 1958, and C. Milton Coughenour, "The Functioning of Farmers' Characteristics in Relation to Contact with.Media and Prac- tice Adoption," Rural Sociolggy, 25:183-279. 25William E. Bell, "Consumer Innovators: A Unique Market for Newness," Stephen A. Greyser (ed.) Toward Scientific Marketiqg, Pro- ceedings of the Winter Conference of the American Marketing Association, Boston, Mass., December 27-28, 1963, p. 93. 26Charles W. King, "Communicating with the Innovator in the Fasion Adoption Process," Peter D. Bennett (ed.) Marketing and Eco- nomip.DevelOpment, Proceedings of the 1965 Fall Conference, American Marketing Association, Washington, September 1-3, 1965, pp. 433-434. 27Ben Gedalecia, "The Communicators: An All-Media Study," 3rd Annual Conference Report, Advertising Research Foundation, New York, November 14, 1957. 2-7 accept new ideas and adopt new produCts. Consequently, it is felt the low dogmatics, who according to the theory are more Open to new ideas, will expose themselves more to mass media channels than their high dogmatic counterparts. lnnovgtiveness In the present study, ipnpygtiveness is defined as a conceptual variable which indicates the time when a given person initially adopts a new product relative to other individuals in the same social systun. This variable has been considered as (l) the acceptance Of new ideas or practices,28 (2) a general behavioral disposition,29 (3) a kind of social action,30 and (4) a tendency to adopt new ideas and practices.31 Innovativeness has been of continuing interest to diffusion researchers. The reason for this unique interest is because this Variable is.directly concerned with the overt behavioral commitment of adoption or rejection of the new idea, which in effect is the culmin- ation of the individual's adoption process. GH 3: LOW DOGMATICS ARE MORE INNOVATIVE THAN HIGH DOGMATICS. A content analysis of past diffusion research indicated more than 28Everett M. Rogers, A_Conceptua1 Variable Analysis pf Technoe logical Change, Ph.D. Thesis, Iowa State University, 1959. 29J. H. Copp, Personal and Social Factors Associated With Adpp- Experiment Station, Bulletin No. 83, 1956. 30Frederick C. Fliegel, A Multiple Correlational Analysis pf Factors Associated With Adoption pf Farm Practices, Ph.D. Theais, uni- versity of Wisconsin, 1955. 31Paul J. Deutschmann and Orlando Fals Borda, Communication and Adoption Patterns pp Ap Andean Village, P.I.I.P., San Jose, Costa Rica, 1962. 28 2,486 research findings relating other (independent) variables to inno- vativeness.32 The findings of the studies relating social and social- psychological variables to innovativeness were presented, respectively, in Table 2 and Table 3. As these tables indicate, there are many dis- crepancies in the findings, and this is especially true with regard to the social correlates. The conflicting evidence most likely occur- red because the innovations being investigated were deliberately dir- ected toward specific market targets, i.e., young people or well- educated people or people with above average incomes. On the other hand, the analyses which included attitudinal concepts generally were concerned with the attributes of the individual, and, therefore, the findings are not necessarily so limited by the nature of the innova- tion. A review of several attitudinal diffusion studies indicated that innovative farm Operators possessed the following socio-psycho- logical characteristics:33 1. They possessed more technical knowledge of agriculture. 2. They tended to act more rationally (factually) in adopt- ing innovations. 3. They had a higher ability to deal with mental abstractions. 4. They tended to view their work as a means to an end rather than as an end in itself. Only five studies, however, directly dealt with the relation- ship between dogmatism and innovativeness. One such study 32Diffusion Documents Center, _p, cit. 33 177-178. Everett M. Rogers, D1£fipaipp_gfi_Innoyations, _p, cit., pp. 29 was conducted by Rogers and Harp.34 In their analysis of the person- ality characteristics of 23 Iowa farm operators, they found that the early adopters scored lower on the dogmatism scale than the less innovative farmers. "The correlation with adoption was minus .15, which is in the expected direction but not significant . . . the explo- ratory nature of this study and the small sample make it necessary that this finding be regarded as tentative."3S In assessing the communication of horticultural information among Boston suburbanites, Troldahl sought to determine the effect that channel exposure had on belief change.36 The evidence somewhat unexpectedly indicated that exposure induced belief changes toward the recommended horticultural practices among closed-minded individuals but not among the low dogmatics included in the sample. Although the findings appeared to contradict the espoused theoretic position, Trol- dahl suggested that the effects may have been in accord with another part of Rokeach's theory which implied that under certain conditions, i.e., non-rational decision making, closed-minded persons are more sus- ceptible to belief changes than less dogmatic individuals. Because Of the methodology which was utilized, it was not possible to deter- mine if this actually were the case. 34Everett M. Rogers and John Harp, "Personality Correlates of the Adoption of Technological Practices," Paper presented to the Mid- west Sociological Society, Des Moines, Iowa, April 5, 1967. 35Everett M. Rogers, "Personality Correlates of the Adoption of Technological Practices," Rural Sociology, 22:267-268. 36Verling C. Troldahl, The Communication pf Horticultural Infor- mation and Influence $3.3 Suburban Community, pp, cit. 30 Jamias, studying the adoptive behavior Of 147 Michigan dairy farmers, tested the hypothesis that highly dogmatic farm operators would be less innovative than their less dogmatic counterparts. General innovativeness scores and the subject's dogmatism scores were negatively related. The relationship was statistically significant. The remaining studies are in the education field. Childs investigated the relationship between.the belief systems of adminis- trators and teachers in innovative and non-innovative school districts. With regard to dogmatism and innovativeness, he found a negative rela- tionship and stated: "The data showed more than a slight degree of association between innovation and the number of individuals having Open belief systems."38 A more definitive statement was precluded because of (1) the size of the sample and (2) the methodology employed. Hudspeth analyzed the effects of dogmatism on attitudes toward 39 He found a signifi- mediated instruction among university faculty. cant correlation between open-mindedness and attitudes toward such innovations, but the data did not statistically support the hypo- thesis that current users Of mediated instruction were more open-minded than similar teachers not using these techniques. In a discussion of the findings, Hudspeth suggested that the presence of uncontrolled variables may have masked the true relationship. 37Juan F. Jamias, pp. cit., p. 78. 38John W. Childs, A Stud y Lf the Belief Systems Lf Adminis- tratOrs and Teachers in Innovative —and Non-Innovative School Dis- tricts, Unpublished Ph. D. Thesis, College of Education, Michigan State University, East Lansing, Michigan, 1965, p. 50. 39DeLayne R. Hudspeth, A Stud y Lf Belief Systems and Acceptance Lf New Educational Media With Userg and _Non-Users.p§ Audiovisual Graphics,Unpublished Ph. D. Thesis, College of Education, Michigan State University, East Lansing, Michigan, 1966. 31 The basic assumptions in Rokeach's theory suggested that low dogmatics utilized both more sources and more channels for Obtaining information than high dogmatics. Therefore, the open-minded person is more likely to be among the first to be aware of the innovation. This in itself suggests that he might be among the first to adopt. Also, the low dogmatic, being less dependent upon authority, may be more inclined to act on his own initiative. On the other hand, the closed- minded person, exposed to a narrower range of communications, would be more inclined to take a wait-and-see attitude and postpone his adoption until the external authority has made a commitment regarding the inno- vation. Product Loyalty Product Loyalty is the degree to which consumers repeatedly purchase a given product or service. Because the goal of most past diffusion research, such as that conducted by rural sociologists, was to Obtain adoption, very little attention was given to post-adoptive behavior. In addition to gaining initial adoption, the great majority of consumer innovations depend upon repeat purchase for their ultimate success. Obviously, the fre- quency Of purchase is dependent upon the product's life expectancy. For example, the average life of a box of breakfast cereal is about one week, whereas the average owner of a new automobile replaces it about once every three years. In either case, the marketer will seek to cultivate a core market composed of loyal customers who will habitually purchase the innovation, its replacement, or another sub- stitute product which is marketed by his company. 32 GH 4: HIGH DOGMATICS EXHIBIT GREATER LOYALTY TOWARD PRODUCTS THAN LOW DOGMATICS. An analysis Of the marketing literature provides an overview Of this important, but little-studied area of product loyalty. Pesse- mier explained that product loyalty, like any other aspect of buyer behavior, is composed of many interacting factors.40 More specifically, the product's price, promotional policies, and the availability of substitutes all have an effect on the individual's purchase decisions subsequent to the initial adoption. In an empirical study assessing loyalty to cake mixes, Draper and Nolin found brand switching to be more prevalent among new cus- tomers.41 Similarly Lipstein reported that individuals rarely changed from one core group of products to another.42 More Often, he found, consumers entered a transition stage where many substitute products were sampled. Some individuals, he observed, remain "switchers" for extended periodsof time, whereas, others may either become loyal to one substitute or return to the original brand. He additionally found that "switchers" were the most likely to try the newest brands. Cunningham reported that there was little carry-over of brand 4oEdgar A. Pessemier, "A New Way to Determine Buying Decisions," Journal p§.Marketipg, vol. 24, no. 2, October, 1959, pp. 41-46. 41Jane E. Draper and Larry H. Nolin, "A Markov Chain Analysis of Brand Preferences," Journal.pf Advertising Research, vol. 4, no. 3, September, 1964, pp. 33-39. ’ 42Benjamin Lipstein, "The Dynamics of Brand Loyalty and Brand Switching," Proceedings, The Annual Conference of the Advertising Research Foundation, New York, September, 1959, pp. 101-108. 33 loyalty across products. "Those who are highly loyal to a brand of one product may have very little loyalty to a brand of another pro- "43 duct. Tucker, on the other hand, perceived that some customers (the reason was not given) were more prone to loyalty than others.44 Rationale for the present hypothesis is an extension of the rationale which supported the hypothesis concerning innovativeness. Both high dogmatic and low dogmatic individuals are probably included in core markets; however, it would seem that there would be a pre- ponderance of the former. Open-minded persons according to belief system theory would continually expose themselves to new product ideas and not feel compelled to "stick" with a product or brand just because their peers or other individuals either recommend or use the product themselves. For him, Eghe Open-minded persog], the power of authority is still there, ut depends upon the authority's cognitive correct- ness, accuracy, and consistency with other information he has about the world. Authority that gives information in conflict with the information he possesses will be judged unreiiable and will there- fore be replaced by more reliable authority. 5 Highly dogmatic individuals, on the other hand, having once adopted a product are more likely to become loyal customers provided the authority they identify with has not recommended a change. "In the closed system, the power of authority does not depend on cognitive 43Ross M. Cunningham, "Brand Loyalty-What, Where, How Much?" Harvard Business Review, vol. 34, no. 1, January-February, 1956, pp. 116-128. 44W. T. Tucker, "The Development of Brand Loyalty," Journal p: flprketing_Resp§rch, vol. 1, no. 3, August, 1964, pp. 32-35. 4SMilton Rokeach, The Open and Closed Mind, pp, cit., p. 63. 34 correctness, but on the authority to mete out reward and punishment, Given a variety of information stemming from an external source, the relatively closed person is forced to accept or reject all in a 'pack- aged deal.'"46 The present hypothesis can be additionally supported with cer- tain aspects Of Festinger's theory of cognitive dissonance.47 Because the decision to purchase an automobile involves choice and compromise, the results according to Festinger may create post-decision dissonance in the mind of the buyer. Ehrlich‘pp‘pl reported that new car buyers restored consonance by reading advertisements of their own automobile more often than of other types of cars which they had or had not con- sidered as alternatives.48 Five laboratory studies considered the relationship existing between dogmatism and tolerance for cognitive inconsistency. Foulkes and Foulkes reported that high dogmatic persons have less tolerance for trait inconsistency than less dogmatic persons.49 Kleck and Wheaton found that high dogmatic individuals showed less tolerance in situations where they were exposed to opinion-consistent and Opinion-inconsistent 46MiltonRokeach, The Open and Closed Mind, pp. cit., pp. 62-63. 47Leon Festinger, é Theory pg Cognitive Dissonance, Row, Peter- son, & Co., Evanston, Illinois, 1957. 48D. Ehrlich, I. Guttman, P. Schonback, and J. Mills, "Post- decision Exposure to Relevant Information," Journal pg Abnormal and Social Psychology, vol. 54, 1957, pp. 98-102. 49D. Foulkes and S. H. Foulkes, "Self-Concept, Dogmatism, and Tolerance of Trait Inconsistency," Journal‘pfi Personality and Social Psychology, 2, 1965, pp. 104-111. i.- -- -. .__.—— ‘1forma denons1 repan theref incons resul: Show . posse atte: agair re? :5- f) I and \- vex Re: 35 50 information. Hunt and Miller reported that closed-minded persons demonstrated significant attitude change in the direction of the dis— crepant position when confronted with belief-discrepant messages and, therefore, suggest that closed-minded persons have less tolerance for 51 On the other hand, the inconsistency than Open-minded persons. results of studies conducted by Wrenn52 and by Fillenbaum53 did not show a relationship between dogmatism and tolerance for inconsistency. Provided there is a relationship between dogmatism and tolerance for inconsistency, it is reasonable to believe that high dogmatics who possess a narrower cognitive structure, will continue to selectively attend to Mustang advertisements. Therefore, they may be more inclined again to buy a Mustang. Opinion Leadership Prior to 1940, the mass media were considered to be a primary communication influence on human behavior. The Voting Study by 50R. E. Kleck and J. Wheaton, "Dogmatism and Responses to Opinion Consistent and Opinion Inconsistent Information," Journalipg Personality and Social Psycholo , 2, 1967, pp. 249-252. 51Martin F. Hunt, Jr. and Gerald R. Miller, "Open-.and Closed- Mindedness, Belief-Discrepant Communication Behavior, and Tolerance for Cognitive Inconsistency," Paper presented at the convention of The Speech AssoOiation of America, New York, 1965.- 52R. L. Wrenn, The Resolution pg Cognitive Dissonance 32 Open [pap Closed Belief S stems, unpublished Ph.D. dissertation, Ohio Uni- versity, Athens, Ohio, 1962. 53 S. Fillenbaum, "Dogmatism and Individual Differences in Reduction of Dissonance," Psychological Reports, 14, 1964, pp. 47-50. 36' Lazarsfeld, Berelson, and Gaudet, however, suggested that interper- sonal communication exerted a higher degree of influence on voting 54 Their seminal study prompted researchers from other dis- behavior. ciplines to investigate the role people played in the communication process.55 In each of these studies, the investigators found certain groups of people, commonly referred to as opinion leaders, who tended to monitor the messages emanating from the media; and frequently found that they "informally" exerted influence on other members of the social system. In the present study, ppinion leaders are defined as individuals who inform and/or influence the opinions and behaviors of other people through interpersonal contact. This is similar to Rogers' definition that opinion leaders are "those individuals in a social system who 56 consistently influence others in a desired direction." GH 5: LOW DOGMATICS EXHIBIT MORE OPINION LEADERSHIP THAN HIGH DOGMATICS. Although not concerned with dogmatism, several diffusion studies led Rogers to generalize that "earlier adopters have more opinion leadership than later adopters."57 Several marketing studies further 54Paul F. Lazarsfeld, Bernard Berelson, and Hazel Gaudet, The People:§ Choice, (second edition), Columbia University Press, New York, 1948. 5SElihu Katz, "The Two-Step Flow Of Communication: An Up-to— date Report on an Hypothesis," Public Opinion Quarter;y, spring, 1958, pp. 61-780 56Everett M. Rogers with F. Floyd Shoemaker, Diffusion pg Innovations: 'A_Cross-Cultural and Communication Approach, Free Press of Glencoe, New York, 1967, in process. 57Everett M. Rogers, Diffusion g; Innovations, pp, cit., p. 313. 37 revealed that early adopters frequently behave as Opinion leaders and tell others about their new acquisition. Bell stated that: Over 65 percent Of the innovators were asked for opinions about their products, almost half of the innovators were asked by friends and neighbors to see the innovistic product. Of the innovators who gave their opinions or demonstrated their product, 68 percent asserted that their questioning friends then purchased the innovation. K138 reported that early buyers had higher involvement in social commu- nication.59 Mueller found that more than fifty percent of the purchasers of new household appliances consulted with others who had previously purchased the innovation. Only one analysis directly considered the relationship between dogmatism and opinion leadership. Troldahl, investigating the diffusion of horticultural information, found that low dogmatics acted as opinion leaders 62 percent of the time, and high dogmatics served in this capa- city 46 percent Of the time.61 According to Rokeach's theory, the Open-minded individual is / more likely to be an Opinion leader for two reasons. In the first place, these individuals will have a greater fund of knowledge about the innovation, and second, they may express their Opinions on various 58William E. Bell, "Consumer Innovators; A Unique Market for Newness," _p, cit., p. 93. 59 Charles W. King, "Communicating With the Innovator in the FashflmlAdoption Process,".pp. cit., pp. 93-94. 60Eva Mueller, "The Desire for Innovations in Household Goods," Lincoln H. Clark (ed.) Consumer Behavior, Harper & Brothers, New York, 1958, pp. 13-37. 61Verling C. Troldahl, Mediated Communicgpion and Personal Influence: .A Field Experiment, Ph.D. Thesis, University of Minnesota, 1963, p. 220. 38 ideas with less fear of being reprimanded by external authority. It is also felt that fewer high dogmatic individuals would have adopted the innovation and, therefore, would have less Opportunity to pass along information than low dogmatic persons.6 Homophily The concept of homophily was introduced to the literature by Merton.63 For him the term summarized the phrase, "a tendency for friendship to form between those who are alike in some respect." He also deveIOped the notion of value homophily, which he defined as the observed tendencies toward correspondence in the values of friends. An investigation of the little-studied concept of homophily appears to be a logical extensiOn of the analysis Of opinion leader- ship. In addition to the identification of_Opinion leaders, a complete analysis of homophily would include an assessment of the relationship existing between the two interacting parties (Opinion giVer and Opinion receiver). Because none of the individuals who might be mentioned by the respondents are to be interviewed in the present investigation, a complete analysis of both parties in the dyad is not possible. Never- theless, some information about homophily or the characteristics mutually held by the Opinion leader and his opposite will be Obtained.* *In the present investigation the respondents are asked to state what percent of the people they talked with about the Mustang live in similar neighborhoods, have similar occupations, and have similar edu- cational backgrounds. 62Juan F. Jamias,.pp, cit. 63Paul F. Lazarsfeld and Robert K. Merton, "Friendship and Social Process: A Substantive and Methodological Analysis," Berger, Abel, and Page (eds.) Freedomlppg Control $p_Moderp Society, Octogon Books, Inc., New York, 1964, pp. 18-66. COS C0? thi pe t. 39 In the present research, homophily is defined as the degree of perceived similarity in selected characteristics between two indivi- duals who interact. This definition closely parallels Merton's defi- nition which was presented in the preceding paragraph. GH 6: HIGH DOGMATICS EXHIBIT GREATER HOMOPHILY THAN LOW DOGMATICS. Several studies cited by Rogerséaindicate that opinion leaders tend to differ from their followers with respect to channel exposure, cosmopoliteness, and innovativeness. But the findings are not entirely consistent. The previously cited Voting Study, for example, reported that opinion leaders were found in all social levels and that inter- personal communication most frequently occurred between people of similar age, occupation, and political Opinion.65 Additional evidence in support of a horizontal flow of informa- tion was presented in the Decatur Study, which indicated that in the areas of food product purchasing, movie-going, and fashions the leaders talked most of all to people like themselves.66 Warland, in his analy- sis of informal communication behavior among Iowa farmers, similarly found that informal communication about an innovation most frequently 64 237-247. Everett M. Rogers, Diffusion.p§ Innovptions, _p, cit., pp. 65Paul F. Lazarsfeld, Bernard Berelson, and Hazel Gaudet, The People's Choic , _p. cit. 66Public Affairs leadership was similarily analyzed. Accord- ingly, Katz and Lazarsfeld stated: "The typical public affairs leader, then, is quite different from Opinion leaders in the arenas of marketing and fashion . . . The flow of influence, too, seems to move more often from higher status to lower status peOple than vice versa." Elihu Katz and Paul Lazarsfeld, Personal Influence, [_p,.p;p,, part two and pp. 294-295. 40 occurred among individuals who possessed similar attitudes, similar levels of competence, and similar socio-economic status. Troldahl and Van Dam, analyzing face-to-face communication of major news topics, used the dyadic relationship as the unit of analy- sis.68 They found that the flow between co-workers was the most active interpersonal channel and that the two interacting parties pos- sessed similar educational backgrounds. Two socio-psychological studies indicated that individuals interact with people who are most like themselves. Festinger-_p._; sought to uncover friendship patterns in a newly' developed housing center.69 An analysis of the socio-metric data suggested a positive relationship between interpersonal communication and uniformity of attitudes. Precker studied the entire population of Bard College to deter- Hine Which peers students would select for friends and which faculty they would choose_for advisors.70 In both instances, he found that students selected individuals who had expressed values similar to their own. Albeit the social and demographic similarities between leaders 67Rex H. Warland, Personal Influence: The Degree p: Shmilarity .9; Those Who Interact, Unpublished M. S. Thesis, Iowa State University, Ames, 1963. 68Verling C. Troldahl and Robert Van Dam, "Face-to-face Communi- cation About Major Topics in the News," Public Opinion Quarterly, Vol. 29. Winter, 1955—1956, pp. 626-634. 9 Leon Festinger, Stanley Schacter, and Kurt Back, Social Pres- sures ip.lpformal Groups: .A Study p£_Human Factors 1p Housin , Harper and wa: New York, 1950. 70Joseph A. Precker, "Similarity Of Valuings as a Factor in Selection of Peers and Near Authority Figures," Journa1.p£_Abnorma1 and Social Psycholpgy, 1952, 47:406-414. 4L 71 concluded that the influential and followers were apparent, Katz possesses: l. The personification p: certain values (who one is) 2. Competence (what one knows) 3. Strategic social location (whom one knows) Evidence tending to support Katz's contention was presented by Lionberger. In his investigation of 279 farm Operators in a Missouri community, he found that opinion leaders were perceived as more compe- tent than those who accepted their advice.72 In a subsequent analysis, Lionberger analyzed the relationship between prestige and interpersonal communication.73 He reported that information-seekers generally tended to seek out more prestigious sources, especially in situations involv- ing Opinion leaders at local levels. In conclusion, he implied that technical proficiency overcame communication barriers which existed because of perceived differences in prestige. Chou, studying interpersonal communication in three Columbian villages, reported (1) that innovativeness, mass media exposure, and social status were relatively effective predictors of friendship dyadic interactions, and (2) that mass media exposure and social participation 71Elihu Katz, "Two Step Flow Of Communication: An Up-to-Date Report on an Hypothesis," ‘pp. cit. 72Herbert F. Lionberger, "Some Characteristics of Farm Operators Sought as Sources of Farm Innovation in a Missouri Community," Rural Sociolo , 1953, 18:327-338. 73Herbert F. Lionberger, "Community Prestige and the Choice Of Sources of Farm Information," Public Opinion Quarterly, vol. 1, Spring, 1959, pp. 110-118. 42 were determinants of homophily in information-seeking behavior.74 She reported further, however, that cosmopoliteness, age, and social status were not important determinants Of homophily in information-seeking dyads. This latter finding, she implied, may have occurred because of the study's design and also because of the unique characteristics of the social systems. No previous studies have investigated the relationship between dogmatism and homophily. According to Rokeach's theory, dogmatism governs the acceptance V 5 There- or rejection of peOple in the same way that it effects ideas.7 fore, several reasons explain why low dogmatics would exhibit less homophily than high dogmatics. First, Open-minded individuals would feel more "free" to communicate with all types of people. In other “words, they would not need the approval of an external authority to engage in conversation. Second, low dogmatics would evaluate new ideas independently of other ideas and thus have little reason to care about other people's Opinions. Third, open-minded persons would realize that most "really new" ideas come from peOple having diesimdlar backgrounds. In such situations, a reciprocal exchange of information may occur. ¢ Conversely, individuals high in dogmatism would be more prone to seek others similar to thanselves as a means of gaining support for ’ ”Teresa M. Chou, Homophily p; Intezgctipn Pgttems y; 2192113" sion pf Innovations ip Cplumpign.yillpgg§, unpublished M. A. thesis, Department of Communication, Michigan State University, East Lansing, Michigan, 1966. 75Milton Rokeach, The Open and Closed Mind, _p, cit. ch. 4. 43 their newly acquired beliefs about an innovation. This point of view is derived from Festinger's theory of cognitive dissonance, which was previously mentioned. Finally, the high dogmatics, whether influencers or influencees have a tendency for cognitive narrowing and thus would restrict conversations to people holding beliefs similar to their own. It is reasonable to expect, therefore, that high dogmatics will exhibit a higher degree of homophily than low dogmatics. Control Variables Past research suggests that there are certain variables which are expected to have an effect on the present hypotheses. Their effect should be controlled. In the present investigation, these variables are (1) sex, (2) age, (3) education, (4) occupation, (5) occupational mobility, (6) occupational prestige mobility, (7) geographical mobility, and (8) social status. Although these variables are expected to be related to communicative and adoptive behavior, they are not of theoretical interest in the present study. Chapter III METHODOLOGY Operationalization Empiricalygypotheses The six general hypotheses and their accompanying empirical hypotheses were enumerated as follows: GH 1: GH 2: GH 3: GH 4: HIGH DOGMATICS UTILIZE FEWER COMMUNICATION SOURCES THAN LOW DOGMATICS. EH la: HIGH DOGMATICS UTILIZE FEWER COMMUNICATION SOURCES THAN LOW DOGMATICS. LOW DOGMATICS EXPOSE THEMSELVES MORE TO MASS MEDIA COMMUNICATION CHANNELS THAN HIGH DOGMATICS. EH 2a: LOW DOGMATICS HAVE A GREATER NUMBER OF HOURS OF EXPOSURE TO MASS COMMUNICATION CHANNELS THAN HIGH DOGMATICS. EH 2b: LOW DOGMATICS EXPOSE THEMSELVES TO MORE MASS COMMUNICATION CHANNELS THAN HIGH DOGMATICS. EH 2c: THOSE LOW IN DOGMATISM SCORES EXPOSE THEMSELVES TO MORE MUSTANG COMMERCIALS THAN THOSE HIGH IN DOGMATISM SCORES. LOW DOGMATICS ARE MORE INNOVATIVE THAN HIGH DOGMATICS. EH 3a: LOW DOGMATICS ADOPT THE INNOVATION RELATIVELY EARLIER IN TIME THAN HIGH DOGMATICS. EH 3b: LOW DOGMATICS PERCEIVE THEMSELVES AS MORE INNOVA- TIVE IN BEING THE FIRST IN THEIR IMMEDIATE NEIGH- BORHOOD TO ADOPT AN INNOVATION THAN HIGH DOGMATICS. EH 3c: LOW DOGMATICS PERCEIVE THEMSELVES AS MORE INNOVA- TIVE IN BEING FIRST AMONG THEIR CIRCLE OF FRIENDS TO ADOPT AN INNOVATION THAN HIGH DOGMATICS. HIGH DOGMATICS EXHIBIT GREATER LOYALTY TOWARD PRODUCTS THAN LOW DOGMATICS. EH 4a: HIGH DOGMATICS TRADED-IN FORD PRODUCTS MORE OFTEN THAN LOW DOGMATICS. GH 5: GH 6: EH 4b: EH 4c: EH 4d: EH 4e: EH 4f: 45 OF THOSE OWNING A SECOND CAR, HIGH DOGMATICS OWN A SECOND FORD PRODUCT MORE OFTEN THAN LOW DOGMATICS. HIGH DOGMATICS REPLACE THEIR ORIGINAL MUSTANG WITH A NEW MUSTANG MORE OFTEN THAN LOW DOGMATICS. HIGH DOGMATICS REPLACE THEIR ORIGINAL MUSTANG WITH A SIMILAR TYPE OF AUTOMOBILE MORE OFTEN THAN LOW DOGMATICS. HIGH DOGMATICS STATE A PREFERENCE FOR REPLACING THEIR ORIGINAL MUSTANG WITH A NEW MUSTANG MORE OFTEN THAN LOW DOGMATICS. HIGH DOGMATICS STATE A PREFERENCE FOR REPLACING THEIR ORIGINAL MUSTANG WITH A SIMILAR TYPE OF AUTOMOBILE MORE OFTEN THAN LOW DOGMATICS. LOW DOGMATICS EXHIBIT MORE OPINION LEADERSHIP THAN HIGH DOGMATICS. EH Se: LOW DOGMATICS EXPRESS THEIR OPINIONS ABOUT THEIR MUSTANG MORE FREQUENTLY THAN HIGH DOGMATICS. EH 5b: LOW DOGMATICS INITIATE DISCUSSIONS ABOUT THEIR MUSTANG MORE OFTEN THAN HIGH DOGMATICS. EH 5c: LOW DOGMATICS ALLOW MORE PEOPLE TO TEST DRIVE THEIR MUSTANG THAN HIGH DOGMATICS. EH 5d: LOW DOGMATICS PERCEIVE THEY HAVE INFLUENCED MORE RECEIVERS THAN HIGH DOGMATICS. HIGH DOGMATICS EXHIBIT GREATER HOMOPHILY THAN LOW DOGMATICS. EH 6a: EH 6b: EH 6c: EH 6d: HIGH DOGMATICS INTERACT WITH A GREATER PERCENTAGE OF TPEOPLE WHO ARE PERCEIVED TO RESIDE IN SIMILAR NEIGHBORHOODS THAN LOW DOGMATICS. HIGH DOGMATICS INTERACT WITH A GREATER PERCENTAGE OF PEOPLE WHO ARE PERCEIVED TO HAVE SIMILAR OCCU- PATIONS THAN LOW DOGMATICSo HIGH DOGMATICS INTERACT WITH A GREATER PERCENTAGE OF PEOPLE WHO ARE PERCEIVED TO HAVE SIMILAR EDU- CATIONAL BACKGROUNDS THAN LOW DOGMATICS. HIGH DOGMATICS INTERACT WITH A GREATER PERCENTAGE OF PEOPLE WHO ARE PERCEIVED AS RELUCTANT TO INNO- VATE THAN LOW DOGMATICS. 46 EH 6e: HIGH DOGMATICS INTERACT WITH A GREATER PERCENTAGE OF PEOPLE WHO ARE PERCEIVED TO HAVE HAD A POSI- TIVE OPINION ABOUT THE MUSTANG THAN LOW DOGMATICS. EH 6f: HIGH DOGMATICS INTERACT WITH GREATER NUMBER OF PEOPLE WHO PREVIOUSLY ADOPTED THE MUSTANG THAN LOW DOGMATICS. Measures The independent variable in all six general hypotheses was dogmatism. It was defined as a personality variable which governs the individual's receptivity or lack of receptivity to new ideas and further includes how a person perceives, evaluates, acts, and reacts to such ideas. Dog- matism was measured by the respondent's answers to the twenty item dog- matism scale1 which indexed the scores from 20 to 140. This scale is a modification of the forty item scale constructed by Rokeach.2 The dependent variable in CH 1 was the utilization of a communi- cation source which was defined as exposure to and perception of any of several messages emanating from a given source over time. This vari- able was measured by the respondent's answers to the following questions: "Thinking back, could you tell me where or from whom you first became aware of the existence of the Mustang?" "Again thinking back, could you tell me where or from whom you first became interested in the Mustang? --Probe to determine similarity between the answer given here and the J/ ' and "Which source of infor- answer given to the preceding question,' mation do you believe was the most influential in your purchase deci- sion? -- Probe to determine similarity between the answer given here ' 1Verling C. Troldahl and Frederic A. Powell, "A Short-Form Dogmatism Scale For Use in Field Studies," Social Forces, vol 44, no. 2, December 1965. 2Milton Rokeach, The Open and Closed Mind, pp. cit., ch. 4. 47 and the answers given to the two preceding questions." The answers given to these three questions were recorded in one of three ways: (1) commercial source, e.g., advertising, contact with dealer and/or salesman; (2) non-commercial source, e.g., immediate family, other relatives, co-workers, friends and/or neighbors, publicity;nand (3) self, e.g., saw on street.* The sources which were designated by the subject were then compared. If three different sources were mentioned, a zero score was recorded. A one was given when the same source was mentioned on two occasions. A two was awarded if the same source was given as the answer to all three questions. The dependent variable in CH 2 was mass media communication channels which were defined as vehicles carrying messages to the receivers or potential adopters. The dependent variables for the accompanying empirical hypotheses were (a) hours of exposure to the mass media, (b) number of mass media communication channels seen or heard, and (c) number of Mustang commercials recalled. Hours Of exposure were ascertained by asking the respondent to state how many hours he was exposed to (1) newspapers, (2) magazines, (3) radio, and (4) television during an average week. One point was given for each hour of exposure. The number of mass media communication channels was measured by asking the subject to state the different (1) newspapers, (2) magazines, (3) radio stations, and (4) television channels seen or heard during the average week. One point was awarded for each channel mentioned. * ,In certain instances the subjects mentioned advertising which is a messpge and not a source. Nevertheless, the message's sponsor, i.e., the source, was readily understood. 48 Mustang commercials were defined as any mediated communication sponsored by the Ford MOtor Company and/or its franchised dealers for the purpose of promoting the Mustang automobile. These communications (messages) were either printed in newspapers or magazines or broadcast over radio or television. Advertising recall was determined by asking the respondent to state the number of (1) newspaper advertisements, (2) magazine advertisements, (3) radio commercials, and (4) television commercials seen or heard during the seven days preceding the date of the interview. One point was given for each commercial recalled. The independent variable in CH 3 was innovativeness which was defined as the time when a given person initially adopts a new product relative to other individuals in the same social system. The dependent variables for the three empirical hypotheses derived from GH 3 were (1) relative time of adoption, (2) perceiving oneself as first in the immediate neighborhood to adopt, and (3) perceiving oneself as first among circle of friends to adopt. The time of adoption was measured by the subject's answer to the question: "When did you buy your Mustang?" The responses were scored from 1 to 31 depending upon the number of months which had elapsed between the date of the interview and the date of purchase. Being first in the immediate neighborhood to adopt was opera- tionalized by the question: "Were you the first person in your immedi- ate neighborhood to own a Mustang?" A one was awarded if the person believed he was first, a two was given if he felt he was one of the first, and a zero was given if he was not among the first to adopt. Being first among one's circle of friends to adopt was measured 49 by asking: "Were you the first person among your circle of friends to own a Mustang?" The answer was recorded in the same way as the response given to the preceding question. The dependent variable in CH 4 was product loyalty which was defined as the degree to which consumers repeatedly purchase a given product or service. The dependent variables for the six empirical hypotheses derived from GH 4 were the make of car traded-in, the make of second car, the make and body type of the car replacing the Mus- tang, and make and body type of purchase intentions. Trade-ins were determined by the question: "Did you sell or trade in another car at or about the time you bought the Mustang?" If the answer was yes, "What make was it?" If the car was produced by the Ford Motor Company a zero was recorded; a one was given if the car had been produced by any other manufacturer. The make of a second car was ascertained by asking: "DO you own a second car?" If yes, "What kind is it?" A zero was recorded if the car had been made by the Ford Motor Company and a one was given if the car had been produced by another company. The make and body type of the replaced car were Operationalized by the following questions: "Have you replaced your original Mustang?" If yes, (a) "With what make of car have you replaced it?" A zero was given if the Mustang had been replaced by another Mustang; a one was given if the car had been built by the Ford Motor Company, and a two was awarded if the vehicle had been produced by any other manufacturer. And (b)'"Which body type did you buy?" A zero was recorded if the car had been replaced by a Mustang, Cougar, Camaro, Barracuda, Marlin, etc.; a one was given for all other makes. 50 Purchase intentions were scored by asking (a) "When you replace your Mustang, with what make of car will you replace it?" and (b) "Which body type would you buy?" These questions were scored in exactly the same way as those pertaining to make and type of replaced auto- mobile. The dependent variable in CH 5 was Opinion leadership which was defined as the degree to which individuals inform and/or influence the opinions and behavior of other people through interpersonal contact. The dependent variables for the derived empirical hypotheses were (1) number of times Opinions were expressed, (2) percentage of dis- cussions initiated, (3) number of peOple who have test driven, and (4) number of receivers influenced. The number of times opinions were expressed was measured by the question: "During the past week, how many times did you talk about your Mustang with the following people: immediate family; other relatives; peOple you work with; friends and/or neighbors; clerks, dealers, salesmen; all others?" One point was awarded for each con- versation recalled by the respondent. The percentage of discussions initiated was indexed by the question: "With regard to these conversations (you have had about the Mustang), about what percent of the time did you begin them?" Per- centage figures were recorded verbatim. The number of people who had test driven was operationalized by the following questions: (1) "Since you have owned your Mustang, have you allowed anyone other than the members of your family to drive it?" If yes, (2) "How many others have driven it?" One point was given for each different individual. 51 The number of receivers influenced was measured by asking: (1) "As far as you know, about how many of the people you talked with about the Mustang, who didn't own one at the time, have subsequently bought a Mustang?" One point was awarded for each person mentioned. And (2) "How many of these peOple, in your Opinion, were influenced in their decision either as a result of riding in or driving your Mustang or by talking with you about your car?" One point was given for each person recalled. The dependent variable in GE 6 was homophily which was defined as the degree of similarity in selected characteristics between two individuals who interact.3 The dependent variables for the accompany- ing empirical hypotheses were measured according to perceived (l) simi- lar neighborhood, (2) similar occupation, (3) similar educational back- ground, (4) reluctance toward innovativeness, (5) positive opinions about the innovation, and (6) previous adoption. Perceived similar neighborhood was determined by asking: "Excluding your family, what percent of the people you talked with 3In a 1958 article entitled "Relational Analysis: The Study of Social Organizations with Survey Methods," Coleman described four methods which had been utilized for measuring homophily. The first method, known as contextual analysis related characteristics of the respondent's social context to a characteristic possessed by the indi- vidual himself. The second method was referred to as boundaries of homogeneity and was concerned with the reference group which had the greatest saliency for the individual in the situation being analyzed. Pair analysis was the third method. It involved a sociometric analysis of A's choosing B; both parties in the dyad were interviewed. The final method known as partitioning into cliques also involved socio- metric mapping. Because of the methodology being used in the present study, none of the methods suggested by Coleman can be used in the present research. See James S. Coleman, "Relational Analysis: The Study of Social Organizations with Survey Methods," Human OrganLgption, vol. 17, no. 4. 52 about the Mustang live in neighborhoods similar to the one you live in?" The percentage answers given to this and other questions under GH 6 were recorded verbatim. Perceived similar occupation was operationalized by the question: "Again excluding your family, what percent of the people you talked with about the Mustang have similar occupations or do about the same kind of work as you do?" Perceived similar education was measured by the question: "Again excluding your family, what percent of the people you talked with about the Mustang spent about the same number of years in school that you did?" Reluctance to innovate was operationalized by the question: "About what percent of the people you talked with about the Mustang would you say are the kind of people who like to try new and different things?" Perceived positive Opinions were operationalized by the follow- ing question: "Since you have owned your Mustang, what percent of the people you talked with about the Mustang, had favorable opinions about the car before they talked with you?" Perceived adaption was measured in terms of the answer given to the question: "About how many of the people you talked with about the Mustang already Owned a Mustang?" One point was given for each person mentioned by the respondent. The control variables in the present study were (1) sex, (2) age, (3) education, (4) occupation, (5) occupational mobility, (6) occupational prestige mobility, (7) geographical mobility, and (8) social status. 53 Sex was indexed as male and female. Age was measured indirectly by asking the subject to state the year in which he graduated from high school or completed his formal education. The responses were recorded in one of fifteen categories which ranged from 00 (1966 or after: 18 or younger) to 14 (before 1901: 84 or older). Each of the thirteen categories between the Open-ended extremes were composed of mutually exclusive five-year time periods, e.g., 04 (1946-1950:. 34-38), 05 (1941-1945: 39-43). Occupation was determined by asking the respondent to state the type of work that he (spouse or parent) does. The answer was then scored on the basis of the North-Hatt occupational rating scale, which ranks various occupations along a continuum ranging from 20 to 100. Occupational prestige mobility was indexed by the number of points separating the respondent's two most diatant occupational ranks as recorded on the North-Hatt scale. If the subject held the same position for five years or longer, a zero score was given. Geographical mobility was indexed by awarding one point for each change of address and one additional point for each different city - ._ "The report of the initial study can be found in Cecil C. North and Paul Hatt, "Jobs and Occupations: A Popular Evaluation." Opinion News, September 1947, pp. 3-13. Since the initial scale rated such a small number of the usual occupations usually recorded, the present study utilizes a list which includes both the original occupations and the interpolations from it. In interpolating the following priority was used: The original North-Hatt values; the Ohio State University interpolations by Leslie Silverman, W. Roy Cook, and A. O. Haller; and the Pennsylvania State University interpolations by Roy Buck and C. H. Brown, and the additions made to the Penn State Scores in 1960 by George Lowe. 54 in which the subject resided during the past five years. Social status was a function of (1) education, (2) occupation, and (3) address; points were awarded for each of these components. Points for education ranged from one to nine and were determined by the number of years in school. Similarly, scale points for occupation ranged from one to nine, and were determined on the basis of the North-Hatt scale. For example, scores ranging from 65 to 74 were given a 7, while those ranging from 85 to 94 were given a 9. Address also ranged along a nine point continuum and was based upon the interviewer's rating of both theidwelling and the surrounding neighborhood; (suggested guide- lines for address ratings were included on the interview schedule). A summation of the education, occupation, and address scores yielded the total sOcial status score which ranged from a possible low of 3 to a high of 27. Instrument Development Pretest A pretest Of twenty Mustang owners residing in the greater, Lansing, Michigan, area was conducted in August 1966. The names of registered Mustang owners compiled by the Michigan Automobile Dealers Association were provided by a Lansing automobile dealer. Ingham County registration lists from the months of March 1965, April 1965, January 1966, and April 1966 were used in the pretest. At the outset, thirty-five people were to be interviewed. This number was reduced to twenty for two reasons. First of all, definite patterns seemed to emerge after the first ten interviews. Second, many of the subjects could not be located for they did not subscribe to telephone services, had moved to another city, or were on vacation. 55 Three people would not participate in the study. The representative- ness of the sample admittedly may be low because registered owners who resided in or near a section of the city where a race riot had recently occurred were deliberately excluded. The pretest was conducted in the following way. The person whose name appeared on the registration list was contacted by telephone (see Appendix A for the Telephone Instruction Form). Once the principal user of the vehicle was identified, he or she was asked to participate in the study and a time and place for the personal interview was established. Each of the twenty subjects was subsequent- ly interviewed at their place of residence. Each respondent was asked about seventy-five questions concern- ing (1) sources of information, (2) media attendance, (3) advertising recall, (4) trade-in and purchase intention information, and (5) information concerning conversations they had had about their automo- bile. Four versions of a pre-coded interview schedule form were used in the pretest. Each respondent additionally completed a self- administered short-form dogmatism test.5 The average length of time for the interview approximated forty minutes. Interview Schedule Form As a result of the pretest, the most suitable of the four interview schedule forms was selected and subsequently reduced to a maximum of fifty-three questions. This reduction was made possible 5Verling C. Troldahl and Frederic A. Powell, "A Short-Form Dogmatism Scale For Use in Field Studies," op. cit. 56 by the elimination of certain questions and the rephrasing of others. Other modifications were occasioned by the differences in setting and by the fact that the Toledo newspapers went on strike three weeks prior to the scheduled data collection period6 (see Appendix B for a copy of the Interview Schedule Form). This pre—ceded schedule and the self-administered dogmatism test were usually administered in about twenty-five minutes. Setting The empirical hypotheses were tested with data obtained from 150 Mustang owners currently residing in Lucas and Wood Counties in Ohio. This particular setting was selected for two reasons. In the first place, time and cost considerations limited the analysis to this particular locality. The second reason was that these two adjacent 2 counties are located in the interurbian strip which extends from Toledo on the north to Cincinnati on the south. 8 The development of interurbias,7 megalopolises, and/or 6The Toledo newspapers were on strike from October 24, 1966 through March 27, 1967. Because of this unanticipated event, it is reasonable to believe that recall figures for the other mass media channels were inflated as people spent more time with these media dur- ing the strike period. Thomas Klein, "The Effect of a Newspaper Strike on Retail Sales and Advertising," Business Research Center, University of Toledo, Toledo, Ohio, in process. 7Interurbia: The Changing Face of America, Memo of the J. Walter Thompson Company, reprinted in Lazer and Kelly (eds.) Managerial Marketing: Perspgctives and Viewpoippg, Richard D. Irwin, Inc., Homewood, Illinois, 1962, p. 92. 8Jean Gottman, MegalOpolis: the Urbpnized Northeastern Seaboard pg the United States, The Twentieth Century Fund, 1961. 57 strip-citiesghas received considerable attention from marketers, urban economists, and urban sociologists. By definition, interubia com- prises... Two or more adjacent metropolitan areas with either two cities of 100,000 or more, or one city of 100,000 and three cities of 25,000 or more, plus adjacent counties with less than 25 percent farm population and more than 100 people per square mile. In effect, it is a combination of standard metropolitan areas which have merged with adjacent areas, thus becoming a new type of urban region.11 As it transcends county and state boundaries, interurbia can be considered as a new market phenomenon. Today, as illustrated in Figure 1, there are fourteen inter- urbias in the United States. Nearly fifty percent of the nation's pOpulation reside in these regions, and more than fifty percent of the country's retail sales are made within its boundaries. By 1975, it is estimated that these metropolitan complexes will include more than sixty percent of the population and account for seventy percent of total retail sales.12 9"Cities as Long as Highways-That's America of the Future," U.S. News.§'World Rgport, 25:31, April 5, 1957. 101nterurbia: The Changing Face of America, O . cit., p. 92. 118. George Walters, Morris L. Sweet, and Max D. Snider, "When Industry Moves to Interurbia," Sales.Management, February 20, 1959, p. 65. 12"Sprawling "Strip-Cities" - They're All Over U.S.," 58 monoum wouHaD ozu afi mound nowouououcH H ouowam 59 Interurbia, however, is far more than an ecological expansion. Several articles indicated that individuals residing in these areas tended to possess social and psychological characteristics different from those who have not participated in the post-war migratory move- ments.13 Some of the more relevant social attributes of the inter- urbian residents included the following:14 1. They are predominantly between the ages of 25 and 40. 2. They are predominantly members of the middle class. 3. They are primarily employed in salaried positions. 4. They are highly mobile geographically. The residents also exhibit psychological characteristics such as the following:15 1. They have a greater expressed desire for achievement. 2. They have a greater need to influence, lead, and dominate others. 3. They have a greater desire to be noticed. 4. They have a greater tolerance for change. 13Interurbia: The Changing Face of America,:ppy cit., p. 98., Everett M. Rogers, Social Change lg Rupp; Society, Appleton, Century, and Crofts, New York, 1960, and John H. Holmes, "Marketing in MegalOpolis," term paper submitted in Marketing 857, Department of Marketing and Transportation Administration, Michigan State University, East Lansing, Michigan, Spring, 1963. 1"Everett M. Rogers, Social Change 1p Rural Society, opL cit., pp. 157-159, and "MegalOpolis: Tomorrow's Society," Business Week, December 2, 1961, pp. 61-62. 15Everett M. Rogers, Social Change ip Rural Society, 020 Cite, pp. 157-1590 60 It is reasonable to believe that the subjects selected in the present investigation are similar with regard to the attributes mentioned above to Mustang drivers residing in the other interurbian regions scattered across the United States. Data Collection Sample Selection Between April 1964, the month of the Mustang's introduction, and September 1966 about 92,000 new Mustang automobiles were regist- ered in the State of Ohio. Of this number approximately 3,10016 were registered to private parties residing in Lucas County (population 482,000) and Wood County (population 79,000). Both of these adjacent counties are located in the interurbian strip described previously. In order to obtain a representative cross-section of these owners, 153 clusters composed of five individual Mustang owners were selected from this two county population on a systematic time ordered basis. The number of subjects contacted in each county approximated the proportionate population of the two counties; the great majority were residents of the city of Toledo and its satellite suburbs. To be included in the study, the subject had to meet two basic qualifications. First, he had to be the principal operator of the vehicle, and second, he had to make or share in the decision to buy the car. The fact that a person had a vehicle registered in his or her name gave no assurance that this was the person to be interviewed. Being aware of this situation and further recognizing the impossibility of locating all the subjects, 153 systematic ordered clusters of re- gistered Mustang owners were purchased from the R. L. Polk Company. 16R. L. Polk & Company, Detroit, Michigan. 61 Each cluster contained the names, addresses, and registration dates for five individuals having a Mustang automobile titled in their name. The skip interval between the clusters was fifteen. The first sub- ject in each of the 153 clusters was assigned the letter A, the second subject B, the third C, the fourth D, and the fifth E. Ideally, all. 153 of the persons interviewed would have been from the A classifica- tion. Nevertheless, it was possible to talk to any one of five people in a given cluster, A through E, and maintain the time ordered sample. Interview Procedure The first step in collecting the data involved the identifica- tion of the 153 individuals who were both the principal users of the Mustang and also made or shared in the decision to buy the car. This was accomplished by telephoning the party whose name appeared on the sample card and inquiring whether or not they met these two criteria. Once the right person was contacted, the study was briefly explained and the subject was asked to participate. The Telephone Instruction Form is contained in Appendix A. Both the local and long distance calls were made by a former speech and English teacher. Personal interviews were scheduled between November 19 and December 13, 1966. Twelve Bowling Green State University male under- graduate students were hired and trained to collect the data. The training consisted of a two hour formal session and two practice interviews, one of which was made in the investigator's presence. Each student was equipped with two letters of introduction, other necessary interviewing materials, and a schedule of his interviewing assignments (see Appendix C). 62 Interview Summegy One hundred and fifty interviews were completed between November 19 and December 13, 1966. One hundred and twenty-two were made by the trained undergraduate students. Twenty-two were made either by the person scheduling the interviews or personally by the investigator. Out of necessity, fourteen interviews were made at the individual's place of employment and six were conducted over the telephone. In these latter cases the self-administrered portion of the instrument was mailed to the respondent. Each completed interview form was edited and the data were punched in IBM cards. Table 4 summarizes the number of respondents who were inter- viewed from each of the five categories. Table 4 - Data Collection Summary Category Number 1. "A" category cards 60 2. "B" category cards 40 2. "C" category cards 24 4. "D" category cards 16 5. "E" category cards 10 TOtal Completed Interviews 150 ‘22 jurp it was hoped that all of the data could have been gleaned from the nameajon the_"A" category cards contained in the 153 sampled clusters.cuThis.g§ facto was not realized. Three clusters were lost ' 1 because all offlthe subjects contained within the cluster were either inaccessible or refused to participate in the study. Eleven percent L I CW8. '1 ' of the total number of people contacted refused to participate. 63 Appendix D presents a summary of who was or was not interviewed and the reasons why individuals were not included in the analysis. Spmple Description This section presents a summary description of the respondents. A more detailed analysis is presented in tables contained in Appendix E. The ages of the 150 respondents varied from sixteen to eighty- nine; the median age is in the 29-33 year bracket. Sixty-nine percent of the respondents were male. All of the respondents had at least eight years of formal education. Ninety percent had completed high school and over nineteen percent had received a college degree. Six percent had taken advanced graduate work. The respondents' occupations indexed according to the North- Hatt occupational ranking scalel7 ranged from 44 to 96. Eighty-eight percent were between 50-79, and half of these were in the 60-69 category which is principally composed of blue-collar jobs or low income white-collar positions. Thirty percent of the respondents had made one or more changes in occupation during the last five years. This includes those sub- jects who were ranked as college students and who are now gainfully employed. Two job changes were reported by thirteen respondents and three changes were reported by four of the subjects. Forty of the fortyefive people who made changes also gained or 17Same as footnote 4 (p. 53). 64 lost prestige in terms of the rankings of the North-Hatt classifica- tion system. Thirteen of these occupational changes exceeded ten North-Hatt scale points. No one experienced more than a twenty point advance or decline over the five year period. The Obtained social status scores ranged between 10 and 25. The modal category was 15 and the median category was 17. Nearly fifty percent of those interviewed had changed address at least once in the five years preceding the date of the interview.* Description of the Variables Under this heading a descriptive summary for each Of the dependent variables and the independent variable is presented. The first general area to be considered is source utilization. Source Utiligption Twenty-five subjects reported that they used a different type of information source at each of the successive stages of the decision making process. Sixty-seven percent relied upon two types of sources before adapting the innovation. Only nine based their buying decision upon a single information source. Three respondents did not know which sources had effected their adoption decision. *This figure probably understates the geographic mobility of the Mustang owner because relocations were one of the principal reasons why the interviews were not made entirely from the "A" category cards. 65 Media Exposure 1. 2. Newspppers All but three of the 150 subjects stated that they read the newspaper at least one hour during an average week. Of the 147 respondents who read the newspaper, fifty-three stated that they read only one newspaper. Sixty-four reported reading two different newspapers, twenty-three stated three, and seven signified four. Whereas the questions pertaining to media exposure were based upon exposure during an average week, questions concerning advertising recall were based upon recall for the past week. Sixty-nine subjects recalled seeing at least one newspaper advertisement for the Mustang during the week preceding the date of the interview.* Magazines One hundred and thirty-two respondents stated that they read magazines for at least half an hour during an average week. The number of different magazines varied from one to sixteen. Seventy-one subjects recalled seeing at least one Mustang advertisement as they paged through magazines during the week preceding the interview. 3. RadiO More than ninety-seven percent of those interviewed reported *This figure is probably lower than usual as the Toledo newspapers were on strike when the data was gathered. 66 that they listened to the radio a minimum of thirty minutes during an average week. Many of the respondents who listened more than twenty hours per week admitted that the radio was on constantly during working hours. 0f the 146 subjects who listened, all but thirty-six tuned to more than one station. Seventy persons indicated that they recalled hearing at least one Mustang commercial in the seven day period preceding the date of the interview. 4. Television One hundred and forty-two persons stated that they watched television at least half an hour during the average week. All but five of the 142 respondents watched more than one channel. Ninety-one persons recalled seeing at least one television commercial for the Mustang during the week preceding the inter- view. The number ranged between one and thirty. gpppyppgveness Mustangs were purchased by the respondents in each of the thirty months included in the investigation. It is interesting to note that the subject in the seventy-fifth sample cluster purchased his Mustang during the fifteenth month that the automobile was on the market. One hundred and three persons believed they were the first person in their immediate neighborhood to own a Mustang. Nine reported that they were one of the first. Thirty—five stated that 67 they were not among the first. Three people, because of change of address shortly before receiving the car, did not know if they were among the first. One hundred and nine subjects felt that they were the first among their circle of friends to acquire the innovation and seven considered themselves to be one of the first. On the other hand, thirty-four indicated that several of their friends owned Mustangs before they bought theirs. Product pralty Forty-seven subjects purchased their Mustang on a "clean deal" or no trade basis. Forty-five traded in other Ford products and fifty-three traded in cars made by another manufacturer. Surpris- ingly, five subjects traded in an older Mustang to acquire their present vehicle. Seventy subjects owned a second car. Forty-one owned a car produced by another manufacturer and twenty-eight owned a second Ford product. One respondent currently owned two Mustangs. Twenty persons had already sold their Mustang by the date of the interview. Eight of these respondents were now driving a car made by another producer and nine had purchased another Ford product. Three had already purchased their second Mustang. 0f the twenty cars purchased, six were of a similar body type; i.e. Monza, Barracuda, Marlin, but fourteen were of a different body type. One hundred and thirty subjects were asked to state their purchase intentions in terms of (1) the make of car they would 68 most likely purchase and (2) the body type they would most likely select. Twenty-seven stated that they "did not know" which make they would prefer. Fifty reported that their next car would be another Mustang and thirty-four were partial to other Ford products. Nineteen stated that their next automobile would be made by another manufacturer. With regard to body type, fifty-eight said they would purchase a similarly styled vehicle, but forty-one preferred a different body type. Opinion Leadership One hundred and thirty-one subjects stated that they had talked about their Mustang at least once during the week preceding the interview. The number of conversations varied from one to fifty-seven. Twenty of these 131 subjects reported that they initiated the conversation on every occasion. On the other hand, twenty-five stated that they never began the conversation, while thirty-seven indicated that they began the conversations about half the time. One hundred and six subjects had allowed someone other than the members of their immediate family to drive their Mustang. The number ranged from one to ninety-six. The respondent who had allowed.ninety-six others to drive his car was employed by a Ford dealer located in the Toledo area. Seventy-seven persons believed that they had influenced at least one other individual to purchase a Mustang. Two subjects did not know whether or not they had influenced anyone, and twenty-four 69 did not believe they were instrumental in effecting the purchase decision of others. Homophily Forty-five subjects stated that all of the people they talked with about the Mustang lived in neighborhoods similar to the ones that they themselves lived in. Conversely, twelve respondents reported that all of the people they talked with about the car lived in dissimilar neighborhoods. Two individuals did not know what percent of the people they talked with ah.ut the Mustang lived in similar neighborhoods. Twenty subjects asserted that all of the persons they talked with about the car had similar occupations. Twenty-five reported that all of the people they talked with had occupations unlike their own. Twenty-six thought about half of the people they talked with did about the same kind of work they did. Two did not know the percentage breakdown. Forty-nine individuals said that all of the persons they talked with had educational backgrounds similar to their own. Nine, on the other hand, indicated that everyone they talked with about the car had either more or less formal education than they themselves had. Seven respondents did not know what percent of the people had educational backgrounds similar to their own. Thirty-two respondents reported that everyone they conversed with about the car was the type of person who likes to try new and different things. Eleven subjects, conversely, stated that none of the people they talked with would be favorably diaposed towards 70 new and different things. Forty persons said that one hundred percent of the people they talked with about the Mustang had favorable opinions about the car before they talked with them. Six, on the other hand, reported that none of the people they talked with had favorable opinions. One hundred and ten subjects stated they had talked with at least one other party who owned a Mustang. The number ranged between one and twenty-five. Dogpatism The Obtained dogmatism scores ranged from 25 to 118. A summary of the scores are presented in Table 5. Table 5 - Obtained Dogmatism Scores Dogmatism Scores Frequency Percentage 20 - 29 l .7 30 - 39 1 .7 4O - 49 12 8.0 50 - 59 26 17.3 60 - 69 28 18.7 70 - 79 38 25.3 80 - 89 23 15.3 90 - 99 16 10.7 100 - 109 3 2.0 110 - 119 2 1.4 120 - 129 0 0 130 - 140 __O 0 Total 150 100.0% Analysis of the Data Indeterminate Responses In order to prepare the raw data for analysis, indeterminate answers had to be recoded. The procedure used in recoding was based 71 upon the probability of receiving a given answer. In several instances the indeterminate answer, i.e., no response, don't know, can't remember, was recoded to the median category of obtained responses because it was believed that this answer had the greatest probability of being given. In other situations, indeterminate responses were recoded to zero. The frequency of indeterminate responses only exceeded two percent of the obtained data on six questions. On each of these six occasions the sample size used in subsequent analyses was reduced accordingly. Oppple Size Variation Because several questions included on the interview schedule form could not legitimately be asked of all 150 respondents, i.e., make of trade-in, make of second car, make of replaced car, seven separate sub-samples which varied in size from twenty to 150 were used in the analyses. Although they are not mutually exclusive the sub-samples reflect a more precise picture of the answers obtained from the given questions. Primapy Analysis The principal statistical method used in the analysis was correlation. Accordingly seven intercorrelation matrices, one for each sample, were generated. The matrices revealed the zero-order correlations existing between the dogmatism scores and the several dependent variables. The matrices indicated further the extent to which any or all of the eight control variables were correlated 72 with (1) the independent variable, (2) the dependent variable(s), and (3) one another. In those situations where there was no 18 between a given control variable and significant correlation the independent variable and/or the dependent variable being analyzed, the control variable was excluded from further analysis. Secondagy Analysis In the event that a control variable significantly correl- ated with either the independent variable or a dependent variable, its effects were statistically eliminated through the computation of partial correlation coefficients. Because there is no sampling distribution for partial correlations, the obtained coefficients were transformed into t scores which reflected allowances for degrees of freedom and size of sample. 18A one tailed test at the five percent level was used for determining statistical significance. See Wilfrid J. Dixon and Frank J. Massey, Jr., Introduction pp Statistical Anal sis, McGraw-Hill Book Company, New York, 1957, p. 468. Chapter IV RESULTS Both the zero-order correlations and the partial correlations existing between dogmatism and the twenty-three dependent variables are presented in Table 6. General Hypothesis.;: HIGH DOGMATICS UTILIZE FEWER COMMUNICATION SOURCES THAN LOW DOGMATICS. The empirical hypothesis, EH la, tested under the first general hypothesis is high dogpatics util$zg fewer communicatipn sources than lpy,dpgpatics. The correlation between dogmatism and number of communication sources is -.058, which is not statistically different from zero. Moreover, the correlation is not in the expected direction. Education and social status are significantly correlated with dogmatism, and age is significantly correlated with the dependent variable. After statistically eliminating the effects of these three control variables, the resulting partial correlation coefficient is -.081, which is not significantly different from zero. Therefore, empirical hypothesis la is rejected. Because this empirical hypothesis is rejected, the first general hypothesis cannot be accepted. General Hypothesis-g: LOW DOGMATICS EXPOSE THEMSELVES MORE TO MASS MEDIA COMMUNICATION CHANNELS THAN HIGH DOGMATICS. The first empirical hypothesis, EH 2a, tested under the second general hypothesis is M _dpgnatg have p gpepter number _o_f_ herp pf_ egposure 3p p:_a_s_s_ 99113 communication phaknnels than _lLigp dogpatics. The correlation coefficient between dogmatism and hours of exposure is .125, which is neither significant nor in the hypothesized direction. 74 Non .OOOOOO Hoaoom .ooauooouo .huHHHoOe .omfiumoum HOOOHOOOOOOO .muaafiooa Hooowuoooooo mom.HI mma.| OmH.I unmoa wouuomouo memo amauoewov me an Mom .OOOOOO HOHOOO .omu .oOHuooouo «o~m.~1 «em~.1 «mm~.1 moH wouuomouo oxoa amfiuoemov we mm meadow . Heaoom .ooauooovo uooaoooaoou .mxamu Huaonumasuuo ”emo.~ ”awe. ”Nan. oN no mess amaumawoe we mm mauouo _ unoamuoamou Hoaoom .ooauooovo ue¢n~.e ”eaen. ”emon. 0N mo oxoa emfiuoawov oe mm Non .OOOOOO HOHOOO . .oo«uooovo .hufiafiooe owflumouo OOHOOOOOOO you .muHHHoOa.HooOauoooooo Hm~.1 woo.| «No.1 On vacuum mo oxoa amauoewov no mm meadow HOHOOO .oOfiuooowo ea .muwafiooa Hooo«uoooooo “wee.a. unqa. ”one. med Iououu «0 oxoe amwuoawoo we mm maumum mooowum Hoaooo .ooaumoovo ~m¢.I nmo.1 moo.1 omH wooed umuam amfiuoawoo on an moumum . voonuoonwaoo . Hanson .aoaumuaem mNm.n aao.u Neo.u omH an.umuau amauuamoe pm mm now .OOOOOO omens Howoom .aoaumoovo Nem.l mmo.l Hoo.l omH than no ouov smauoemov mm mm moumum Hamoou Hoaoom .ooaumoooo «no. Mano. ”Hmo. omH Hafiouoeeoo amwuoawov ON mm moumum . Howoom .ooauooovo masseuse waves .xouu HOOOHuoOOOOO ”mom.H ”med. ”moo. omH moms mo .Oo amfiuoewov AN mm maumum unamomxo Heaoom .OOHOOOOOO ”mmw. .umno. ”mNH. oma mo mason Beaumemov ON mm mwm .mououm nowumuwafiuo .HOfiOOm .oowumoooo ”mum.1 ”Hwo.| ”mmo.l oma mouoom amauoewov OH mm vowaouuooo moanmwum> u u u Hovuo onwo unaccomon uaowaonovoH mmmonuOONm Heauuom Iowan oaoewm moaoowuo> Hoowuanem moaomwuo> uoovoooon,onu vow oaouwum> uoouooooveH one cmosuom wewumfixm mooaumaouuoo Hoquuom vow mooauoaouuoo nowuo ouou 1 o magma 75 :OHuwmooeoo uaouomm«v mo one moaoaom ponamonuooan none Ou ouflmoooo coauuoufin mum. 1 ON 2 com. I 0* z moa. I moa 2 sea. I oma z .Avoawmu mgovao>oa unoonoo m Ono no unmeaMHome hHHmOHumaumum « - maumum Huaoom .ooaumoouo OOHuoopm .huuafiooe HOOOHOOOOOOO Newa.l undo.l ”omo.l omH mo kHHnO060: amfiuoewov we mm maumuo Hofioom .oOHumO OOHOHOO loco .xomu HOOOHOOOOOOO ”m¢N.HI ”moH.I Noma.l oma mo %chooeoe amauoewov mo mm onumum omooo>fiuo>ooofi Hofioom .oowuooouo wmw.l mmo.l who.| omH mo maficooeon Beaumawov no mm mauoum mawnmoaon Hofioom .ooaumoovo ”Nam.l ”meo.1 ”mqo.1 omH HOOOfiuooouo Boguoewou 00 mm moumum hafinmoaon H366... €638.66 32.- ”30.: ”8o... o3 3:633:86 83638 no we Mom .m9umum hawsmoaos 368 5638.66 ”93.7 $8.: ”$2.: 03 432638.. amEmamoe 8 mm maumum Howoom .oOHumoouo voooooawafi .5236... 3833366 33. $8. $8. o2 33on? .6: 83.688 an em muo>fiup nausea Howoow .oowuooooo ”ooh. Mano. ”omo. ona umou mo .oo amwumawoo on mm emooauwofi meadow mooaummuo> Howoom .OOHOOOOOO o o ”coo. omH Iooo mo unwound amfiuoewov on mm meadow HOHOOO .oowomo cowofioo taro .xoou Hoooaummouoo mmq.HI oma.l «no.1 omH mo odoamooumxo aneuoawov mm mm poqflouuooo moHomHum> u u u Houuo swam unouoooon uooocomwooH mamosuoowm Hoauuom IOHON oaoamm moaoowum> HOOHuaoem 3533... no 633. 76 Because education and social status significantly correlate with both the independent variable and the dependent variable, they are statis- tically controlled. The partial correlation of .093 is neither signi- ficantly different from zero nor in the expected direction. Conse- quently, empirical hypothesis 2a is rejected. The second empirical hypothesis, EH 2b, is gpy’dogpatics eypose themselves Q BEE £3.33 medfit. communipetion channels pm h_13p do - mptics. The correlation between dogmatism and number of communication channels is .043, which is not significantly different from zero. Occupational rank is significantly correlated with the number of channels; education and social status are significantly correlated with both the independent variable and the dependent variable. By statistically controlling these three variables, a partial corre- lation of .108 is obtained. Neither the zero-order correlation nor the partial correlation are significant or in the hypothesized direction. Therefore, EH 2b is rejected. The third empirical hypothesis, EH 2c, isuppppp.;pywip.gpgpp- - _tiLIE scores egpose themselves _tp m Mustapg commercials _t_ppp _t_h_O§_e_ ‘pégp.;pldogpatism scores. The correlation between dOgmatism and com- mercial recall is .091, which is not significantly different from zero. By controlling social status, which significantly correlates with both the independent variable and the dependent variable and education, which significantly correlates with dogmatism, a partial correlation of .052 is produced. This correlation is not significantly different from zero. Since neither the zero-order correlation nor the partial correlation are significant, or in the expected direction, EH 2c is rejected. 77 Because all three of the empirical hypotheses tested under the second general hypothesis are rejected, the second general hypothesis is similarly rejected. Genergi Hypothesis.§: LOW DOGMATICS ARE MORE INNOVATIVE THAN HIGH DOGMATICS. The first empirical hypothesis, EH 3a, tested under general hypothesis 3 is ipy,dogpatics.pgppp.ppe innovation relatively earlier L11 imp ph_ap_ my, dogpatics. The correlation existing between dogmatism and date of purchase is -.001. A partial correlation of -.029 is obtained when the effects of education and social status, both of which significantly correlate with dogmatism, and sex, which significantly correlates with date of purchase, are controlled. Because neither the zero-order correlation nor the partial correlation are significantly different from zero, EH 3a cannot be accepted. The second empirical hypothesis EH 3b, is lpy;dogpatics pp;- pejye themselves pp more innovative pip m _pljp iiLst _ip their imme- .O;ppp neighborhood ppdpgppp pp innpygpiqp than plgp dogpatics. The correlation of -.042 between dogmatism and being the first to adOpt in the neighborhood, although in the predicted direction, is not signifi- cantly different from zero. A partial correlation of -.077 is obtained when social status and education, both of which significantly correlate with dogmatism, are statistically controlled. Because neither the zero- order correlation nor the partial correlation are significantly differ- ent from zero, EH 3b is refuted. The third empirical hypothesis, EH 3c, is lpy Opgpptics perceive themselves pp more innovative 3p being first among their circleng 78 friends.pp.pgppp.ppu;pnovation than p;gp dogpatics. The correlation between dogmatism and first among circle of friends to adopt is -.005, which is not significantly different from zero. The statistical elim- ination of education and social status, which significantly correlates with dogmatism, yields a partial correlation of -.037 which similarly is not significant. Consequently, EH 3a is rejected. Although all of the correlations in the three empirical hypo- theses are in the predicted direction, none are significantly differ- ent from zero. Therefore, the third general hypothesis cannot be accepted. Gpneral Hypothesis-3: HIGH DOGMATICS EXHIBIT GREATER LOYALTY TOWARD PRODUCTS THAN LOW DOGMATICS. The first empirical hypothesis, EH 4a, tested under general hypothesis 4 is pggpugpgmatics traded gp Ford products more Often than ‘lpy,dogpatics. The correlation coefficient between the two variables, dogmatism and make of car traded-in, is .060 which is not significantly different from zero. Statistically removing the effects of (l) educa- tion, which significantly correlates with both the independent variable and the dependent variable, (2) social status, which significantly correlates with dogmatism, and (3) occupational rank, which signifi- cantly correlates with make traded-in, generates a partial correlation of .145. As neither the zero-order correlation nor the partial cor- relation are either significant or in the expected direction, EH 4a is rejected. The second empirical hypothesis, EH 4b, is pg those ownipg 5 second pa_r, _h_:l._gl_1_ _clpgmatipp own 3 second Ford goppct more mp M do - matics. The correlation coefficient of -.022 between dogmatism and 79 make of second car, although in the hypothesized direction, is not significant. Statistically controlling education, social status, and sex, all of which significantly correlate with dogmatism, and occupa- tional mobility and occupational prestige mobility, which correlate with make of second car, generates a partial correlation of -.068. The partial correlation is not significantly different from zero. There- fore, EH 4b cannot be accepted. The third empirical hypothesis, EH 4c, is O;ggwgpgmatics replgpe pppgp original Opetang‘with.p sbmilar.pypeup§.§ptppobile more often ‘ppppn;py_dogpatics. The correlation between dogmatism and replacement of original Mustang withasecond Mustang is .502 which is significantly different from zero. But the significance is Opposite to the direction hypothesized. A partial correlation of .741 which similarly is statis- tically different from zero in the Opposite direction is obtained by controlling social status which significantly correlates with dogmatism, and education which significantly correlates with the make of car pur- chased.’ As both the zero-order correlation and the partial correlation are statistically significant in the direction Opposite to that which was predicted, EH 4c is rejected. The fourth empirical hypothesis, EH 4d, is plgp_dogpatics replace their origippliMustepguyipp p shmilar‘pypp.p§ automobile more Often [ppppnipy,dogpatics. The correlation of .192 between dogmatism and type of replaced automobile is neither significant nor in the hypothesized direction. The partial correlation of .489 which is produced by con- trolling social status, which significantly correlates with both the 80 independent variable and the dependent variable, and both occupational rank and education, which correlate with the dependent variable, simi- larly is neither significant nor in the predicted direction. Conse- quently, EH 4d is rejected. The fifth empirical hypothesis, EH 4e, is high dogmatics state _a_ ,Lreference for replacing ppejp; original Mustang £9.32 3 pg Mustapg more.p£ten‘phpp.lpy’dogpatic . The correlation coefficient between dogmatism and preferred make is -.238 which is both in the expected direction and significantly different from zero. Statistically con- trolling education, social status, and sex, all of which correlate with dogmatism, and age, which correlates with the make of car preferred, produces a partial correlation of -.254 whichsimilarly is both signi- ficant and in the right direction. Therefore, EH 4e is accepted. The sixth empirical hypothesis, EH 4f, is pigp dogmatics state 5 preference for replapggg their origipa_1 Mustang M _a_ similar _typ_e_ pgnputomobile more often than low dogmatics. The correlation of -.150 between dogmatism and preferred body-type is in the right direction, but not significantly different from zero. A partial correlation of -.133 is obtained by statistically controlling the effects of educa- tion, social status, and sex which significantly correlate with dogma- tism and occupational mobility, occupational prestige mobility and age which significantly correlate with the dependent variable. The partial correlation, like the zero-order correlation, is in the predicted dir- ection, but not significant. Therefore, EH 4f cannot be accepted. Of the six hypotheses tested under general hypothesis 4, only one, EH, 4e, was statistically significantly different from zero in the expected direction. Empirical hypothesis 4b, on the other hand, 81 was statistically significant in the direction opposite to that which ‘was predicted. Because of these conflicting findings and the results in the four other empirical hypotheses, general hypothesis 4 cannot be accepted.‘ Generpi Hypothesis.§: LOW DOGMATICS EXHIBIT MORE OPINION LEADERSHIP HIGH DOGMATICS. The first empirical hypothesis, EH.5a, tested under general hypothesis 5, is lpy_dogmatics express their Opinions about their Mus- pppg'more frequently than high dogmatics. The correlation between dog- matism and the number of times opinions are expressed is -.084 which, although in the predicted direction, is not significantly different from zero. Statistically controlling the effects of (1) social status which significantly correlates with both the independent variable and the dependent variable, (2) education which significantly correlates with the independent variable, and (3) occupational rank which correlates with the dependent variable yields a partial correlation of -.120. The partial correlation similarly is not significant. Consequently, EH 5a is rejected. The second empirical hypothesis, EH 5b, is lpyygpgmetics‘lnitigpp lpipcussions about the;p_Mu§pppg_more often than plgp dogmatics. The correlation between dogmatism and percentage of conversations initiated by the respondents is .060 which is neither significant nor in the hypothesized direction. A partial correlation of 0 is obtained when social status, which significantly correlates with the independent variable, and education, which correlates with both the independent variable and the dependent variable, are controlled. 82 Because neither the zero-order correlation nor the partial correlation are significantly different from zero, EH 5b is rejected. The third empirical hypothesis, EH 5c, is lpy dogpatics pllpy more people pp test drive their Mustang than high dogpatics. The correla- tion of .076 which exists between dogmatism and the number of people who had test driven the Mustang was neither significantly different from zeroxxnrin the predicted direction. The statistical control of both education and social status, which significantly correlate with dogmatism, produces a partial correlation of .059, which similarly is neither significant nor in the expected direction. Consequently, EH 5c is rejected. The fourth empirical hypothesis, EH 5d, is low dogmatics er- pem _t_l'fl h_ay_e_ influenced more receivers M high dogmatics. The zero-order correlation between dogmatism and number of people influenced is .056 and is neither significant nor in the hypothesized direction. Statistically controlling occupational mobility, education, and social status all three of which significantly correlate with dogmatism yields a partial correlation of .078. The partial correlation, like the zero- Order correlation, is not significant. Therefore, EH 5d is not con- firmed. Because none of the four empirical hypotheses tested under gen- eral hypothesis 5 have correlations significantly different from zero, the fifth general hypothesis is rejected. Oenerpl Hypothesis 6: HIGH DOGMATICS EXHIBIT GREATER HOMOPHILY THAN LOW DOGMATICS. The first empirical hypothesis, EH 6a, which is tested under general hypothesis 6 is high dogmatics interact with p greater 83 percentpge pg pepple who are perceived pp reside ;p_similar neighbor- ,pppgp_£ppp.lpy,dogpatics. The correlation between dogmatism and per- centage“of peOple residing in similar neighborhoods is -.158 which is statistically different from zero, but opposite in direction to that which was predicted. A partial correlation -.096 is obtained when both education and social status, which significantly correlate with both the independent and dependent variable, and sex, which significantly correlates with the dependent variable, are controlled. The partial. correlation is in the opposite direction, but not significantly different from zero. As a result, EH 6a cannot be accepted. The second empirical hypothesis, EH 6b, is high Opgmatics inter- pc_t_ w_i_t_:_l_1_ 5 wer perceppege o_f_ people who Lrp perceived _gp ppyp similar occupations than ipy dogmatics. The zero-order correlation of -.068 which exists between dogmatism and percentage of people having similar occupations is neither significantly different from zerOonor* in the hypothesized direction. The statistical control of education and social status, both of which significantly correlate with dogmatism, produces a partial correlation —.045 which similarly is not significant. Therefore, EH 6b is rejected. The third empirical hypothesis, EH 6c, is high ppgmatics intepgpp £13.11 _a_ ggpiter percentgagp pf_ Bople who are pgrceived pp flyp similar educational backgrounds than low dogmptics. The correlation between dogmatism and people having similar educational backgrounds is -.049 which is neither statistically different from zero nor in the predicted direction. A partial correlation of -.049 is obtained by controlling education and social status both of which significantly correlate with 84 dogmatism. As neither the zero-order correlation nor the partial corre- lation are significant, EH 6c cannot be accepted. The fourth empirical hypothesis, EH 6d is high dogmatics inter- ppp 33p pmater percentage 5E mople who are grceived fl reluctant _pp innovate than low dogmatics. The correlation of -.077 existing between dogmatism and percentage Of people who are perceived as reluc- tant to innovate, although in the predicted direction, is not statis- tically different from zero. A partial correlation of -.073 is gener- ated by controlling education and social status, both of which signi- ficantly correlate with dogmatism. The partial correlation is not significant. Consequently, EH 6d is rejected. The fifth empirical hypothesis, EH 6e, is pigp dogmatics‘lpppp- £3 w p greater percentage .o_f_ people who are perceived _t_o_ m _lgp .p_positive Opinion about the Mustang. The correlation of -.l36 between dogmatism and percentage of peOple perceived to have had a positive opinion about the Mustang is neither statistically different from zero nor in the predicted direction. A partial correlation of -.103 is produced when (1) education and social status, both of which signi- ficantly correlate with both the independent variable and the depend- ent variable, and (2) occupational rank, which significantly corre- lates with the dependent variable, are controlled. As a result, EH 6e is rejected. The sixth empirical hypothesis, EH 6f, is high dogmatics interact 1_v_i_._t_:_l_1. p greater number pf; people who have adopted the Mustang than low dogmatics. The correlation between dogmatism and the number of people 85 already owning a Mustang is -.050 which is neither statistically dif- ferent from zero nor in the hypothesized direction. The control of education and social status, both of which significantly correlate with dogmatism, and occupational prestige mobility, which significantly correlates with the dependent variable, produces a partial correlation of -.015. The partial correlation similarly is not significant. Therefore, EH 6f is rejected. None of the correlations for the six empirical hypotheses tested under general hypothesis 6 were significantly different from zero in the expected direction. As a result, general hypothesis 6 cannot be accepted. CHAPTER V SUMMARY AND DISCUSSION Summary The ability to identify those particular individuals within a given market segment who are most likely to be among the first to adopt an innovation, to purchase the product repeatedly, and to pass informa— tion about the product on to others is a critical element which must be considered by the innovative marketer. The present study proposed the thesis that the socio- psychological theory of beliefs as formulated by Rokeach1 would generate new insights into some of the communication behavior involved in the diffusion and adaption of a consumer innovation, the Ford Mustang automobile. The six distinct areas of communication behavior investi- gated were (1) communication sources, (2) communication channels, (3) innovativeness, (4) product loyalty, (5) Opinion leadership and (6) homophily. The basic objective was to identify and measure the relationship which existed between dogmatism and each of these six con- cepts. Dogmatism was defined as a personality variable which governs a person's receptivity or lack of receptivity to new ideas and further includes how a person perceives, evaluates, acts, and reacts to such ideas. A review of the literature combined with an assessment of the theoretical position of belief systems led to the formulation of six 1Milton Rokeach, The Open and Closed Mind, pp. cit. 87 general hypotheses, which were investigated. The hypotheses were as follows: GH 1: HIGH DOGMATICS UTILIZE FEWER COMMUNICATION SOURCES THAN LOW DOGMATICS. GH 2: LOW DOGMATICS EXPOSE THEMSELVES MORE TO MASS MEDIA COMMUNI- CATION CHANNELS THAN HIGH DOGMATICS. GH 3: LOW DOGMATICS ARE MORE INNOVATIVE THAN HIGH DOGMATICS. GH 4: HIGH DOGMATICS EXHIBIT GREATER LOYALTY TOWARD PRODUCTS THAN LOW DOGMATICS. GH 5: LOW DOGMATICS EXHIBIT MORE OPINION LEADERSHIP THAN HIGH DOGMATICS. GH 6: HIGH DOGMATICS EXHIBIT GREATER HOMOPHILY THAN LOW DOGMATICS. Twenty-three empirical hypotheses, derived frOm the six general hypo- theses, were tested. 'A systematic ordered sample of 150 Mustang owners residing in two adjacent Northwestern Ohio counties located in the interurbian strip extending between Toledo~and Cincinnati were personally inter- viewed during a four-week period which extended from November 17 through December 13, 1966. In addition to responding to questions concerning the six aspects of communication behavior, each subject 2 dogmatism test. Dogmatism scores were correlated completed a short-form with the answers given to the other questions. Correlational analy- ses of the data led to the rejection of all but one of the twenty-three enpirical hypotheses. Eight of the zero-order correlation coefficients were in the expected direction, but fifteen were in the direction 2Verling C. Troldahl and Frederic A. Powell, "A Short-Form Dogmatism Scale for Use in Field Studies," _p, cit. 88 opposite to that which was predicted. Two of the latter were sig- nificantly different from zero. The findings were not appreciably affected by a secondary analysis consisting of the statistical elimination of control vari- ables which significantly correlated with either the independent variable and/or the dependent variable(s). The only correlation which was significantly different from zero in the expected direction was for empirical hypothesis 4e, pggp dogmatics ppppp p preference for peplacing their original Mustepg ‘yipp.p new Mustang more often than low dogmatics. Because twenty-two of the twenty-three hypotheses were rejected none of the six general hypotheses can be accepted. Explanptions for the Results The fact that only three of the correlations obtained in the present research were significantly different from zero leads to the general conclusion that the personality variable of dogmatism as measured in the present investigation is in itself of little value in predicting the communicative and adoptive behavior of individuals who recently acquired a new generic product. Several factors inherent in the investigation may have con- tributed to the unexpected results. First of all, the measurements used in the analysis may not have been sensitive enough to Operational- ize the intended meanings of concepts. If this measurement insensitivity were present, it most likely occurred with those questions pertaining to the unaided recall of past events especially those concerning sources of information, mass media exposure, and conversations with other people about the Mustang. 89 Another factor which may have contributed to the unanticipated results is the innovation which was selected for analysis. In all probability, beliefs about the Mustang would be incorporated at the peripheral level of the individual belief system. -As such, new be- liefs about the innovation could be more readily accepted by dogmatic individuals than beliefs concerning self or the nature of reality. Furthermore, the innovation in the present research is a relatively expensive consumer product; therefore, it is reasonable to believe that several prospective adopters may have excluded themselves from the sample by postponing adoption for as long as their present auto- mobile was still in good operating order. A third source Of uncontrolled variance pertains to the representa- tiveness of the respondent. Even though interviews were obtained from 150 of the 153 systematic time ordered sample clusters, it must be noted that eleven percent of the subjects chosen for analysis refused for one reason or another to participate in the study. Additional variance may have been occasioned because of different uncontrollable circumstances present in the homes of the respondents. Data were collected from eight in the morning until eleven in the evening; distractions and interruptions during the interviews were reported by the interviewers. Finally, as is the case in any study where numerous field workers are employed, there is an undetermined amount of inter- viewer bias. Relationships Between the Variables The correlation analysis used in the present investigation assumed linear relationships between the independent variable and the 90 several dependent variables. In the event that the linear assump- tions have been violated, the obtained zero-order correlation co- efficients (presented in Table 6) understate the true relationship between the variables. Therefore, an investigation into the validity of the linear relationships was undertaken. The analysis involved the computation of eta coefficients. This statistic was selected because, if a curvilinear relationship were present, the eta coefficients would be significant even though the zero-order correlation coefficients were not significantly different from zero. In order to compute the eta coefficients, the data were grouped by segmenting the dogmatism scale into five cate- gories which included the complete range of dogmatism scores obtained from the 150 respondents. The categories are: (1) 25-43; (2) 44-62; (3) 63-80; (4) 81-99; and (5) 100-118. The subjects were grouped into these categories according to their respective dogmatism scores. Once the eta coefficients were calculated, they were tested_ for significance by F ratios. As can be seen in Table 7, only one of the F scores (EH 4e) is statistically significant at the five percent level and.in this particular instance the zero-order correlation Obtained before the data were grouped similarly was significantly different from zero. As a result of this analysis, one can conclude that the low relations obtained in the present study were not the result of curvilinear relationships between dogmatism and the several dependent variables. Table 7 - Zero-Order Correlations, Eta Coefficients, F of Eta* Empirical Dependent Zero-Order Eta F off: Hypotheses Variables N Correlptions Coefficients Ete__ 1a source utilization 150 -.01 .15 .837 2a hours of exposure 150 .06 .15 .833 2b number of channels 150 .02 .15 .841 2c commercial recall 150 .05 .09 .294 3a date of purchase 150 .05 .13 .632 3b first in neighborhood 150 -.01 .18 1.220 3c first among friends 150 .05 .18 1.173 4a make of trade-in 103 .07 .15 .588 4b make of second car 70 -.10 .13 .299 4c make of replacement 20 .34 .51 1.350 4d type of replacement 20 .09 .54 1.500 4e make preferred 103 -.22 .29 2.262b 4f type preferred 103 -.12 .18 .851 5a number of conversations 150 .01 .15 .865 5b percent initiated 150 -.05 .09 .300 5c number of test drivers 150 -.12 .17 1.024 5d number allegedly influ- enced. 150 .06 .22 1.289 6a residential homOphily 150 .00 .11 .406 6b occupational homophily 150 -.04 .09 .317 6c educational homophily 150 -.05 .16 .974 6d innovative homOphily 150 -.10 .14 .747 6e homophily of opinion 150 .01 .08 .231 6f homophily of adoption 150 -.O3 .07 .179 a4 degrees of freedom in the numerator b *The coefficients illustrated in the Table were computed after grouping significant at the five percent level the data by segmenting the dogmatism scale as follows: (2) 44-62; (3) 63-80; (4) 81-99; and (5) 100-118. (1) 25-43; 91 92 Represeptptiveness.p£ the Subjects Another factor which possibly could have contributed to the unanticipated results is concerned with the subjects included in the sample. All of them had purchased and had been instrumental in the decision to adopt the innovation. This would lead one to suspect that the sample might be weighted in favor of low dogmatics and exclude those having higher dogmatism scores. However, an inspection Of the dogmatism scores reveals a nearly normal distribution. The scores ranged be- tween 25 and 118 and the median was 69.5. The arithmetic mean was 70.9 and the standard deviation was 16.2. The present distribution of dogmatism scores does not appreciably differ from the distribution obtained by Rokeach and others who have worked with various versions of the dogmatism scale.3 An interesting comparison of dogmatism scores between adopters and nonadopters of the innovation could have been made by sampling other individuals in the two county area who purchased a different make of car during the time period being investigated. The analysis was not made because such investigation was not included in the original objectives of the present thesis. 3See for example Juan F. Jamias, The Effects-pg Belief _System Styles-pp The Communication and Adoption_p§ Farm Practices, _p, pip., and Frederic A. Powell, "Open and Closed-Mindedness and the Ability To Differentiate Source and Message," pp._p;p. 93 Interpretation of the Results There are many other contributing factors, in addition to those previously mentioned, which may have led to the obtained results. Obviously in those situations where the correlations were not signifi- cantly different from zero, the results could be completely attributed to chance. Nevertheless, a brief discussion of each of the empirical hypotheses is presented in the paragraphs which follow. Communication Sources The predicted inverse correlation between dogmatism and number of communication sources utilized was neither statistically different from zero nor in the predicted direction. Although the finding is contrary to the theoretical expectation, one might speculate that the dogmatic individual was consulting with a wider variety of respected sources before incorporating the new belief into his system. More specifically, the high dogmatics may have exposed themselves to more evidence before accepting the new idea and adOpting the innovation. Although there is no evidence generated from the present study, one might conjecture that had the investigation analyzed rejectors as well as adopters, those high in dogmatism may have rejected the innova- tion after conferring with one authority, whereas, high dogmatics who accepted the new idea had to be confronted with evidence from several respected authorities. CommunLcation Channels In accordance with Rokeach's theory, those high in dogmatism are inclined to avoid contact with stimuli which may threaten their existing 94 beliefs. Therefore, it was expected that high dogmatic individuals would have less exposure to the mass media in terms of number of hours of exposure. Nevertheless the obtained correlation between dogmatism and hours of exposure was contrary to the hypothesized direction. Perhaps, as Rokeach has implied, high dogmatics may have used the mass media for added exposure to respected authorities and/or to reinforce existing beliefs; there is no evidence in the data,however,which can support this supposition. The relationship between dogmatism and number of communication channels attended similarly was neither significantly different from zero.nor in the expected direction. The finding is difficult to explain. It is possible that high dogmatics, and perhaps low dogmatics as well, used the media more for entertainment than for information. If-such were the case, the findings would not necessarily be in- congruous with the theory. Unfortunately, the study did not ask why and/or for what purposes the media were used. Consequently, there is no evidence to suppose that high dogmatics use the media for different purposes than low dogmatics. The correlation between dogmatism and exposure to Mustang commerci- als similarly was neither significantly different from zero nor in the expected direction. Although this finding was contrary to that which was hypothesized, it was consistent with the correlations obtained for the first two hypotheses considered in this section. There are two possible explanations for the obtained correlation between dogmatism and commercial recall. First, it may have occurred because those high 95 in dogmatism spent more time with the media. Second, it-cou1d be proffered that high dogmatics who adopted the innovatiOn experienced greater cognitive dissonance than low dogmatics who purchased the car.4 This position appears plausible as it may have been more difficult for a high dogmatic individual to accept the innovation in the first place; whereas, the less dogmatic person would have found his original adoption decision somewhat easier to make. If such were the case, it would be expected that close-minded individuals would seek informa- tion which would reinforce their purchase decision and thus were able to recall more commercials and/or advertisements than open-minded persons. Innovativeness Although the correlation relating dogmatism and purchase date was not significantly different from zero, it was in the expected direction. This finding is consistent with the results reported by Rogers and Harp,5 7 8 Jamias,6 Childs, and Hudspeth. The fact that the obtained relationship 4Gerald R. Miller and Milton Rokeach, "Individual Differ- ences and Tolerance for Inconsistency," in P. N. Tannenbaum, R. P. Abelson, E. Aronson, W. J. McGuire, T. M. Newcomb, and M. J. Rosenberg, (eds.) Theories p£_Opgnitive Consistenpy, Rand McNally, Chicago, in process. 5Everett M. Rogers and John Harp, "Personality Character- istics of the Adoption of Technological Practices," _p, cit. 6Juan F. Jamias, pp, cit. 7John W. Childs, pp, cit. 8DeLayne R. Hudspeth, pp, cit. 96 was not greater in the present study is difficult to explain. One possible explanation for the low relationship is that those high in dogmatism, who relied heavily upon the authority of commercial sources, and to a lesser extent upon noncommercial sources for information dur- ing the early stages of the product's life were influenced and accord- ingly changed their beliefs sooner as a result of the information which they had received. The second empirical hypothesis considered the respondent's adoption vis-a-vis those living in his immediate neighborhood, and the third pertained to adoption vis-a-vis one's circle of friends. Although the correlations were not significant, both were in the pre- dicted direction,* The findings of EH 3b and EH 3c lead one to speculate that those low in dogmatism acted independently of both their neighbors and their 9 and that closed minded individuals were perhaps making up friends their own minds or were being influenced by authority figures who did not practice as they preach or by commercial sources, i.e., advertising sponsored by the Ford Motor Company and/or its franchised dealers. *It is interesting to note that while fifty percent of the individuals interviewed purchased their Mustang after the car had been on the market fifteen or more months, more than two-thirds be- lieved they were the first in their neighborhood to own a Mustang and an even larger percentage stated they were first among their circle of friends to buy the car. This situation occurred in all likelihood because of the very low percentage of families owning Mustangs by July 1965. 9Leon Festinger, "Behavioral Support for Opinion Change," Public Opinion Quarterly, Vol. 28, No. 3, Fall, 1964, pp. 404-417. 97 Product pralty Generally speaking, satisfied automobile owners tend to purchase the same make of car, or another make made by the same manufacturer, time after time. Nevertheless, there is a certain amount of switch- ing which takes place. It was expected (1) that individuals scoring high in dogmatism would have been previous owners of Ford products, and (2) that fewer closed minded people would have traded-in cars made by other manufacturers. Because high dogmatics are more reluctant to accept new ideas until they have received the approval of authority figures, it was predicted that these persons would have at least remained faithful to the automobile producer if not to the Specific make. If their loyalty were to the Ford MOtor Company, they could have purchased the Mustang without changing their beliefs about automotive manufact- urers, on the other hand, if they traded-in a vehicle produced by another company, this would be overt evidence of belief change. Notwithstanding the relationShip between dogmatism and make traded- in was neither significantly different from zero nor in the expected direction. One possible explanation for this unanticipated result is that high dogmatics who were satisfied with cars produced by a given manufacturer remained loyal, but those who were in some way dissatisfied had no qualms about switching. Unfortunately no evidence to support this contention is available as questions pertaining to the reasons for switching from one manufacturer to another were not asked. 98 The Obtained relationship between dogmatism and make of second car, although not significantly different from zero, was in the expected direction. This finding suggests that high dogmatics are perhaps slightly more inclined to consider the make of their present car before purchasing a second automobile than are low dogmatics. The possession of two cars made by the same company is visible evidence of consistency of beliefs toward a particular producer; whereas, the ownership of cars built by two competing manufacturers would in- dicate a greater breadth and/or differentiation of beliefs on the part of the owner. The fact that the correlation was not significant suggests further that high dogmatics are perhaps just as likely as their less dogmatic counterparts to simultaneously own two automobiles made by two different companies in the event they are not completely satisfied with the first automobile. The finding additionally suggests that closed- minded individuals who were satisfied may have been slightly more in- clined to purchase their second car from the same manufacturer than the more open-minded respondents. No evidence is available to support this conjecture. It was additionally expected that high dogmatic individuals who had purchased the Mustang would be the most enthusiastic and loyal customers. This position was postulated for two reasons. First, it was belieVed that several of the more dogmatic adopters would have been the traditional or habitual buyers of Ford products the type of person who would blindly defend his purchase. It was believed further that high dogmatics who had changed their beliefs in the first place would defend their beliefs in an overt way. Nevertheless, a statisti- cally significant negative relationship was obtained between dogmatism 99 and purchase of a second Mustang. But in terms of purchase intention, the correlation was significantly different from zero in the predicted direction; a positive relationship exists between dogmatism and the desire to replace the present Mustang with a second Mustang. The in- consistency between the two sets of findings once again suggests satisfaction in ownership as a critical intervening factor in product loyalty. Niether the correlation relating dogmatism and the pypp of replaced automobile nor the correlation relating dogmatism to the pypp of car preferred in the future were significant. The former was opposite in direction to that which was hypothesized; albeit, the latter was in the expected direction. These findings are similar to the findings for the two hypotheses concerned with the make of car and, therefore, could be explained in a similar way. _Opinion Lepdership The correlation between dogmatism and number of conversations about the innovation was in the hypothesized direction. The fact that it was not significant, however, suggests that people high in dogmatism talk about the innovation almost as frequently as their more Open-minded counterparts. This does not mean that they interact with more people nor does it necessarily imply that high dogmatic individuals are obtaining information and/or advice from the people with whom they are engaging in conversation. Although it was predicted that low dogmatics would initiate more conversations about the Mustang than high dogmatics, the Opposite situation was found in the present study. Rokeach's theory implies 100 that closed-minded persons tend to be "true believers", and the obtained correlation suggests that these individuals may be slightly more inclined to talk about the idea that they had accepted, the be- liefs which they had embraced, and the innovation which they had adopted. Similarly, the obtained correlation between dogmatism and the number of people who luui test driven the Mustang was not signifi- cantly different from zero. The finding indicates that the more dogmatic individuals allowed a greater number of people to drive their Mustang, which may be evidence of their desire to proselytize others toward the innovation. Although the correlation relating dogmatism and the number of people perceived to have been influenced was neither significantly different from zero nor in the hypothesized direction, the finding is consistent with the findings for the two preceding empirical hypotheses. It was expected that low dogmatics would be more in- fluential because they would be interacting with a greater number of others. Data pertaining to the total number of peOple talked with was not collected; consequently, there is no evidence upon which to refute or confirm this proposition. Nevertheless, the obtained data suggests, if anything, that high dogmatics were possibly more convincing at least in terms of the number of people perceived to have been influenced than their less dogmatic counterparts. Once again it appears that high dogmatics who adopted the innovation may have been trying, and in more instances may have succeeded, in convincing others about the merits of the product. In other words, 101 high dogmatics may have been interacting with fewer people, but they seem to have exerted more total influence on others, than the more Open-minded individuals. As a result of the analysis one might argue that opinion leader- ship should be redefined in order to discriminate between breadth of leadership and depth of leadership. Judging by the obtained correlations, one might speculate that low dogmatics exhibit some- what greater breadth of leadership as they possibly interact with a greater number of others, whereas the high dogmatics possibly exhibit greater peppp_in leadership based upon their perceptions of influenc- ing Others. HomOphily The relationship between dogmatism and residential homOphily was significantly different from zero in the direction opposite to that which was predicted. Perhaps open-minded individuals are more gregarious than are more closed-minded persons and thus became better acquainted with their immediate neighbors and with a greater total number of others residing in similar neighborhoods. The finding suggests further the possibility, which would be in accordance with the theory, that the more dogmatic individuals were more selective in their choice of friends and used criteria other than neighborliness as bases upon which to build an acquaintance and/or friendship. The fact that there was a negative relationship between dogmatism and occupational homophily suggests, if anything, the proposition set forth in the preceding paragraph that high dogmatics were using criteria other than co-workers in choosing peOple with whom they wished to 102 associate and talk. Additional evidence in support of this select- ivity is suggested by the negative relationship which existed be- tween dogmatism scores and educational homophily. Even though the relationship between dogmatism and perceived reluctance to innovate was not significant, it was in the predicted direction. This finding implies that those high in dogmatism were interacting with people who were more conservative in their attitudes towards new things. Although no support can be obtained from the present research, one might conjecture that these individuals were high in dogmatism. The negative, although not significant, relationship between dogmatism and perceived homophily of Opinion about the Mustang suggests that the more closed-minded individuals could have been interacting with people who tended to;view the innovation with caution; possibly others who were high in dogmatism. Finally, the negative relationship between dogmatism and the number of other peOple owning a Mustang shows that high dogmatics were not interacting with as many people who already owned a Mustang as were their less dogmatic counterparts. There are two possible explanations as to why the correlation was in the direction Opposite to that which was anticipated. First, if those high in dogmatism selected others who were high in dogmatism as their friends it can be seen why high dogmatic individuals knew fewer persons who had adopted the innovation. Second, it appears that low dogmatic in- dividuals possibly interacted with a greater total number of peOple than the more closed-minded persons. If such is the case, Open-minded 103 individuals probably would know more people owning any specific kind of automobile than would closed-minded individuals. Commentapy It is difficult to support the conjectural statements appearing in the preceding paragraphs because of the methodology employed in the present investigation. More information about the relationship existing between dogmatism and innovativeness could have been obtained if both rejectors and adopters had been included in the analysis. Similarly, more definitive statements in the area of product loyalty could have been made had questions concerning the reasons for buying the Mustang been asked. Finally, a more concrete analysis of the relationship between (1) dogmatism and homophily, and (2) dogmatism and Opinion leadership would have been possible had the study been expanded to include those individuals who allegedly interacted with and/or were influenced by the respondents. Comparing Laboratory Research with Field Studies Although it is discouraging to find that Rokeach's theory was not supported in this field situation, it is not altogether surprising. Discrepancies between the results of experimental laboratory research and field studies have frequently appeared in the literature. For example, Festinger cited three field studies on attitude change where the induced attitude change had little, if any, effect on the subjects' subsequent behavior.9 In discussing the findings reported in the 9Leon Festinger, "Behavioral Support for Opinion Change," ,Ppplig Opin$pn Ouepteply, Vol. 28, No. 3, Fall, 1964, pp. 404-417. 104 three studies,10 Festinger commented that such change effected in the laboratory setting may be unstable unless a change in the subject's environment can be similarly effected. He implied that the environ- mental factors which led to the formation and support of the initial Opinion will continue to operate in such a way as to nullify the attitude change which took place in the laboratory. Several factors, described by Hovland,11 frequently produce divergent findings between laboratory experiments and field studies. Several of these conditions can be identified in the present investiga- tion, as Rokeach's theoretical tenets were taken from the laboratory into a field setting. In the first place it should be noted that the majority of dogmatism studies performed in the laboratory dealt with fundamental issues on which subjects have rather strong opinions. The present research, on the other hand, was concerned with peripheral beliefs which are probably less important and the more readily changed. Further- more, it is recognized that the decision to purchase an automobile is 1°11. Maccoby, at .a_1.. W Pepiods: _13 Seeking pup Accepting Information, Paris--Stanford Studies in Communication, Institute for Communication Research, Stanford, California, 1962, C. Fleischman, E. Harris, and H. Burtt, Leadership ppp Supervision ‘gp Industpy: 'pp_Evaluati n ping Supervisory Training Program, Bureau of Educational Research, Ohio State University, Columbus, Ohio, 1955, and I. Janis and S. Feshback, "Effects of Fear Arousing Communication," Journal pg Abnormal ppO_Soeip; Psycholpgy, Vol. 48, 1953, pp. 78-92. 11Carl I. Hovland, "Reconciling Conflicting Results Derived from Experimental and Survey Studies of Attitude Change," American Psychologist, Vol. 14, 1959, pp. 8-17. 105 based upon a multiplicity of influences which possibly exerted a greater degree of influence than any given intrapersonal considera- tion. Obviously, the influence of these intervening variables could have been more tightly controlled in a laboratory situation. A third difference pertains to the population and the subjects included in the present analysis. Each of the respondents had made an overt adOption decision before being interviewed. It is reasonable to believe that the purchase of the Mustang may have predisposed the post-adoptive behavior of the subjects in various ways. Admittedly, a diffusion study which is designed in the form of a longitudinal field experiment may generate findings more closely , akin to those reported in a laboratory. Nevertheless, the present investigation, conducted in a rather naturalistic situation, clearly found the personality variable of dogmatism inadequate for making predictions about (1) utilization of communication sources; (2) mass media channel attendance; (3) innovativeness; (4) product loyalty; (5) opinion leadership; and (6) homophily among subjects who have adOpted the innovation. Personality Variables in Perspective One of the primary reasons for focusing attention on a personal- ity variable is because the relationship obtained may prove useful in other research areas. A study which focuses on a personality variable actually centers principal attention on people, and only on other variables in an almost incidental way. Thus, if a personal- ity variable correlates with a certain type of behavior regarding one 106 innovation, it is reasonable to hypothesize that individuals poss- essing the particular trait will generally exhibit similar behavior when confronted with a different innovation. It was hOped that dogmatism may have been a predictor variable which would have helped in the p priori identification of innovators, core customers, and Opinion leaders. Unfortunately, the results reveal that dogmatism is a very poor predictor of such behavior. One might ask how the findings of this study compared with the results of other studies which attempted to link personality variables with overt behavior. The majority of such findings were derived from laboratory experiments. Hovland and Janis,12 based upon an extensive review of the literature, concluded that there is evidence of an attribute of personality which is independent of both the appeals and the subject matter, and suggest that authoritarianism is related to the acceptance of persuasive communication. A limited number of diffusion studies attempted to relate per- sonality characteristics with various aspects of communicative and adoptive behavior. The results are summarized by Rogers. Rogers (1957b) found that more innovative farmers scored lower on a dogmatism scale and on a rigidity scale. Copp (1956) found in a study of Kansas cattlemen that innovators had greater mental flexibility than laggards. Dean and others (1958), Emery and Oeser (1958), Bemiller (1960), and Coughenour (1960b) found that more innovative farmers utilized more rational means to reach their goals. 12Carl I. Hovland and Irving L. Janis (eds.), Personplity and Persuasibility, Yale University Press, New Haven, Conn., 1959. 107 Burdge (1961), Goldstein and Eichorn (1961), Rogers and Burdge (1962), and Copp (1956) concluded that laggards were relatively more work-oriented, that is, they viewed work as a goal in itself rather than as a means to Other/ends. Sutherland (1959) found that laggard cotton- spinning firms regarded the future only in terms of the short run, and claimed the best poificy was to simply hang on to the ideas presently on hand. More recently, Jamias and Troldahl reported a significant relationship between dogmatism and rate of adoption, and showed further that the social system's "value for innovativeness" affected the behavior of high dogmatics more than their less dogmatic counterparts.14 With the one exception noted above, the great majority of the studies which have focused on the relationship between personality variables and overt behavior have produced very low relationships similar to those found in the present investigation. Implications for Future Research At the present time, one can conclude that dogmatism as measured in the present study is of little value for predicting communicative and adoptive behavior. Nevertheless, the relationships between dogmatism and product loyalty and between dogmatism and opinion leader- ship might be explored further. Analyses pertaining to product loyalty should seek data concerning why consumers change from one product to l3Everett M. Rogers, Diffusion pf Innovations, pp. cit., p. 178. l4Juan F. Jamias and Verling C. Troldahl, "Dogmatism, Tradition, and General Innovativeness," unpublished manuscript, Department of Communication, Michigan State University, 1965. 108 to another. In order to determine whether purchase behavior agrees with expressed purchase intentions, longitudinal studies should be conducted. Investigations focusing on the relation- ship between dogmatism and Opinion leadership should include a consideration of both the Opinion giver and the party allegedly being influenced; due consideration should likewise be given to the gre- gariousness of both high dogmatic and low dogmatic individuals. Because the issues considered in the present study are of importance to marketing theorists and practitioners, the search for a variable which will aid in predicting and explaining innovative- ness, product loyalty, and opinion leadership in general should be continued. The search, if it is to be meaningful, must begin with a strong theoretical framework. The concepts should initially be tested in laboratory situations and then carried into the field for empirical validation. BIBLIOGRAPHY BIBLIOGRAPHY A. BOOKS Adorno, T. W., Else Frenkel-Brunswik,Daniel J. Levinson and R. Nevitt Sanford. The Authoritarian Personality. New York: Harper and Brothers, 1950. Backstrom, Charles H. and Gerald D. Hursh. Survey Research. Chicago: University of Illinois Press, 1962. Barnett, H. 0. Innovation. New York: McGraw-Hill Book Company, 1953. Berg, Thomas L. and Abe Shuchman (eds.). Product Strategy and Manage- ment. New York: Holt, Rinehart and Winston, 1963. Berlo, David K. The Process pg Communication. New York: Holt, Rine- hart and Winston, 1960. Bliss, Perry (ed.). Marketing and the Behavioral Sciences. Boston: Allyn and Bacon, 1963. Brown, Roger, Eugene Galanter, Eckhard H. Hess, and George Mandler. New Oirections Op Psychology. New York: Holt, Rinehart and Winston, 1962. Cohen, Arthur R. Attitude Changp_and Social Influence. New York: Basic Books, Inc., 1964. Dixon, Wilfrid J. and Frank J. Massey, Jr. Introduction pp Statistical Anal sis, second edition. New York: McGraw-Hill Book Company, 1957. Festinger, Leon. .A,Theopy.p£_Cogn;tive Dissonance. Evanston, Illinois: Row, Peterson, and Co., 1957. Festinger, Leon, Stanley Schachter, and Kurt Back. Social Pressures.;p Informal Groups: IA_Stgdy.p§_Human Eeptors $3 Housing. New York: -Harper and Row Publishers, 1950. Gottman, Jean. Megalopolis; the Urbepized Northeastern Seaboepg p: the United States. New York: The Twentieth Century Fund, 1961. Field, George A., John Douglas, and Laurence X. Tarpey. Marketipg Manpgement e Behaviore; Systems Approach. Columbus, Ohio: Charles E. Merrill Books, Inc., 1966. Hollander, E. P. and Raymond S. Hunt. Current Perspectives $3 Social Psychology. New York: Oxford University Press, 1963. lll Katz, Elihu and Paul F. Lazarsfeld. Personal Influence. New York: The Free Press of Glencoe, 1955. Lazarsfeld, Paul F. , Bernard Berelson, and Hazel Gaudet. The People's Choice (second edition). New York: Columbia University Press, 1948. Laharsfeld, Paul F. and Robert K. Merton. "Friendship as Social Process: A Substantive and Methodological Analysis," in M. Berger, T. Abel, and C. H. Page (eds.) Freedom and Control ;p_Modern Society. New York: D. Van Norstrand, 1954, pp. 18-66. Lazer, William and Eugene Kelley (eds.). Manfiagerial Mprketing: Perepectives and Viewpoints. Homewood, Illinois: Richard D. Irwin, Inc., 1962. Lionberger, Herbert F. Adoption of New Ideas and Practices. Ames Iowa: Iowa State University Press, 1960. Maccadby, Eleanor E., Theodore M. Newcomb, and Eugene L. Hartley (eds.), Readings $3 Social Psychology. New York: Holt, Rinehart and Winston, Inc., 1958. lflflNemnar, Quinn. Psycholpgical Statistics, third edition. New York: John Wiley and Sons, Inc., 1962. Merton, Robert K. Socia_1 Theogy and Social Structure. New York: The Free Press of Glencoe, 1957. Merton, Robert K., Leonard Broom, and Leonard S. Cottrell, Jr. Sociology Today. New York: Basic Books, Inc., 1959. Miller, Gerald R., and Milton Rokeach. "Individual Differences and Tolerance for Inconsistency," in P. H. Tannenbaum, R. P. Abelson, E. Aronson, W. J. McGuire, T. M. Newcomb, and M. J. Rosenberg (eds.). Theories_p§ Cognitive Consistenyy. Chicago: Randh MCNally, in press. Mueller, Eve. "The Desire for Innovations in Household Goods," in Lincoln H. Clark (ed.). Consumer Behavior. New York: Harper and Brothers, 1958. ‘.Nielsen,cA. 'C., Jr. ."Consumer Product Acceptan.cex;Rates,'"; in'1Lincoln Hi Clark (ed;). Consumer Behavior. New York: Harper and Row, 1958. Roberts, W. Rhys, "Rhetorica," in W. D. Ross (ed.). The Works p_f_ Aristotle. New York: Oxford University Press, 1946. R"gel-s, Everett M. Social Change _ip Rural Society. New York: Appleton, Century, and Crofts, 1960. 112 Rogers, Everett M. The Diffusion pg Innovations- New York: The Free Press of Glencoe, 1962. Rogers, Everett M. with F. Floyd Shoemaker. Diffusion pf: Innovptions: O‘Cross-Cultural and Communication Approach.. New York: The Free Press of Glencoe, 1967, in process. Rokeach, Milton. The OJen and Closed Mind. New. York: Basic Books, Inc., 1960. . "The Nature and Meaning of Dogmatism," in E. P. Hollander and Raymond G. Hunt (eds.). Current Perspectives_ip Social Psychology. New York: Oxford University Press, 1963, pp. 162-172. Schralmn, Wilbur. "How Communication Works," in Wilbur Schramm (ed.). Process and Effects p£_Mass Communication. Urbana, Illinois: University of Illinois Press, 1954, pp. 3-26. Secord, Paul F. and Carl W. Backman, Social Psychology. New York: McGraw-Hill Book Company, 1964. Shannon, Claude and Warren Weaver. The Mathematical Theory pt Communi- cation. Urbana, Illinois: University of Illinois Press, 1949. Staudt, Thomas A. and Donald A. Taylor. _A Managerial Introduction 3; Marketing. Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1965. Walker, Helen M. and Joseph Lev. Statistical Inference. New York: Holt, Rinehart and Winston, 1953. Zeltman, Gerald. Marketing: Contributions from the Bepgviopa; Sciences. New York:. Harcourt, Brace and World, Inc., 1965. B. ADDRESSES,.BULLETINS, AND PUBLISHED RESEARCH REPORTS 3881, George M. and Everett M. Rogers. The Adoption p_f_ Tw0 Farm Prac- tices.;pup Central Iowa Community. Iowa.Agricultura1 and Home Economics Experiment Station, Special Report 26. Ames, Iowa: 1960. Bell, William E. "Consumer Innovators: A Unique Market for New- ness," in Stephen.A.Greyser (ed). Toward Scientific Marketipg. Proceedings of the Winter Conference of the American Marketing Association. Boston: Dec. 27-28, 1963, pp. 85-95. Copp, James H. Personal and Social Factors Associated With the Adoption ng Recommended Farm Practices Amopg Cattlemen. Agricultural Experiment Station, Technical Bulletin 83. Manhattan, Kansas: 1956. 113 Cox, Donald F. "The Audience as Communicators," in Stephen A. Greyser (ed.). Toward Scientific Marketipg.. Proceedings of the Winter Conference of the American Marketing Association. Boston: Dec. 27-28, 1963, pp. 58-72. Crane, Edgar. "Communications: A Marketing Analysis," in Peter D. Bennett (ed.). Marketipg and Economic Development. Proceedings of the 1965 Fall Conference of the American Marketing Associ- ation. Washington: Sept. 1-3, 1965, pp. 400-410. Diffusion Documents Center, Michigan State University, East Lansing, Michigan. Deutschman, Paul J., and Orlando Fals Borda. Communication and Adoption Patterns Op.pp Andean Village. San Jose, Costa Rica: P. I. I. P., 1962. Emery, F. E., and O. A. Oeser. Informption, Decision and Action: .5 Study.p£ the Peychological Determinants pg ChapgesILO Farming Techniques. New York: Cambridge University Press, 1958. Frank, Ronald E., and William F. Massy. "Innovation and Brand Choice: The Folger's Invasion," in Stephen A. Greyser (ed.). Toward Scientific Marketipg. Proceedings of the Winter Conference of the American Marketing Association. Boston: Dec. 27-28, 1963, pp. 96-107. Frank, Ronald E., William F. Massy, and Donald G. Morrison. "The Deter- minants of Innovative Behavior With Respect to a Branded, Frequently Purchased Food Product," in L. George Smith (ed.). Reflections pp_Progress ip_Marketing. Proceedings of the 1964 Educators' Conference, American Marketing Association. Chicago: Dec. 38-39, 1964, pp. 312-323. Gedalecia, Ben. The Communicators: .pp’All-Media Study. 3rd Annual Conference Report, Advertising Research Foundation, New York: Nov., 1957. Hunt, Martin F., Jr., and Gerald R. Miller. "Open and Closed-Mindedness, Belief-Discrepant Communication Behavior, and Tolerance for Cognitive Inconsistency." Paper presented at the convention of the Speech Association of America. New York: 1965. "Interurbia: The Changing Face of America." Memo of the J. Walter Thompson Company. New York: May 10, 1960. King, Charles W. "Fashion Adoption: A Rebuttal to the ”Trickle Down' Theory," in Stephen A. Greyser (ed.). Towarg Scientific Market- 'ipg. Proceedings of the American Marketing Association.Confer- ence. Boston: Dec. 27-28, 1963, pp. 108-125. 114 King, Charles W. "The Innovator in the Fashion-Adoption Process," in L. George Smith (ed.). Reflections pp Progress ;p_Marketing. Proceedings of the 1964 Educators Conference, American Marketing Association. Chicago: Dec. 28—29, 1964, pp. 324-339. . "Communicating With the Innovator in the Fashion Adoption Process," in Peter D. Bennett (ed). Proceedings of the 1965 Fall Conference, American Marketing Association. Washington: Sept. 1-3, 1965, pp. 425-439. Klein, Thomas. "The Effect of a Newspaper Strike on Retail Sales and Advertising," Business Research Center, University of Toledo, F Toledo, Ohio, in process. Kreitlow, Burton W., and James A. Duncan. The Acceptance p£_Educational Programs Lg Rural Wisconsin. Wisconsin Experiment Station, Bulletin, 525. Madison, Wisconsin: 1956. Lionberger, Herbert F., and C. Milton Coughenour. Social Structure and Diffusion.p§ Farm Information. Agricultural Experiment Research, Bulletin 631. Columbia, Missouri: 1957. Lipstein, Benjamin. "The Dynamics of Brand Loyalty and Brand 'Switch— ing.'" Proceedings: 5th Annual Conference, Advertising Research Foundation. New York: Sept., 1959, pp. 101-108. Meloan, Taylor W. "New Products-—Keys to Corporate Growth." An address presented at the Winter Conference of the American Marketing Association. St. Louis, Missouri: Dec. 28-30, 1960. Nicosia, Francesco M. "Opinion Leadership and the Flow of Communi- cation: Some Problems and Prospects," in L. George Smith (ed.). Reflections_pp Progress 12 Marketing. Proceedings of 1964 Educators Conference, American Marketing Association. Chicago: Dec. 28-29, pp. 340-358. Opinion Research Corporation, America's Tastemakers: .A New Strategy for Predicting Change $3 Consumer Behavior. Princeton, New Jersey: April, 1959. Photiadis, John D. Contacts with Agricultural pgents. South Dakota Agricultural Experiment Station, Bulletin 493. Brookings, South Dakota: 1961. Rogers, Everett M. Characteristics p§_égricultural Innovptors and Other AdOpter Categories. Agricultural Experiment Station, Research Bulletin 882. Wooster, Ohio: 1961. 115 Rogers, Everett M. Bib;iography pt Research pp the ngfusion.p§ ‘ Innovations., East Lansing, Michigan: Michigan State Univer- sity, July, 1966. Rogers, Everett M. and Rabel J. Burdge. Muck Vegetable Growers: DiffusiOn.p£_Innovations Among Specialized Eprmers. Ohio Agri- cultural Experiment Station, Research Circular 94. Wooster, Ohio: 1961. Rogers, Everett M., and John Harp. "Personality Correlates of the Adoption of Technological Practics," Paper presented to the Midwest Sociological Society. Des Moines, Iowa: 1957. Rogers, Everett M., and J. David Stanfield. "Adoption and Diffusion of New Products: Emerging Generalizations and Hypotheses." Paper presented at the Conference on the Application of Sciences to Marketing Management, Purdue University. W. Lafayette, Indiana: July, 1966. Rogers, Everett M., and Rabel J. Burdge. ‘Oppmunity Norms, Opiniop Leadershi , ppg Innovatgveness Amopg Truck Growers. Ohio Agricultural Experiment Station Research Bulletin. Wooster, Ohio: 1962. Sizer, Leonard M., and Ward F. Porter. The Relation pg Knowledge_pp Adoptionng Recommended Practices. .West Virginia Agricultural Experiment Station, Bulletin 446. Morgantown, West'Virginia: 1960. Troldahl, Verling C. The Communication.p£_Horticulture Information and Influence.;pqp Suburban Community. .Communication Research Center, Report NO. 10. Boston: Boston University, March, 1963. Yellow Pages National Usage Study. Audits and Survey Company, Inc., 1964. ' Yeracaris, Constantine A. "Social Factors Associated with the Accep- tance of Medical Innovations." Paper presented at the American Sociological Association. St. Louis, Mo.: 1961. ' C. ARTICLES FROM JOURNALS AND PERIODICALS Barber, Bernard and Lyle S. Lobel. "Fashions in Women's Clothes and the American Social System," Social Forces, 31:124-131, December, 1952. ‘ Belcher, John C. "Acceptance of the Salk Polio Vaccine," Rural Socio- logy, 23:518-170, 1958. 116 Bowers, Raymond V. "Differential Intensity of Intro-Societal Diffu- sion," American Sociological Review, 3:21-31, 1938. Brooks, Robert 0., Jr. "'Word-of-Mouth' Advertising in Selling New Products," Journa1,p£_Marketi , 22:154-161, October, 1957. Chaparro, Alvaro. "Soziale Aspekte des Kulturellen Wandels: Die Diffusion neuer Techniken in der Landwertschaft," Kolner Zeitschrift.§. Soziologie and Sozial-peychologie, 8:567-594, 1956. "Cities as Long as Highways--That's America of the Future," .OOO. News and Worlg Report, 25:27-31, April 5, 1957. Cohen, Arthur R. "Attitudinal Consequences of Dissonance," Public Opinion Quarterly, 24:297-318, Summer, 1960. Coleman, James S. "Relational Analysis: The Study of Social 0188n128t10ns with Survey Methods," Human Organization, 17: 4, 28-36, Winter, 1958-59. Coughenour, C. Milton. "The Functioning of Farmers' Characteristics in Relation to Contact with Media and Practice Adoption," Rural Sociology, 25:183-297, 1960. Cunningham, Ross M. "Brand Loyalty--What, Where, How Much?" Harvard Business Review, 34:116-128, January-February, 1956. Draper, James E., and Larry H. Nolin. "A Markov Chain Analysis of Brand Preferences," Journal.p§ Advertising Research, 24:33-39, September, 1964. Ehrlich, D., I Suttman, P. Schonbach, and J. Mills. "Post—decision Exposure to Relevant Information," Journal p§_Abnormal and Socie; PeycholOgy, 54:98-102, 1957. Ehrlick, Howard J. "Dogmatism and Learning," Journal.p£ Abnormal and Social Psychology, 69:148-149, January, 1961. Fallers, Lloyd A. "A Note on the 'Trickle Effect,'" Public Opinion Oparterly, 18:314-321, Fall, 1954. Festinger, Leon. "Behavioral Support for Opinion Change," Public Opinion Ouarterly, 28:404-417, Fall, 1964. Fliegel, Frederick C. "A Multiple Correlation Analysis of Factors Associated with Adoption of Farm Practices," Rural SociolOgy, 21:284-292, 1956. 117 Foulkes, D., and S. H. Foulkes. "Self-Concept, Dogmatism, and Toler- ance of Trait Inconsistency," Journal_p§ Personality and Social Psychology, 2:249—252, 1967. Fourt, Louis A., and Joseph W. Woodlock. "Early Prediction of Market Success for New Grocery Products," Journal p§_Marketin , 25:31-38, October, 1960. Graham, Saxon. "Class and Conservatism in the Adoption of Innovations," Human Relations, vol.IX, No. 1, pp. 91-100, 1956. Hovland, Carl 1. "Reconciling Conflicting Results Derived from Experimental and Survey Studies of Attitude Change," American Psychologist, 14:8-17, January, 1959. Katz, Elihu. "The Two-Step Flow of Communication: An Up-to-date Report on an Hypothesis," Public Opinion Ouapperly, pp. 61-78, Spring, 1958. Kleck, R. E., and J. Wheaton. "Dogmatism and Responses to Opinion Consistent and Opinion Inconsistent Information," Journal‘pg Personality and Social PsychologY. 5:249-253, 1967. Klein, L. R., and J. B. Lansing. "Decisions to Purchase Consumer Durable Goods," Journal pg Marketin , 20:109-132, October, 1955. KOponen, Arthur. "Personality Characteristics of Purchasers," Journal .9; Advertising Research, 1:6-12, September, 1960. Kuehn, Alfred A. "Consumer Brand Choice as a Learning Process," Journal of Advertising Research, 2:10.13, December, 1962. Lionberger, Herbert F. "Some Characteristics of Farm Operators Sought as Sources of Farm Innovation in a Missouri Community," Rural Sociolo , 18:327-330, 1953. Lowry, Sheldon, and others. "Factors Associated with the Acceptance of Health Care Practices Among Rural Families," Rural Sociology, 23:198-202, 1958. "Marsh, C. Paul and A. Lee Coleman. "Differential Communication among Farmers in a Kentucky County," Rural Sociology, 20:93-101, 1955. Marsh, C. Paul and A. Lee Coleman. "The Relation of Farmer Character- istics to the Afloption of Recommended Farm Practices," Rural Sociology, 20:289—296, 1955. May, Frederick. "Buying Behavior: Some Research Findings," Journal ,9; Business, 38:379—396, October, 1965. 118 3Mills, Harlan D. "Dynamics of New Product Campaigns," Journal-pg Marketin , 28:60-63, October, 1964. North, Cecil C., and Paul Hatt. "Jobs and Occupations: A Popular Evaluation," Opinion News, pp. 3-13, September, 1947. Powell, Frederic A. "Open and Closed-Mindedness and the Ability to Differentiate Source and Message," Journal 9; Abnormal and Socie; Psychology, 65:61-64, 1962. Pessemier, Edgar A. "A New Way to Determine Buying Decisions," Journal pg Marketin , 24:41-46, October, 1959. Precker, Joseph A. "Similarity of Valuings as a Factor in Selection of Peers and Near-Authority Figures," Journal.p§ Abnormal and Social Psycholo , 47:406-414, 1952. Robertson, Thomas S. "The Process of Innovation and the Diffusion of Innovation," Journal p§_Marketing, 31:14-19, January, 1967. Rogers, Everett M. "A Note on Innovators," ourna;.p§ Farm Economics, 41:132—134, 1959. . "Personality Correlates of the Adoption of Technological Practices," Rural Sociology, 22:267-268, 1957. Rogers, Everett M, and George M. Beal. "The Importance of Personal Influence in the Adoption of Technological Changes," Social Forces, 36:329-335, 1958. Rosenberg, Milton J. "A Structural Theory of Attitude Dynamics," Public Opinion Oparterly, 24:319—340, Summer, 1960. Ryan, Bryce- and Neal C. Cross. "The Diffusion of Hybrid Seed Corn in Two Iowa Communitiesi'Rural Sociology, 8:15-24, 1943. Shaw, Steven J- "Behavioral Science Offers Fresh Insights on New Product Acceptance," Journal.p§.Marketipg, 29:9-13, January, 1965. "Sprawling 'Strip-Cities'--They're All Over U. 8.," O, O, News and World Report, 51:73-78, September, 18, 1961. Troldahl, Verling C., and Frederic A. Powell. "A Short-Form Dogma- tism Scale for Use in Field Studies," Social Forces, 44:211—214, December, 1965. Troldahl, Verling C., and Robert Van Dam. "Face to Face Communication About Major News Topics," Public Opinion Quarterl , 29:626-634, Winter, 1965-66. 119 Tucker, W. T. "The Development of Brand Loyalty,‘ ing, 31:32-35, August, 1964. Journal pg Market- Van den Ban, A. W. "Some characteristics of progressive Farmers in the Netherlands," Rural Sociology, 22:205-212, 1957. ‘Wasson, Chester R. "What is 'New' About a New Product?," Journa1.p§ Marketing, 25:52-56, July, 1960. Walters, S. George, Morris L. Sweet, and Max D. Snider. "When Industry Moves to Interurbia," Sales Management, 82:65-67, February 20, 1959. Whyte, William H., Jr. "The Web of Word of Mouth," Fortune, 50:140- 144f, November, 1954. Wilkening, Eugene A. "Informal Leaders and Innovators in Farm Prac- tices," Rural Sociology, 17:272-275, 1952. Robert B. Zajonc. "Balance, Congruity, and Dissonance," Public Opinion Ouarterly, 24:280-296, Summer, 1960. D. UNPUBLISHED MATERIAL Bell, William E. Consumer Innovation: ‘Ap Investigation pg Selected Characteristics pg Innovators. Unpublished Ph.D. thesis. Department of Marketing and Transportation and Administration, East Lansing, Michigan: Michigan State University, 1962. Childs, John W. .5 Study pi the Belief Systems pp Administrators-ppp Teachers ;p_lnnovative.ppp Non-Innovative Schop;,Distr;pts. Unpublished Ph.D. thesis, College of Education, East Lansing, Michigan: Michigan State University, 1965. Chou, Teresa M. Homophily lp_Interaction Patternslgp theIngfusion pghlpnovations.;p Columbipp Villages. Unpublished M.A. thesis, Department of Communication, East Lansing, Michigan: Michigan State University, 1966. Holmes, John H. "Marketing in Megalopolis." Unpublished term report for Marketing 857, Michigan State University, East Lansing, Michigan: Spring, 1963. 120 Hudspeth, DeLayne R. .5 Study pg Belief Systems.and.Acceptance.pg New Educational Media With Users and Non Users p§_AudiO Visual Graphics. Unpublished Ph.D. thesis, College of Education, East Lansing, Michigan: Michigan State University, 1965. Jamias, Juan F. The Effects'pg Belief System Styles ppuppp Communi- cation ppg Adoption p: Farm Practices. Unpublished Ph.D. thesis, Department of Communication, East Lansing, Michigan: Michigan State University, 1964. Jones, Gwyn E. ‘Epptors Affectigg the Adoption pg Opp Farm Practices, with gprticular Reference-pp Central Wales ppguppp East flip- lands p; Epgland. B. Litt. Thesis, Oxford, England: Oxford University, 1960. Rogers, Everett M. .A,Conceptual Variable Analysis pg Technological Change. Ph.D. thesis, Ames, Iowa: Iowa State University, 1957. Shoemaker, F. Floyd. "Personality Dimensions of Innovativenss," unpublished term report for Psychology 936, Michigan State Uni- versity, East Lansing, Michigan: Winter, 1966. . "A Reconceptualization of a Process," unpublished term report for Communication 470, Michigan State University, East Lansing, Michigan: Spring, 1966. Troldahl, Verling C. Mediated Communication and Personal Influence: ,§.§$eld,Experiment.. Ph.D. thesis, Minneapolis: University of Minnesota, 1963. Warland, Rex H. Personal Influence: The Degree pg Similarity pnghose Who Interact. Unpublished M.S. thesis, Ames, Iowa: Iowa State University, 1963. Wrenn, R: L; The Resolution_pf Cognitive Dissonance_$p Open and Closed _Oelief Systems. Unpublished Ph.D. dissertation, Athens, Ohio: Ohio University, 1962. APPENDICES APPENDIX A Telephone Instruction Form 123 Appendix A Telephone Instruction Form ‘ __n‘ Follow this format in contacting individuals by telephonefiand arrang- in ,personal interviews. Hello: i May I please speak to the person in your home who owns and drives the Mustang automobile? My name is Mrs. Holmes. My husband, John Holmes, is.a professor in marketing at Bowling Green State University. At the present time he is engaged in a research project studying people who bought a new Mustang in the past two and a half years. My records indicate that you bought a Mustang during that period of time, is that correct? Did you or some other member of your family make the actual purchase decision? Are you the principal user of the car? In the_event that the person who made the decision is someone other than the principal user, thank the person for 'his time and terminate the interview.> Make certain that you talk to the person who (1) made the buying decision ppO_(2) is the principal user. If you find your- self talking with a second party, it will probably be necessary for you to repeat the aboveyparagraph. I am calling because I would like to include you in this study. I wonder if you would be willing to answer some questions for a stu- dent from Bowling Green who could come to your home at your conven- ience. The interview will take only twenty-five minutes of your time. Set up a definite time, date, and place for the interview and record 'this information on the followipg page. Thank you for your cooperation, Mr. , and I hope you will enjoy the interview with Mr. (Mrs.) , who will be cal- ling on you at a.m. (p.m.) (day) , (date) . Goodibye Mr. . APPENDIX B Interview Schedule Form 125 Appendix B Interview Schedule Form bleep code MUSTANG DIFFUSION STUDY ' Subject DEPARTMENT OF COMMUNICATION number MICHIGAN STATE UNIVERSITY letter Res idence of Person Interviewed Address Phone Date of Personal Interview Time In terviewer Sp ecial Instructions 3L2} 126 3 209 Project number 5 Deck number 8 Subject number 10 0; Card number 14 1. 15 Score: 0 - if made by Ford Motor Company 1 - if made by another manufacturer When did you buy your Mustang? 00 - this month (November 1966) 01 - one month ago (October 1966) 02 - two months ago (September 1966) 27 - twenty-seven months ago (July 1964) L .97 - no response 98 - don't know Did you sell or tradeiin another car at or about‘ time you bought the Mustang? yes 'If no, go t J no 'question #4 _ What make was it? no response don't know question not asked Thinking back, could you tell me where or from whom you first became aware of the existence Of the Mus- tang? e - cOmmercial source advertising contact with dealer and/or salesman n - non-commercial source immediate family other‘relatives co-workers friends and/or neighbors publicity made up own mind observed on street rode in or drove someone else's Mustang? x - no response don't know 0 l N I 16 17 127 5. Again thinking back, could you tell me where or from whom you first became interested in the Mustang? Probe to determine similarity between answer given here and the answer given to the preceding questiOn.. c - commercial source 'advertising contact with dealer and/or salesman n - non-commercial source a immediate family other relatives co-workers friends and/or neighbors publicity o - made up own mind Observed on street rode in or drive someone else's Mustang x - no response 2 - don't know 6. Which source of information do you believe was the most'gpfluential in your purchase decision. Probe to determine similarity between answer given here and the answers given to the two preceding? questions. c - commercial source advertising contact with dealer and/or salesman n - non-commercial source immediate family other relatives Co-workers friends and/or neighbors publicity o - made up own mind observed on street rode in or drove someone else's Mustang x - no response 2 - don't know Score: 0 - all different 2 - three alike 1 - two alike 7 - no response 8 - don't know 7. Were.y0u the first person in your immediate neighbor- hood to own a Mustang? 0— 1 2 7 8 no yes one of the first no response.. don't know 128 18 8. Were you the first person among your circle of friends to own a Mustang? 0 - no 7 - no response 1 - yes 8 - don't know 2 - one of the first Now I am going to ask you some questions about your newspaper and magazine reading habits and your television and radio listening habits. I would like to remind you at this time that the Toledo newspapers are on strike. 9. How much time would you estimate that you spend per day reading the newspapers in an average week? 00 - none 01 - between % and 1% hours Ly_ . 02 - between 1% and 2% hours onvert answers to a% 03 - between 2% and 3% hours er week basis befor 04 - between 3% and 4% hours ecordi . . 21 - between 20% and 21% hours 97 98 no response 4 if 00, 97, 98, gpl don't know to guestion #13._ 10. How many different newspapers (hi you read in an average week? 01 - one 02 - two 03 - three 04 - four 05 - five 16 - sixteen 97 - no response 98 - don't know 99 - question not asked 11. As you paged through these newspapers last-week, did you notice any Mustang advertisements? no if no, go :3 yes uestion #1 12. How many do you recall? 01 - one 97 - no response 02 - two 98 - don't know 03 - three 99 - question not asked 04 - four 16 - sixteen 129 25 - 26 .13. During an average week, how much time would you estimate that you spend reading magazines? 00 - none 01 - between % and 1% hrs. 02 - between 1% and 2% hrs. 03 - between 2% and 2% hrs. 04 - between 3% and 4% hrs. 24 - between 23% and 24% hrs. _ FfOO,97,98gL 97 - no reSponse 'to question #17. 98'- don't know rm- 27 - 28 14. How many different magazines do jyou read in an average week? 01 - one 02 - two 03 - three 97 - no response by 98 - don't know 99 - question not asked 15. As you paged through these magazines last week, did you notice any advertisement for the Mustang? no hf no go to question #ifl yes 29 - 30 116. How.many can you recall? 01 - one 02 - two 03 - three 04 - four 12 - twelve 97 - no response 98 - don't know 99 - question not asked 130 21 - 32 17. During an average week, how much time would you estimate you spend per day listening to the radio both at home and in your car?‘ 00 - none 01 - between % and 1% hrs. 02 - between 1% and 2% hrs. _ 03 - between 25. and 315 hrs. f 00, 98, 97, go tj uestion #21. . 28 - between 27% and 28% hrs. 97 - no response Convert answers to a per A 98 - don't know week basis before recordi . 33 - 34 18. How many different stations 1k) you listen to in an average week? 01 - one 02 - two 03 - three 04 - four 10 - ten 97 - no response 98 - don't know 99 - question not asked 19. As.you listened to the radio laSt week, do you recall hearing any commercials about the Mustang? no _f no, go to J yes uestion #21._ 35 - 36 20. How many do you recall? 01 - one 02 - two 03 - three 20 - twenty 97 - no response 98 don't know 99 - question not asked 131 37 - 38 21. How much time would you estimate that you spend per day watching television in the average week? 00 - none . . 01 - between % and 1% hrs. Eonvert to a per week. 02 - between 1% and 2% hrs. asis before recordi . 03 - between 2% and 3% hrs. 24 - between 23% and 24% hrs. If 00, 97, 98 go to uestion #25. 97 98 no response don't know 39 - 40 22. How many different channels do you watch in the average week? 00 - none 97 - no response 01 - one 98 - don't know 02 - two 99 - question not asked 04 - four 05 - five 11 - eleven 23. As you watched television last week, do you recall seeing any commercials about the Mustang? AL no [If no, gp_to question #ZSJ yes 41 - 42 24. How many can you recall 01 - one 02 - two, 03 - three 15 - fifteen 97 - no response 98 - don't know 99 - question not asked Now I am going to ask you some questions about conversations you may have had with other people about your car. ‘45 - 47 43 - 44 438 - 50 25. 26. 27. 132 During the past week how many times did you talk about (your) Mustang with the following people? immediate family other relatives people you work with friends/neighbors clerks, dealer, salesmen all others Total 00 01 02 03 04 15 97 98 none one two three four fifteen If 00, 97, or 9 appears in IBM column 43-44 go to question #27. no response don't know With regard to these conversations, about what percent of the time did you begin them? 000 001 002 050 100 997 998 999 Since 000 001 002 003 004 050 100 997 998 never one percent two percent fifty percent one hundred percent no response don't know question not asked you have owned your Mustang, what percent of the peOple you talked with about the Mustang, had favorable opinions about the car before they talked with you? none one percent two percent three percent four percent fifty percent one hundred percent no response don't know 133 (Since respondent has owned Mustang) 51 - 52 28. About how many of the people you talked with about the Mustang, already owned a Mustang? 00 - none 01 - one 02 - two 03 - three 04 - four 15 - fifteen 97 - no response 98 - don't know 29. As far as you know, about how many of the people you talked with about the Mustang, who didn't own one at the time, have subsequently bought a Mustang? 00 - none 01 - one If 00, 97, or 98, gj 02 - two to question #31. 03 - three 12 - twelve 97 - no response 98 - don't know 533 - 54 30. How many of these people, in your opinion, were influenced in their decision either as a result of seeing or riding in your Mustang or by talking with you about your car? 00 - none 01 - one 02 - two 03 - three 97 no response 98 - don't know 99 - question not asked 55:5 -57 31. Excluding the members of your family, what percent of the people you talked with about the Mustang spent about the same number of years in school that you did? 000 - none 001 - one percent 002 - two percent 003 - three percent 100 - one hundred percent 997 - no response 998 don't know 58 - 60 61 - 63 64 - 66 32. 33. 34. 35. Excluding your family, what percent of the people you talked with about the Mustang live in similar 134 neighborhoods to the one you live in? 000 - 001 002 003 100 997 998 none one percent two percent three percent one hundred percent no response don't know Again excluding your family, what percent of the peOple you talked with about the Mustang have similar occupations or do about the same kind of work as you do? 000 - 001 002 003 100 - 997 - 998 - About about none one percent two percent three percent one hundred percent no response don't know what percent of the people you talked with the Mustang would you say are the kind of people who like to try new and different things? 000 - 001 - 002 - 003 - 100 - 997 - 998 - Since none one percent two percent three percent one hundred percent no response don't know you have owned your Mustang, have you allowed anyone other than the members of your family to drive no yes it? 'u'o-T'E.'l A.-‘ v. n Ff no, hand respondent the clipboard. fi 67 - 68 69 70 135 36. How many others have driven it? 00 - none 01 - one 02 - two 03 - three 04 - four 14 - fourteen 97 — no response 98 - don't know 99 - question not asked MD RESPONDENT THE cum -gf it is obvious that the respondent still has his] ri inal Musta o to question #41. j 37. Have you replaced your original Mustang? no ‘If no, go to ‘ yes uestion #40a 38. With what make of car have you replaced it. ficore: 1 0 - Mustang 1 - Ford product 2 - other make 7 - no response 8 - don't know 9 - question not asked Score: 0 - similar type (Mustang, Cougar, Camaro, Barracuda, Marlin) 1 - dissimilar type (all others) 7 - no response 8 - don't know 9 ' question not asked 39. Which body type did you buy? 71 72 73 136 40. Do you own a second car? yes If no, go to no ggestion #42. 41. What kind is it? WScore: 0 - if made by Ford Motor Company - no response don't know question not asked “>me I if made by another manufacturer 42. When you replace your Mustang with what make will you replace it? DO NOT ASK IF THE PERSON HAS ALREADY REPLACED HIS MUSTANG. Igcore: 0 - Mustang Ford product other make no response don't know question not aske \DWNNH I Score: Marlin) dissimilar type (all others) - no response don't know question not_g§ked \OmNH I 0 - similar type (Mustang, Cougar, Camaro, Barracuda, 43. Which body type would you buy? Now in conclusion I am going to ask you a few questions about yourself. 44. 45. 46. 137 Do you own or rent this house (apartment)? 0 - rent 1 - own 2 - living with parents 3 - other . If living with parents, ask 7 - no response Ifor father's occupation. 8 - don't know 'What type of work does your father do? NH: About how long have you lived at this address? 00 - lessthan six months 01 - between % and 1% yrs. 02 - between 1% and 2% yrs. 03 - between 2% and 3% yrs. If other than 00 - 05 04 - between 3% and 4% yrs. record a00 in IBM 05~- between 4% and 5% yrs. column 74-75 and go 06 - between 5% and 6% yrs. to question #48. 07 - between 6% and 7% yrs. 27 - between 26% and 27% yrs. 97 - no response 98 - don't know How many times have you moved during the last five years? 01 - once 02 - twice 03 - three times 04 - four times 12 - twelve times 97 - no response 98 - don't know 99 - question not asked 74 - 75 Card 2 1 - 3 209 138 47. Would you please indicate any other towns that you have lived in during the last five years and the approximate dates that you lived there? Towns From To #‘2_ core: Award one point for each move Award one point for each city u mentioned in the preceding 3; questions (exclude military transfer) * Total Geographical Mobility Score [Enter here and iniiBM coiumn 74 - 75 roject number eck number ubject number 02 ard-number 48. Do you work for pay outside your home? no f no, or part time, ask for part time father's occupation and record full time elow or if interviewing wife _ask for husband's occupation. 13 - 14 49. What type of work do you do? 139 (Insert father's occupation if respondent (l) is living at home, (2) is a minor, (3) is unemployed) or (insert hus- band's occupation if interviewing wife) 00 01 97 98 NH: unemployed retired no response don't know enter in IBM col. 13 - 140 50. How long have you (your father, your husband) been ' doing this kind of work? 00 - 01 - 02 - 03 - 27 - 97 r 98 - less than six months between % and 1% yrs. between 1% and 2% yrs. between 2% and 3% yrs. between 26% and 27% yrs.L_i If other than 00 - 05, no response record a 0 in column #15 don't know nd go to question #52. 51. Would you please indicate the other types of work that you (your father) have done during the peat five years? NH: NH: NH: NH: NH: 15 -§core: Award one_point for each change] 16 - 17 Score: Award the number of points which separate the 18 - 19 52. What 00 - 01 - 02 - 03 - 10 — ll - 12 - 13 - l4 - on the NH scale. in IBM column 15. two most distant occupational ranks as reported Enter a 99 if a 0 or 9 appears was the last year of school completed? none 15 - 3rd yr. of college lst grade 16 - completed college 2nd grade 17 - master's degree 3rd grade 18 - a.b.d. 2nd yr» of high 19 - doctor's degree 3rd yr.;6f high completed high school 97 lst yr. of college 98 2nd yr. of college no response don't know 140 20 - 21 53. In which year did you finish high school (complete your education)? 00 - 1966 or after (18 or younger) 01 - 1961-1965 (19-23) 02 - 1956-1960 (24-28) 03 - 1951-1955 (29-33) 04 - 1946-1950 (34-38) 05 - 1941—1945 (39—43) 06 - 1936-1940 (44-48) 07 - 1931-1935 (49—53) 08 - 1926-1930 (54—58) 09 - 1921-1925 (59-63) 10 - 1916-1920 (64-68) 11 - 1911-1915 (69-73) 12 - 1906-1910 (74-78) 13 - 1901-1905 (79-83) 14 - before 1901 (84 or older) 97 - no response 98 - don't know 22 - 23 [Score: Award poinis for education, occupationjiand addressJ Education* Key Occupation oouokar-o neighborhood. did not finish did not finish completed high some college bachelor's degree master's degree hi sc 9 - doctor's degrei *add one point 0 , respondent is over 45 yrs. old gh school hool 2, 4, if grade school Kgy,Address** 1- OWNOUI-I-‘WN I **Deduct one point if renting This ranges on a continuum ranging from 1-9. should consider both the respondent's residence and the immediate Suggested Scale: slum slum renewal - po fringe pre-war small post-war small pre-war moderate post-war large estate - prospero large estate Education 1-05--14 2—15—-24 3—25--34 4-35--44 5-45--54 6-55--64 7-65--74 8-75--84 9-85--94 fiéfifiééfififi The interviewer or farm farm us farm Occupation Address A Total Enter here and in IBM Colgg§_22 - 23 .__.-.., ~".. A .».-—————..—~.—. .fififl‘ . 24 25 - 27 141 _ex of the respondent - male - female Egst ScoreI 142 .DOGMATISM TEST Michigan State University (This and the following page would be handed to the respondent on the previously mentioned clipboard). Several statements on a variety of topics are listed below. You may find yourself agreeing strongly with some of the statements . . . disagreeing just as strongly with others . . . and perhaps uncertain about others. Whether you agree or disagree with any statement, you can be sure that many people feel the same as you do. We want your personal opinion on each statement. After you read each, please write an A or a D in the first column which tells me whether . . . in general . . . you agree or disagree. In the second column, place a 1 if you agree or disagree a little, a 2 if you agree or disagree on the whole, and a 3 if you agree or disagree very much. A or D 1,2,3 1. There are two kinds of people in this world; those who are for the truth and those who are against the truth. If 2. Most people just don't know what's good for them. 3. The highest form of government is a democracy and the highest form of democracy is a government run by those who are most intelligent. 4. Most of the ideas which get printed nowadays aren't worth the paper they are printed on. 5. In the long run the best way to live is to pick friends and associates whose tastes and beliefs are the same as one's own. 6. The present is all too often full of unhappiness. It is only the future that counts. 7. To compromise with our political opponents is dangerous because it usually leads to betrayal of our own side. 8. The main thing in life is for a person to want to do / something important. 9. Even though freedom of speech is a worthwhile goal, it is unfortunately necessary to restrict the freedom of certain political groups. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. 20. 143 Most peOple just don't give a "damn" for others. 1 It is only natural for a person to be rather fearful of the future. When it comes to differences of opinion in religion we must be careful not to compromise with those who believe differently from the way we do: Unfortunately, a good many people with whom I have discussed important social and moral problems don't really understand what's going on. A group which tolerates too much difference of Opinion among its own members cannot exist for long. H In this complicated world of ours the only way we can know what's going on is to rely on leaders or experts who can be trusted. , My blood boils whenever a person stubbornly refuses to admit he's wrong. L Of all the different philosophies which exist in the world there is probably only one which is correct. It is only when a person devotes himself to an 1 ideal or cause that life becomes meaningful. In times like these it is often necessary to be more on guard against ideas put out by people or groups in one's own camp than by those in the opposing camp. In a discussion I often find it necessary to repeat myself several times to make sure I am being understood. Score: Enter in IBM column 25 - 27 on page 16 of the appropriate Interview Schedule Form. Make certain that the subject number on this statement agrees with the subject number on the interview schedule form. APPENDIX C Letters of Introduction MICHIGAN STATE UNIVERSITY EAST LANSING . MICHIGAN 48323 COLLEGE OF COMMUNICATION ARTS - DEPARTMENT OF COMMUNICATION 0 CABLE: COMLJDBPT November 2, 1966 Dear Mustang Owner: The interviewer calling upon you at this time is associated with the Mustang Diffusion Study being conducted by Mr. John Holmes, a graduate student in Communication, at Michigan State University. This person will ask you a number of questions about your communication patterns and your Mustang at this time. All information provided by you will be held in strict con- fidence. Your name will not in any way be identified with the answers you may give. So, may we urge you to express yourself freely. If you have any questions about this interview at this time, you may contact either: Mr. John H. Holmes Dr. Everett M. Rogers Student Director Faculty Advisor 329 Union Building 320 Union Building Michigan State University Michigan State University East Lansing, Michigan East Lansing, Michigan Phone: 353-3820 Phone: 355-3480 We would like to thank you in advance for your cooperation in this research study. Yours very truly, flaw ohn H. Holmes Student Director 93%: Everet . Rogers Faculty AdVisor 145 BOWLING GREEN STATE UNIVERSITY BOWLING GREEN. OHIO 43402 November 19, 1966 Dear Mustang Owner: The young gentleman who is about to interview you is a student at Bowling Green State University. The information which you give him will be kept in strict con- fidence. The material is being compiled for my Ph.D. dissertation at Michigan State University. If you have any questions concerning the interview please feel free to contact me. My telephone number is 353-6341 in Bowling Green. Thank you very much for your time, cooperation, and assistance. Yours very truly, £25m) John H. Holmes Marketing Department 146 APPENDIX D Summary of Data Collection 148 Appendix D Summary of Data Collection Completed Interviews: "A"-60; "B"-40; "C"-24; "D"-16; "E"-lO -150 Inaccessible: no telephone listing; no answers; geographic relocations-139 Refusals: too busy; not interested - 39 Disqualified: decision maker not principal user - 12 Aborted: subject could not comprehend questions - 1 APPENDIX E Tabular Description of Sample 150 Table 8 - Age Distribution of the Respondents Ages Frequency Percentage 18 or younger 3 2.0 19-23 42 28.0 24—28 21 14.0 29-33 13 8.7 34-38 18 12.0 39-43 13 8.7 44—48 16 10.7 49-53 10 6.7 54-58 8 5.3 59-63 4 2.7 64-68 1 .7 84 or older __1 .7 Total 150 100.01 Mean age category: 29-33 Median age category: 29-33 151 Table 9 - Occupational Ranks Held by Respondents (North-Hatt Classifications Frequency Percentage 40-49 1 .7 50-59 23 15.3 60-69 66 44.0 70-79 43 28.7 80-89 14 9.3 90-99 3 2.0 Total 150 100.0% 152 Table 10 - Geographic Mobility Scores Geographic Mobility Scores? Frequency Percentage 0 77 51.32 1 32 21.3 2 20 13.3 3 6 4.0 4 4 2.7 5 5 3.3 6 2 1.3 7 3 2.0 8 ___i .7 Total 150 100.0% *One point is awarded for each change of address and for each different city in which the subject has re- sided during the five year period preceding the date 'of the interview. 153 Table 11 - Number of Hours of Exposure to the Mass Media in an Average Week Hours of Exposure Newspapers Magazines Radio Television 3 l8 4 8 1 - 5 85 103 43 14 6 - 10 53 22 39 35 11 - 15 7 4 18 36 16 - 20 1 l S 6 21~- 25 l 2 8 25 26 - 30 6 12 31 - 35 4 9 36 - 40 2 l 41 - 45 2 2 46 - 50 6 1 51 - 55 2 l 56 - 60 l 61 - 6S 3 66 - 70 4 70+ __g Total 150 150 y-d U! 0 150 154 Table 12 - Number of Mass Media Communication Channels Seen and/or Heard in An Average Week Number of Mass Media Channels Newspapers Magazines Radio Television 0 3 18 4 8 1 53 14 36 5 2 64 21 4o 45 3 23 28 41 3o 4 7 34 16 22 5 17 5 15 6 9 5 22 7 6 2 2 8 2 1 9 10 1 20 .__1 Total 150 150 150 150 155 Table 13 - Number of Mustang Commercials Recalled During the Week Preceding the Interview Number of Commercials Recalled Newspgper Magazine Radio Television 0 89 79 80 59 l- 5 52 67 52 82 6-10 ' 8 2 l3 5 ll-lS 2 2 2 16-20 1 3 21-25 1 26-30 __1_ Total 150 150 150 150 156 Table 14 - Adoption Dates for 150 Subjects Month of Adoption April 1964 May June July August September October November December January 1965 February March April May June July August September October November December January 1966 February March April May June July August September Total Median Category: June 1965 Adoptions I—‘ [—0 WNCDMO‘UI-I-‘I-‘bmwwNO\UOCDOO\O\O\-I>CDO\J>NNw-I>w 150 Percentage N NuNO‘H-fiWO‘mw-bJ-‘kaflbNI-‘I-‘NNN OWWWOUJNNNWOOWOWNUNOOONWONWWONO NI—‘UWkUJN 100.0% 157 Table 15 - Number and Frequency of Conversations about the Mustang during the week Preceding the Interviews Number of Conversations Frequency Percentage 0 19 12.7. 1 - 5 54 36.0 6 - 10 37 24.7 11 - 15 19 12.7 16 - 20 9 6.0 21 - 25 5 3.3 26 - 30 3 2.0 31 - 45 2 1.3 36 - 40 1 .7 More than 40 1 .7 Total 150 100.0% 158 Table 16 - Number of Conversations About the Mustang with People Living in Similar Neighborhoods, Having Similar Occupations, and Having Similar Educations as a Percent of Total Conversations about the Mustang Percentage Having Number Residing Number Having Number Having Similar Social in Similar Similar _Similar Characteristics Neighborhood Occupation Education None 6 3 8 01-09% 17 29 12 10-19 8 7 7 20-29 8 12 9 30-39 1 5 1 40-49 5 7 3 50-59 18 26 14 60-69 6 7 6 70-79 l6 l7 19 80-89 5 5 7 90-100 .92 _32 _6_4 Total 150 150 150 159 Table 17 - Number of Others Who Drove Respondents' Mustangs i; Number of Others Who Drove Respondents' Mustangs Frequency Percentagep 0 44 29.3% 1 - 5 68 45.3 6 - 10 26 17.3 11 - 15 8 5.3 16 - 20 1 .7 20 - 25 2 1.3 More than 25 ___i .7 Total 150 100.0% 160 Table 18 - Number of Persons Allegedly Influenced by the Respondents Number of Persons Allegedly Influenced Frequency_, Percentage 73 48.7% 1 - 3 68 45.3 4 - 6 4 2.7 7 - 9 l .7 10 - 12 3 2.0 More than 12' __i. .7 Total . 150 100.0% 161 Table 19 - Number of Other Mustang Owners Known by the Respondents Number of Other Mustang Owners Frequency Percentage 0 37 24.7% 1 - 5 94 62.7 6 - 10 12 8.0 11 - 15 3 2.0 16 - 20 1 .7 20 - 25 3 2.0 Total 150 100.0%