OPINION LEADERSHIP AND NETWORK CENTRALITY WITH RESPECT TO TEACHING INNOVATIONS WITHIN ACCOUNTING HIGHER EDUCATION Dissertation for the Degree of Ph. D. MICHIGAN STATE UNIVERSITY VINCENT FRANCIS McCORMACK 1977 I II II; IIZIIJILIIIIII III I III III IMIIIIIII I . r \.a"" i i I R 1‘ V T ' ' I; .‘ . An .' -_ . I. " r - pi - . . I? t . ‘ y I, I1] .I \ (1‘ {.1 ‘ I‘ I LJP: ' “’ i U - A l . ' {If}; E. L: . m 1 --!m= anew-mg This is to certify that the , thesis entitled 5‘ '7 u OPINION LEADERSHIP AND NETWORK CENTRALITY WITH RESPECT TO TEACHING INNOVATIONS WITHIN ACCOUNTING'HIGHER EDUCATION presented by Vincent Francis McCormack has been accepted towards fulfillment of the requirements for Ph . D. degree in Accountlng -’-r [’- I /?7/ "i 7 5' 7 ...c cc; my vv// 7,4“ { / Major professot/ 7 » x/i 1" ’/-:s7 ./ Date /uj / 57,7 u 0-7639 [mew xI-d ABSTRACT OPINION LEADERSHIP AND NETWORK CENTRALI'TY WITH RESPECT TO TEAGIING INMJVATIONS WITHIN ACCOUNTING HIGHER EDUCATION BY Vincent Francis McCormack This study applied portions of the methodologies of diffusion of innovations research and comication network analysis research to the field of university-level accounting education, in order to help bring about an understanding of the ways in which developments in teach- ing technology are disseminated among accounting educators. Since prior application of these methodologies to the context of accounting educa- tion had never been made, this research represents a pioneering, explora- tory, tentative, descriptive work. The study has attempted to provide a start toward accomplishing the long-run objective of securing maximal rates of adoption, of improvements in instructional technology, by accounting educators. The methodology employed in this research attempted to identify key relationships existing within the cammicationgctivities of depart- ments of accounting faculty with respect to teaching-related topics. Twenty dependent variables were operationalized in order to measure the extent to which individuals performed two key roles in the commmication process I Vincent Francis McCormack 1. The role of opinion leader, from diffusion of innovations research, consists of being a potential influential and fbcus of advice- seeking cannunication within the department; 2. The role of occupying a central position in a communication network-~network centrality-~consists of serving a linking function in the transmission of infbrmation between individuals in a department, and is a product of the structure of the communication network in the department. A census of all full—time, permanent, accounting faculty mem- bers from ten AACSB schools was conducted to obtain the data from which the twenty dependent, and ferty-two independent, variables were gener- ated. Although the overall response rate fer the ten schools was in ex- cess of ninety per cent, concentrations of non-respondents prohibited the calculation of dependent variable measures at two schools. .After testing for, and finding no appreciable evidence of, response bias, ninety-seven individuals from the remaining eight schools were identified as the respondent set to be analyzed. The independent variables-~categorized as biographic characteris- tics, interpersonal communication variables and mass media communication variables--were based upon generalizations from diffusion research re- garding the social status, cosmopoliteness, social participation, extent 0f change agent contact, exposure to mass media, innovativeness, and technical competence of Opinion leaders. All variables were standardized hdthin each department, resulting in sixty-two measures of relative individual differences. Vincent Francis McCormack Initially, the existence of linear relationships between all de- pendent and independent variables was tested through the use of Pearson product-moment correlation coefficients. The relationships githin the variable sets were then explored utilizing the results of principal com- ponents factor analyses with varimax rotation, with respect to each of the dependent and independent variable sets. Factor scores were calcu- lated for each of the resulting significant factors, creating twenty new factor score variables which represented the significant components of the variability within each of the original variable sets. Finally, line- ar relationships between the independent variable factor score sets, and each significant factor from the dependent variable factor score sets, were identified using the results of multiple linear progression procedures. Limitations of this research consisted of the assumption of a linear model, and the potential effect of violations of the assumption of multivariate normal distributions. The results of this study may, strictly speaking, be generalized only to the schools and individuals analyzed. Selected characteristics, of the ten departments in which the data was gathered, are presented in order to assist the reader who wishes to infer the results of this research to a specific population of interest. OPINION LEADERSHIP AND NETWORK CENTRALITY WITH RESPECT TO TEACHING INNOVATIONS WITHIN ACCOUNTING HIGHER EDUCATION By Vincent Francis McCormack A DISSERTATION Submitted to Michigan State University. in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Accounting and Financial Administration 1977 © Copyright by VINCENT FRANCIS MCCORMACK 1977 Ca Te lo Oil DEDICATION I believe most accounting graduate students enter a doctoral program in order to become teachers. Many gradu- ate students whom I have been privileged to know have ex- pressed the feeling, at least early in their careers, that they hoped teaching would be a more personal and satisfying way of providing service to humanity than at least some of the other alternatives available to individuals with train- ing in the field of accounting. Over the years, I have seen much of this idealism slowly diminish; largely, I believe, due to a reward struc- ture within higher education that all too often forces an individual to devote more and more of his efforts to activi- ties other than teaching. It is to my fellow faculty within accounting higher education that this dissertation is dedi- cated. It is my hope that the results of this and similar research in the future will enable us, in spite of ourselves, to offer our students what they so well deserve—~the best education that we can give them. iii ACKNOWLEDGEMENTS Very early in this writer's doctoral program, he promised himself that he would attempt a dissertation whose results would at least have a chance of benefitting students in accounting. This writer was extraordinarily fortunate to find an individual, in the person of Professor Gardner M. Jones, who was not only willing to chair a dissertation that was multidisciplinary in nature, and unfashionable in the current vogue of doctoral research in accounting; but who was also willing to put up with the trials and tribula- tions of this writer as he attempted to teach a thousand students a year and write this dissertation at the same time. Professor Roland F. Salmonson has been this writer's advisor throughout his doctoral program, and has provided assistance in many different ways. Professor Everett M. Rogers can be held accountable for first sparking this writer's interest in the field of communication theory and research--an:hnmnest which, in large measure, was responsi- ble for this writer continuing his doctoral program during a period in which personal difficulties made the effort barely seem worthwhile. iv Professor Richard V. Farace opened the world of communication network analysis to this doctoral student. His encouragement and patience with an accounting doctoral student attempting to assimilate a resonable core of know- ledge in a research area unknown to most accountants, has been monumental. Heartfelt appreciation is extended to all these individuals; this writer's major regret is that this dissertation is neither an accurate, nor adequate, represen- tation of their contributions. Sincere thanks are also due to many other friends-- doctoral students, faculty and staff-~at Michigan State Univer- sity, The Pennsylvania State University, and elsewhere. They have provided needed friendship, criticism, encouragement and help at innumerablezpoints in time. Rather than attempt to identify each individual, and thereby inadvertently omit persons who should be mentioned, this writer hopes that his inadequate collective thanks will be understood by all. This writer also wishes to express his appreciation to the Touche Ross Foundation for financial support during part of the period in which this dissertation was written. Most importantly, this writer would like to thank his wife, Pat. She has suffered, with this writer, through too many impoverished years, and the seemingly interminable delays and frustrations in making progress on this disserta- tion. To her, this writer expresses his love, thanks and h0pe that accomplishment of the objective makes the sacra- fices seem worthwhile. Finally, this writer can only hope that, in some later year, his now-young son, Michael, will stumble upon this obscure piece of research that has occupied so much of his father's time, and understand. TABLE OF CONTENTS Page LIST OF TABLES. . . . . . . . . . . . . . . . . . . ix LIST OF FIGURES . . . . . . . . . . . . . . . . . . xiii Chapter I. INTRODUCTION . . . . . . . . . . . . . . . 1 Overview . . . . . . 1 Nature of the Problem. . . . . 1 Prior Research and Methodological Approach of the Study . . . . . . . . . 7 Organization of the Thesis . . . . . . . l4 Footnotes to Chapter I . . . . . . . . . 15 II. GENERATION OF THE DATA BASE. . . . . . . . 22 Population and Sample. . . . . . . . . . 22 Data Gathering Procedures. . . . . . . . 26 Initial Distribution. . . . . . . . . 27 Second Distribution . . . . . . . . . 28 Dependent Variable Measures. . . . . . . 30 Opinion Leadership Indexes. . . . . . 31 Network Centrality Indexes. . . . . . 41 Independent Variable Measures. . . . . . Sl Biographic Variables. . . . . . . . . 52 Interpersonal Communication Variables. . . . . . . 59 Mass Media Communication Variables. . 62 Response Bias. . . . . . . . . . . . . 66 Independent Variables . . . . . . . . 66 Dependent Variables . . . . . . . . . 69 Data Modification Procedures . . . . . . 73 Footnotes to Chapter II. . . . . . . . . 77 III. ANALYSES OF THE DATA BASE. . . . . . . . . 84 Pearson Correlation Analysis . . . . . . 85 Biographic Variables. . . . . . . . . 85 Interpersonal Communication Variables. . . . . . . . . . . . . . 96 vii Chapter Social Participation Cosmopoliteness. . . Change Agent Contact Exposure to Mass Media. . . . . Opinion Leadership with etwork Centrality . . . . . . . . . . . Factor Analysis. . . . . . z. o o 0 Factor Analysis Procedures Employed : Biographic Variable Set . . . . . Interpersonal Communication Variable Set . . . . . . . . Mass Media Communication Vari- able Set . . . . . . . . . Teaching Innovation Dependent Variable Set . . . . . . . General Teaching Dependent Variables. . . . . . . . . Multiple Regression Analysis . . Multiple Regression Procedures. Biographic Variables. . . . . Interpersonal Communication Variables. . . . . . . . . . . Mass Media Variables. . . . . . . Footnotes to Chapter III IV. SUMMARY AND CONCLUSIONS. . . . . Summary. . . . . . . . . . . . Methodology . . . . . . . Opinion Leadership. . . . . . . . Network Centrality. . . . . . . . . Conclusions. . . . . . . . . . . . Limitations. . . . Final Note . . . . . . . Footnotes to Chapter IV. . . . . SELECTED BIBLIOGRAPHY . APPENDIX. . . . . . . . . . . . . . . . viii Page 96 101 106 109 115 125 126 133 135 141 144 148 153 154 160 164 168 173 179 179 179 192 193 195 206 208 209 210 218 Table 1. 10. 11. 12. 13. 14. LIST OF TABLES Page Summary of Descriptive Information Per- taining to Departments Who Received Questionnaires . . . . . . . . . . . . . . . 25 Binary Matrix of Figure 1 Choice Data. . . . 34 Weight Matrix of Figure 1 Choice Data. . . . 36 Directed Centrality Opinion Leadership Index Calculations for Figure 4 Choice Data. . . . . . . . . . . . . . . . . 38 Non-directed Centrality Index Calculations for Figure 6 Choice Data . . . . . . . . . . 44 Non-directed Centrality Index Frequency set weights. 0 O O O O O O O O O O I O O O I 50 Weights Used in Calculating Innovative- ness Index . . . . . . . . . . . . . . . . . 58 Mass Media Variable Awareness Codes. . . . . 64 Respondent Categories. . . . . . . . . . . . 68 Independent Variable Response Bias ReSUItSO O O O O O O O O I O O O O I O O O O 68 Dependent Variable Response Bias Results . . 70 Missing Data Cases for Independent Variable Groups. . . . . . . . . . . . . . . 74 Data Bases With and Without Missing Value Substitutions. . . . . . . . . . . . . 75 Significance Levels of Selected Pearson Product-Moment Correlation Coefficients Using a Two-Tail Test With 95 Degrees of Freedom . . . . . . . . . . . . . . . . . 86 Table 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. Page Pearson Correlations of Social Status With Opinion Leadership. . . . . . . . . . . 87 Pearson Correlations of Social Status With Teaching Innovation Network Centrality . . . . . . . . . . . . . . . . . 88 Pearson Correlations of Social Status With Combined Teaching Network Centrality O O O 0 O O O O O O O I O O O O O 89 Pearson Correlations of Technical Compe- tence With Opinion Leadership. . . . . . . . 91 Pearson Correlations of Technical Compe— tence With Teaching Innovation Network Centrality and Combined Teaching Network Centrality . . . . . . . . . . . . . 92 Pearson Correlations of Innovativeness With Opinion Leadership. . . . . . . . . . . 94 Pearson Correlations of Innovativeness With Teaching Innovation Network Centrality and Combined Teaching Network Centrality . . . . . . . . . . . . . 95 Pearson Correlations of Social Par- ticipation With Opinion Leadership . . . . . 97 Pearson Correlations of Social Par~ ticipation With Teaching Innovation Network Centrality . . . . . . . . . . . . . 99 Pearson Correlations of Social Par- ticipation With Combined Teaching Network Centrality . . . . . . . . . . . . . 100 Pearson Correlations of Cosmopoliteness With Opinion Leadership. . . . . . . . . . . 102 Pearson Correlations of Cosmopoliteness With Teaching Innovation Network centrality O O O O O O O O O 0 O O O I O O O 103 Pearson Correlations of Cosmopoliteness With Combined Teaching Network Centrality . . . . . . . . . . . . . . . . . 105 Pearson Correlations of Change Agent Contact With Opinion Leadership. . . . . . . 107 X Table 29. 30. 31. 32. 33. 34. 35. 36. 37. 38. 39. 40. 41. 42. 43. Page Pearson Correlations of Change Agent Contact With Teaching Innovation Net- work Centrality and Combined Teaching Network Centrality . . . . . . . . . . . . . 108 Pearson Correlations of Mass Media Exposure With Opinion Leadership . . . . . . lll Pearson Correlations of Mass Media Exposure With Teaching Innovation Network Centrality . . . . . . . . . . . . . 112 Pearson Correlations of Mass Media Exposure With Combined Teaching Network Centrality . . . . . . . . . . . . . 113 Pearson Correlations Within Opinion Leadership Variable Set. . . . . . . . . . . 116 Pearson Correlations Within Teaching Innovation Network Centrality Variable Set . . . . . . . . . . . . . . . . 118 Pearson Correlations Within Combined Teaching Network Centrality Variable Set . . . . . . . . . . . . . . . . 119 Pearson Correlations of Teaching Inno- vation Network Centrality With Combined Teaching Network Centrality. . . . . . . . . 121 Pearson Correlations of Opinion Leadership With Teaching Innovation Network centrality O O O O O O O O O O I O I O 0 O O 122 Pearson Correlations of Opinion Leadership With Combined Teaching Network Centrality . . . . . . . . . . . . . . . . . 123 Factor Analysis of Biographic Variables. . . 134 Factor Analysis of Interpersonal Communi- cation Variables . . . . . . . . . . . . . . 136 Factor Analysis of Mass Media Communica- tion Variables . . . . . . . . . . . . . . . 142 Factor Analysis of Teaching Innovation Dependent Variables. . . . . . . . . . . . . 146 Factor Analysis of Combined Teaching Dependent Variables. . . . . . . . . . . . . 150 xi Table 44. 45. 46. 47. 48. Biographic Independent Variable Factors Regressed With Combined Teaching and Teaching Innovation Dependent Vari- able Factors . . . . Interpersonal Communication Independent 0 Variable Factors Regressed With Combined Teaching Dependent Variable Factors. Mass Media Communication Independent Variable Factors Regressed With Combined Teaching Dependent Variable Factors. Summary of Significant Relationships Between Independent and Dependent Variables. . . . . . Importance of Interpersonal and Mass Media Information Sources for the Average Respondent . xfi O Page 161 166 170 185 196 Figure 0‘ U1 -5 04 N o o o o o 10. 11. 12. 13. 14. LIST OF FIGURES Page Opinion Leadership Choice Listing Data Set 0 O O O O O O O O O O O O O O O O O 32 Non-reciprocated and Reciprocated Dyads. . . 32 Sociogram of Figure 1 Choce Data . . . . . . 33 Three Member Chain Sociogram . . . . . . . . 37 Opinion Leadership Variable Designations . . 40 Network Analysis Sociogram with Liaison. . . . . . . . . . . . . . . . . . . 41 Network Analysis Sociogram with Bridges. . . . . . . . . . . . . . . . . . . 42 Unweighted Non-directed Centrality Variable Designations. . . . . . . . . . . . 49 Weighted Non-directed Centrality Variable Designations. . . . . . . . . . . . 51 Biographic Variable Designations . . . . . . 58 Interpersonal Communication Variable Designations . . . . . . . . . . . . . . . . 62 Mass Media Communication Variable Designations . . . . . . . . . . . . . . . . 66 Complete Listing of Standardized Vari- able Names and Designations. . . . . . . . . 181 Complete Listing of Factor Score Vari- able Names, Designations and Primary variables. 0 O O O O O O O O 0 O O I O O O O 189 xiii CHAPTER I INTRODUCTION Overview This study applied portions of the methodolog1es of diffusion of innovations research and commun1cat1on network analysis research to the field of univers1ty- level accounting education, in order to help br1ng about an understanding of the ways in which developments 1n teach1ng technology are disseminated among account1ng edu- cators. Since prior application of these methodolog1es to the context of accounting education had never been made, this research represents a pioneering, exploratory, tenta- The study has attempted to provide tive, descriptive work. a start toward accomplishing the long-run object1ve of securing maximal rates of adoption, of improvements in instructional technologY, by accounting educators. Nature of the Problem Instructional Technology (IT) has been def1ned as . . a systematic way of designing, carrying out, and evaluating the total process of learning and teaching in terms of specific objectives, based on research in human learning and communication, and employing a com- bination of human and nonhuman resources to bring about more effective instruction. 2 rtance to the learning process has been Its impo g statement by the Co d up in the followin mmission on summe echnology: In the conviction that technology individual Instructional T can make educa- and powerful, ma e tion a more tion more productive, learning more immediate, give instruc scientific base, and make access to education more that the nation equal, sion concludes should increase its investment in instructional technology, thereby upgrading the quality of educa- ’ of individuals' tion, and ultimately, the qual1§y lives and of society generally. y teaching assistants and new faculty mem hing an accounting c "throw them in the the Commis bers, Man ourse for the given the task of teac have been exposed to a " philosophy w pool first time, and they'll learn how cular course might b to swim ith respect to e taught most effectively.3 how the parti c of a rela- in this write symptomati r's opinion, This is, ques ect ive teaching techni emphasis accorded eff ent American tive lack of for accounting topics. In the words of a rec sociation (AAA) Committee: e opinion that the major . . . the Committee is of th impediment to the marriage of IT and accounting edu- cation is the 1ow status which learning theories and research now occupy on the scale de rigueur of account— ing intellectualism. . . If IT becomes fashionable in the order of our pursuits into quantitative analysis a solid framework will have and behavioral science, eady and meaningful progress. been established for st e above and other reasons advanced Accounting As As a result of th such as are ipso facto exper lieves that there has "the fallacious assumption by the AAA Committee, ts of the that holders of the Ph.D. "5 the Committee be learning process, ress to date: been relatively little prog 3 While the Committee has discovered some exemplary applications of IT in accounting education, we are 'on that a thorough beginning throughout of the opin1 ducation has not yet begun. This is not ducators accounting e intended as an indictment--for accounting e are not uniquely ineffective or remiss in his regard--but rather a to action. The signs are clear that great changes will occur in education and in truth have within the decade of the 70's, already begun. . . The immediate challenge to ace in part from the fact that som plines have already begun to e in a variety of ways, indicating both the conviction ' thering the academic objec- as to its importance 1n fur tives of their disciplines, and their intention 0 taking organized and systematic action in this field. ears the profession ounting educators arises e other academic disci- 1evate the status of has been taking tentative It app 5 in the direction 0 8 edition of "A Guid f IT research and application. For step example, the 196 e to Accounting Instruc- Concepts and Practices, " prepared by an AAA Committee, tion: states: ous developments In recent years there have been numer which are of considerable importance to the teaching of accounting. The use of television as an instruc- tional medium has increased. Recent technological progress in data processing has provided the accoun- tant with a powerful tool--the computer. . . Programmed instruction is another teaching method which is find— ing increased application for accounting education. This teaching me d in combination with other traditional methods in the process of developing more effective learning by the student. f the Teacher's Clinic and a review 0 ions of The Accounting Review in can experimentation with In addition, Education Research sect icates there has b accounting educators. port on the role of recent years ind ety of IT topics by content include a re wo professional comm a vari Articles related to course iting courses by t ittees,8 EDP in and 4 as a suggested emphasis for introductory cles related to teac generated assign- and topics such hing tech- ting courses.9 Arti s the use of computer- y answered examinatio terized applications, accoun nology encompas 10 probabilisticall 2 various compu ns,11 and ments, modularized learning; 4 statistical 3 and simulations;1 g case studies1 6 and the use of such includin sampling programs15 and simulations;1 technology as progra nce telephone calls1 mmed instruction,17 teaching diverse 9 and microwave TV 8 confere . 20 1n a case course. machines,1 published "Accounting Education: AAA contains article y 200 pages, The recently Problems and Prospects," by the s on learning and motivation theory, devotes nearl ations and contains sec- n instructional innov nation and research i AAA has offered limited to articles 0 n accounting tions on performance eval Since April 1971, the s specifically for p experience in one or education.21 cial assistance grant rojects with finan promise of advancing knowledge and l innovations. Recently, the bases of educationa ting Educator Award, n Outstanding Accoun ching and contributio more p AAA established a ns to ch excellence in tea n are criteria for for whi nomination. accounting educatio A review of Dissertat s which have investi ion Abstracts has revealed a number of dissertation gated IT tapics ounting education. at least For example, applied to acc s have conducted seven doctoral dissertations in recent year ing the effectiveness of programmed experiments test 0f Ca 0n 5 24 instruction in the teaching of accounting principles. These and other dissertations along similar lines have experimented with rate-controlled speech, 5 free operant 6 computer-assisted instruction,27 random access 30 learning, 9 and multi-media formats. tapes,28 business gaming, inasmuch as half of these studies are Ed.D. dis- However, the sheer number of the dissertations is a sertations, deceptively large indicant of the extent of IT dissertation research by accounting educators. According to the results of a survey published in a recent monograph by Paul Garner, the percentage of accounting education-related dissertations is a smallo-four per cent-~percentage of the total disser- tations written by accounting doctoral candidates. Garner finds this somewhat surprising: It is a little more unexplainable why subjects relating to Accounting Education have not been pursued more frequently and vigorously, since it is well known that more than 75% of the doctoral candidates in accounting thus far go into academic careers and it would be therefore somewhat of a 'natural' for the budding professors to do their doctoral research and writing on educational topics. For the period under observation, however, there is no trend toward pedagogical topics.31 In addition, it is likely that IT topic studies comprise only a part of the dissertations classified by Garner as being related to accounting education. Finally, although the above sources provide evidence of at least some research in IT related to accounting edu- cation, a word of caution has been sounded by the Committee on Instructional Technology concerning a lack of quality 6 exhibited by IT research in general up to the time of the Committee's report: There is too little research, too much of it is of low quality, too little is relevant to the most serious problems of education; and, in general, there is too little direct relationship between research and implementation.32 To bring the above criticism closer to home, the AAA Committee on Multi-Media Instruction makes the follow- ing comment regarding research on programmed instruction: . . . The use of programmed materials in accounting education could benefit from more disciplined experi- ments. . . we venture that most applications lack thg statistical authenticity to reach valid conclusions. 3 To sum up the situation to date: 1. There is a need for quality research regarding the application of newer IT methods to accounting education; 2. There appears to be an emerging awareness of the importance of IT research and its application by accounting educators; and 3. There is limited current implementation within accounting education of the existing newer instructional methods. This dissertation focuses on the last of these three items. No matter how good any specific instructional 34 its overall effectiveness in account- innovation may be, ing higher education will be a function of the extent of its implementation. It is inconceivable to this writer that there will be unlimited resources available, much less 7 expended, for achieving maximal rates of adoption of newer teaching methods within accounting education. The patterns of increasing resistance by state legislatures in granting the budget requests of state institutions, cut-backs in monies available for federally funded research and dissem- ination programs, and the financial difficulty of many smaller and private institutions, have become all too evi- dent in recent years. Given scarce resources, and/or the desire to use the resources that are available for securing the adoption of educational innovations most efficiently, a strategy of being able to focus resource expenditures where they will be most effective is highly desirable. This study attempts to provide a start toward iden- tifying elements of a strategy whereby the more timely and efficient implementation of newer instructional methods may be secured. This research does not make value judg- ments concerning the desirability of using specific teach- ing techniques in accounting courses; it attempts to facil- itate the future adoption, of present or future instruction- al technology innovations, within accounting higher education. Prior Research and Methodological Approach of the Study A relatively recent IT innovation in accounting education is programmed learning, an example of which is the Edwards, Hermanson, Salmonson programmed text.:SS Inas- much as a textbook is a commercial product, the publisher 8 designed a marketing strategy with the objective of secur- ing maximum sales volume for the product. The marketing strategy is, in many ways, analogous to the overall objec- tive of this research: to facilitate the adoption of existing, or future, IT innovations in accounting education. Studies on this general themet-the adoption of innovations over time in a social system-~have been carried out for many years in a variety of academic disciplines: anthrOpology, sociology, education, medical technology and marketing, among others.36 It eventually became apparent to researchers such as Rogers, that many of these studies, although set in the framework of differing disciplines and covering a wide variety of social systems, were reaching substantially similar conclusions. A concerted effort has been made in the last fifteen years to bring together the results of the separate research traditions, culminating in the listing of 112 generalizations regarding the workings of the diffusion process.37 Because diffusion research, now considered a subset of communications research, specifically deals with the adoption of ideas and practices perceived as new by the members of the social system being considered, this writer believes diffusion theory has particular promise for appli- cation to the problem area being considered. In addition, since many of the generalizations from diffusion research have been developed from studies covering a wide variety 0f innovations and social systems, it is likely that 9 concepts and relationships from these prior studies have relevance for the present problem, although the validity of their use with reference to IT innovations in account- ing education needs to be empirically tested. The only major study of which this writer is aware, that has applied elements of diffusion theory to a social system which included accounting faculty members, is the 1967 book by Richard Evans, "Resistance to Innovation In 38 Using the semantic differential as a Higher Education." measurement method, this study examines the attitudes of faculty members from various departments--including account- ingt-at one school toward instructional television. Al- though Evans, in the early chapters of his book, draws heavily upon material from Everett Rogers' 1962 book on the 39 the attempted relationships are, diffusion of innovations, in this writer's opinion, often inappropriate and modified to coincide with the form of the author's data base. It is interesting to note that results of the same study were originally published in 1962 under the title, "The Univer- sity Faculty and Educational Television: Hostility, Resis- "40 No mention was made of diffusion re- tance and Change. search in the 1962 version. The diffusion research tradition in education was led in the early years by Paul Mort41 of Columbia University Teacher's College, and in recent years by Ronald Havelock42 of the Institute of Social Research. The great bulk of the studies which comprise this research tradition have 10 examined primary or secondary educational systems.43 Havelock identifies four major strategic orientations for securing the adoption of educational innovations: problems solving; social interaction (SI); research, development 44 Of these and diffusion; and his own linkage concept. strategies, the SI approach has the largest empirical foun- dation and is, in this writer's opinion, the most appropri- ate for application to accounting education. A significant element is the design of a diffusion strategy using an SI approach is the concept of opinion leadership, which has been defined as follows: Opinion leadership is the degree to which an individual is able to informally influence other individuals' attitudes or overt behaxior in a desired way with relative frequency. 5 The concept of opinion leadership developed from the assumption of a two-step flow communication model as the foundation of the diffusion process. The steps in the flow were posited as follows: The first step, from sources to Opinion leaders, is mainly a transfer of information; whereas the second step, from opinion leaders to their followers, involves also the spread of influence.46 More recent theory has assumed a multi-step flow model, which incorporates the two-step flow model, the one-step 47 flow model and the hypodermic needle model. The multi- step flow model "suggests that there are a variable number of relays in the communication flow from a source to a receiver."48 11 The importance of opinion leaders in planning a diffusion strategy may be seen in the following: Several researches indicate that when the social system is modern, the opinion leaders are quite innovative; but when the norms are traditional, the leaders also reflect this norm in their be- By their close conformity to the system's havior. norms, the opinion leaders serve as an apt model for the innovation behavior of their followers.4 Thus, Opinion leaders function as potential influ- entials in their system, serve a linking function in the transfer of information, and must be considered in designing a diffusion strategy regardless of the location Of the sys- tem's norms on a modernism-traditionalism continuum. Communication network analysis, a subset of communi- cations research, also provides a means of identifying in- dividuals who play key roles in the communication activities A description of a communication net- within their system. work, and a brief summary of network analysis, follow: Communication networks arise in a social system where repetitive, recurring patterns of interaction take place among the system's members. Communica- tion networks, then, are derived from an aggregate or sum of the interactions in a system, occurring across time and space. The networks provide the means by which messages move from member to member throughout the system. The basic unit of inter- action is the linka e or communication relation between pairs of system members, i.e., the dyadic linkage. 0 The initial goals of communication network analysis are essentially descriptive or classificatory in the initial analytic task is to nature. That is, reduce the membership of the system to some smaller number of categories that allow the investigator to describe the networks in whatever manner best fits the purpose of the research. Given that the rela- tions under study reflect various aspects of the 12 communication or message exchange process among system members, one logical set of categories to use are those that delineate various communication These roles may be defined in differing roles. ways, but Often they are of three main types: (2) inter- (1) member Of a communication group, group linker, and (3) isolate, or non-participant in the network. Thus, although network analysis does not necessarily deal with messages that are perceived as new by members of the network analysis does enable the classification system, of individuals in a defined network by functional roles such as group member, bridge, liaison, tree node, and 150- late.53 Individuals in certain of these roles-~liaisons and bridges--provide a linking dimension between groups in the network and thereby play key roles in the dissemination of information throughout the network. This writer is unaware Of any major study that has applied network analysis to a system-~large 23 small--which has included accounting faculty members and for which the communication topic area has been instructional technology or teaching innovation. In addition, since the analytical tools for identifying roles and network structure in larger 54 there have systems have only recently become available, been relatively few empirical studies that have examined d.55 The role characteristics in larger systems of any kin advent of the analytical tools for larger systems has Spurred consideration of structural variables at many differ- ent levels of analysis-~individual, group, sub-system and SYStem,56 resulting in refinements of the measurement pro- 57 cesses at all levels of analysis. 13 In summary, elements of the methodologies of both diffusion research and network analysis have been utilized in this study. Each methodology attempts to identify key relationships existing within the communication activities in a given system: The concept of Opinion leadership, from diffu- l. sion research, focuses on potential influentials, and advice-seeking relationships, in the system, and 2. The concept of functional communication roles from network analysis focuses on the linkage and structure dimensions within a communication network. By the application of these tools in the context of higher education in accounting, this dissertation examines aspects of the communication process occurring within selected systems of accounting educators, with the hope of identify- ing focal points potentially useful in the formulation of a strategy for securing maximal rates of adoption. This study, in many respects, is truly an explora- The ground being covered is virgin, and in some tory one. instances has proven either barren or resistant to close Nonetheless, it is the hope of this writer that scrutiny. the research has provided a significant start toward ad- dressing a problem that should be of real concern to account- ing educators. TI 14 Organization of the Thesis Chapter II of the dissertation discusses the selec- tion and Operationalization of the dependent and independent the data-gathering procedures used, response re~ variables, sults, bias considerations, and specification of the data sets analyzed. The statistical analysis of the data, presented in Chapter III, begins with a Pearson product moment correla- This is followed by a discussion of the tion analysis. factor analysis procedures employed, the determination of significant factors, and the results of the factor analyses. Finally, the multiple regression procedures used, and the results Of the multiple regression analyses, are discussed in the closing section of the chapter. The initial section of Chapter IV consists of a summary of the results of the analyses contained in the pre- ceding chapter; subsequent sections of Chapter IV detail the major conclusions of the study, discuss the major limi- tations Of the analyses, and provide suggestions for future research. 0:: Up SC in; Acc IIEI 15 FOOTNOTES TO CHAPTER I 1Commission on Instructional Technology, :9 Improve Learning, ed. by Sidney G. Tickton (2 vols.2;1 New York: R. R. Bowker Company, 1970), Vol. I, p 21bid., p. 10. reflects this writer's 3This statement, of course, Opinion, based upon his personal eXperience, as well as upon discussions with junior faculty at many major institu- It appears to this writer that the practice de- tions. scribed represents the rule, rather than the exception. 4Committee on Multi-Media Instruction in Account- ing, "Report of the Committee on Multi-Media Instruction in Accounting," Supplement to Volume XLVII Of The Accounting Review, 1972, p. 115. SIbid. 61bid., pp. 115-16. American Accounting Association Committee to Prepare a Revised Accounting Teachers' Guide, A Guide to Concepts 6 Practices (2d ed.; Accounting Instruction: C1nc1nnati, Ohio: Scut51Western PubliShIng Co., 1968), p. 84. American Accounting Association Committee on Accounting Education and American Institute of Certified Public Accountants Computer Education Subcommittee, "Inclusion Of EDP in an Undergraduate Auditing Curriculum: Some Possible Approaches," The Accounting Review, Vol. XLIX (October, 1974), pp. 859-64. Barry E. Cushing and Charles H. Smith, "A New Emphasis for Introductory Accounting Instruction," The Agcounting Review, Vol. XLVTI (July, 197?), pp. 599-601. 10Eugene L. Zieha, "Computer-Generated Accounting Assignments,” The AccountingwReview, Vol. XLIX (July, 1974L pp. 600-02. 11Irvin N. Glein and John B. Wallace, Jr., A Field Test," "Probabilistically Answered Examinations: 1b; Accounting Review, Vol. XLIX (April, 1974), pp. 363-66. 16 lor and Harold Western, for Intermediate 2Jay M. Smith, Dale Tay XLIX (April, 1974), "Experiment in Modularized Learning Accounting," The Accounting Review, Vol pp. 385- 0. . Ricketts, "A 13James J. Benjamin and Donald E Profit Planning Project in the Management Accounting Coursefl The Accountin Review, 0 XLVIII (October, 1973), pp. 794-97. 14William R. Kinney, Jr., ”The Use of the Time- ter in Audit Education," The 90-94; Shared Interactive Compu l. XLIX (July, 1974), pp. 'mulation Accountin Review, 0 J. Timotfiy Sale, "Using Computerized Budget Si Models as a Teaching Device," The Accounting Review, I (October, 1972), pp. 836-39. Vol. XLVI , "Audit-Aid: Generalized Computer- as an Instructional Device," The Accountin Audit Program Review, Vol. XLV (October, 1970), pp. 774-78; V. Thomas Dan M. Guy and Doyle 2. Williams, ”Integrating the An Approach in Auditing," Th3 Computer in the Classroom: Accounting Review, Vol. XLIX (January, 1974), pp. 149- 3. 16Alvin A. Arens, Robert G. May and Geraldine Dominiak, "A Simulated Case for Audit Education," Thg Accountin Review, VO . XLV (July, 1970), pp. 573-78; Paul H. WalgenEach and Werner G. Frank, "A Simulation Model for Applying Audit Sampling Techniques," The Accounting Review, Vol. XLVI (July, 1971), pp. 5 8. 17Billy E. Askins, "Determining the Effectiveness ruction--A Training Course Example," The Of Programmed Inst Vol XLV (January, 1970), pp. 159-63; "Programmed In- Accounting Review, . William Mar e l and Wilfred A. Pemberton, struction in Elementary Accounting-~15 It Successful?" The Accounting Review, 01. XLVII (April, 1972), pp. 381-84. 186. Fred Streuling us Lectures i Machines Vers Experiment," The Accountin Review, 1972), pp. 806-10. 19Michael H. Granof, Means to Bridge the Academic-- 1. XLVIII (July, 1973), Accounting Review, V0 h, "The Case Method of Accounting 20Andrew M. McCos e Television," The Accountigg Instruction and Microwav 831133, Vol. XLVII (January, 1972), pp.rr - Holstrum, "Teaching Education: An 1. XLVII (October, and Gary L. n Accounting Vo "Conference Telephone Calls: A 'Real World' Gap," The pp. 612'14. 21James Don Edwards, ed., Accountin and Education: : American Accounting Problems and Pros ects (n. p. Association, 1974I. 17 22See The Accounting Review, Vol. XLVI (April, 1971), p. 397. 23See The Accounting Review, Vol. XLVIII (April, 1973), pp. 440-41. 24Franklin Eugene Butts and Gary L. Prickett, "The Effect of Audio-Tutorial and Programmed Instruction Labor- atories on Achievement in Accounting Principles” (unpub- lished Ed.D. dissertation, Colorado State University, 1969); Charles Douglas Cloud, "An Experimental Study Comparing the Effectiveness of Programmed Instruction and the Conventional Method Of Teaching First-Semester Principles of Accounting" (unpublished D.B.A. dissertation, Arizona State University, 1971); Victoria Lee DeFore Daily, "The Effect of Programmed Instruction in the Teaching of Principles of Accounting" (unpublished Ed.D. dissertation, Colorado State University, 1969); Mildred Williams Glover, "An Experiment in the Use of Programmed Instruction in Elementary College Accounting" (unpublished Ed.D. dissertation, University of Georgia, 1970); Sunion Theodore Hong, "An Empirical Study of the Effectiveness of Programmed Instruction and Computer- Assisted Instruction in Elementary Accounting" (unpublished Ph.D. dissertation, New York University, 1972); Joseph Lee Humphrey, "An Inquiry Into Programmed Instruction as a Pedagogical Technique in Accounting Education” (unpublished D.B.A. dissertation, Texas Tech University, 1971); Dominick Salvatore Orefice, "An Experiment to Determine the Effec- tiveness of Programmed Instruction in Elementary Accounting" (unpublished Ed.D. dissertation, Rutgers University, The State University of New Jersey, 1971). 25Frederick Miller Cole, ”A Study of Comprehension Levels of College Students Studying Elementary Accounting Via Rate-Controlled Speech" (unpublished Ed.D. dissertation, University of Florida, 1971). 26Milton Mike Will, "The Effect of Free Operant Learning on Achievement in the Principles of Accounting Course" (unpublished Ph.D. dissertation, University of North Dakota, 1970). 27Hong, "An Empirical Study of the Effectiveness of Programmed Instruction and Computer-Assisted Instruction in Elementary Accounting." 28Stephen Michael Flanagan, "The Effectiveness of Random Access Tapes in the Instruction of Elementary Accounting (unpublished Ed.D. dissertation, University of Northern Colorado, 1970)- 18 l Caldwell, "An Inquiry Into Business chnique in Accounting Education" 29Jimmy Car University of Alabama, Gaming as a Pedagogical Te (unpublished Ph.D. dissertation, 1970). 30Butts and Prickett, "The Effect of Audio- Tutorial and Programmed Instruction Laboratories on Achieve— ment in Accounting Principles;" Julius Onvorah Onah, "An Experimental Study Using the Audio-Visual Tutorial System s Of Accounting to Community College to Teach Principle Students” (unpublished Ph.D. dissertation, Michigan State University, 1971). 31Paul Garner, §9me Reflection on Research bx Doctoral Candidates in Accounting (Univer51ty, a ama: CEnter for Business and Econom1c Research, University of Alabama, 1973), p. . 32Commission on Instruction Technology, To Improve Learning, Vol. 11, p. 917. 33Committee on Multi-Media In ing, "Report of the Committee,“ p. 9 the first 34This writer hopes that progress in ' ted to accounting area mentioned--quali h in IT rela ' n for judging the desir- 1 methods. education-~will provide a foundatio ability of implementing Specific instructiona search settings-~multiple experimenta- Certainly there are abundant re section courses in large schools--for careful as the AAA would fund a this writer believes sub- tion. If an organization such rt in this direction program of such experimentation, stantial progress would be made. A sta ' r of research has been made 1n the form of a limited numbe grants that have been available since 35James Don Edwards, Roger H. Hermanson, Xt (2 vols.; Salmonson, Accountin A Pro rammed Te Homewood, Illinois: R1chard D. IrW1n, Inc., 1967). 36See in particular Everett M. Rogers with F. Floyd of Innovations, A Cross—Cultural ihoemaker, Communication 1971) pproach (23 ed., New York: The Free Press, , pp. ' O. 371bid., pp. 346-85. 38Richard 1. Evans, Resist Higher Educatigp (San Francisco: Inc., 1970). ion (New York: 39E. M. Rogers, Diffusion of Innovat Free Press, 1962). struction in Account- and R. F. ance To Innovationlp ossey-Bass Publishers, 19 40Richard I. Evans, Ronald G. Smith, and William K. Colville, The University Faculty and Educational Television: Host111ty, Res1stance, and Change (Houston, Texas: Univer- sity of Houston, 1962)1 41Rogers with Shoemaker, Communication of Innovations, p. 58. 42 for example, Ronald G. Havelock, A Guide to Institute fOr See, Innovation in Education (Ann Arbor, Michigan: SociaIResearch, 1970); Ronald G. Havelock and Mary C. Havelock, Training for Change Agents (Ann Arbor, Michigan: Institute for SoCiaITRESearch, 1973); Ronald G. Havelock, The Change Agent's Guide to Innovation in Education (Englewood Cliffs, New Jersey: Educational Technology Publications, 1973). 43Good examples in this category include the summary of the Columbia studies compiled by Donald H. Ross, Administration for Adaptability: A Source Book Drawing TOgether the ResuIts of Here Than 150 Studies Related to the Questibn ofhWhy_and'How SchooIs Improve (Neinorki Metropolitan School Study CounciI, 1958), and the work by Richard O. Carlson, Adoption of Educational Innovations (Eugene, Oregon: The Center for the Advanced Study of— Educational Administration, 1965). 44Havelock, Change Agent's Guide, pp. 151-68. 45Rogers with Shoemaker, Communication of Innovations, p. 35. 461bid., p. 205. 471bid., pp. 203-09. 481bid., p. 209. 491bid., p. 35. 50Richard V. Farace, William D. Richards, Peter R. Monge, and Eugene Jacobson, "Analysis of Human Communica- tion Networks In Large Social Systems," unpublished paper, Department of Communications, Michigan State University, May, 1973. p. 3. 511bid., p. 4. I‘NAIINs-IICVIMA DZgCHa 20 52 . Although 1t may by specification of the content of the communication for which the network is defined. As an example, a number of researchers have subdivided work- related communication into content areas such as production, innovation, and maintenance, and then defined separate net- works for each content area. See, for example, Richard V. "Comparative Analysis 0 lected Formal Organiza- a paper presented at the tions" (mimeographed copy Of ting in New International Communication Association mee Orleans, April, 1974). "Large Social 53See, for example, Farace, et 31., Systems," pp. 13-14. 54The most frequ analysis has been sociome systems can becomes unwieldy with larger 5 method of storing sociometric nd column reordering rows 3 and identify structural patterns: Robert Stuart Weiss, "Processes of Organization" (unpublished Ph.D. dissertation, University of Michigan, 1954). This technique proved reasonable for moderate size systems, but has been sub- stantially improved for the analysis of large systems by a developed by Richards. See William D. computer program "An Improved Conceptually-Based Method for ructures Of Large ently used data source for network tric choice data. Although smaller d by hand, the process ' 5 describes a rix form, and then to form groups Richards, Jr., Analysis of Communication Network St anizations" (mimeographed; East Lansing, ' Michigan State Complex Org Michigan: Depar University, 1971 sis in Large Comple Tools" (mimeographe national Communicat April, 1974). 55See, for example, Richard V. Farace and James A. Danowski, "Analyzing Human Communication Networks in Organi- zations: Applicat1ons to Mana Problems" (mimeo— graphed copy of paper presented at the International Association meeting, March, 1973); Donald Communication MacDonald, "Communication Roles and Communication Content In a Bureaucratic Setting" (unpublished Ph.D. dissertation, Michigan State University, 1970). 56See, in particular, William D. Richards, Jr., "Network Analysis in Large Complex Systems. Theoretical Basis" (mimeographed copy of paper presented at the Inter- national Communication Association meeting in New Orleans, April, 1974); William D. Richards, Jr., "Network Analysis in Large Complex Systems: Metrics" (mimeographed copy of paper presented at the International Communication Association meeting in New Orleans, April, 1974); Farace, et 31., "Large SOCial Systems," pp. f Communication, m D. Richards, Jr., Techniques and Methods-- presented at the Inter- ting in New Orleans, tment 0 "Network Analy- ); Willia x Systems: d copy of paper ion Association mee 21 57This is exemplified by the increasing complexity of the data-gathering forms used in recent years, which now often include, in addition to identification of the contact, multiple content areas, frequency levels and direction of initiation with respect to the communication activity being measured. See Peter R. Monge and George H. Lindsay, "The Study of Communication Networks and Communication Structure in Large Organizations" (mimeographed copy of paper pre- sented at the International Communication Association meet- ing in New Orleans, April, 1974) for a good introduction to network analysis in general, as well as sample data instru- ments. More comprehensive examples include Edwin H. Amend, "Liaison Communication Roles of Professionals in a Research Dissemination Organization” (unpublished Ph.D. dissertation, Michigan State University, 1971); MacDonald, "Communication Roles and Communication Content;" and Donald F. Schwartz, "Liaison Communication Roles in a Formal Organization” (unpublished Ph.D. dissertation, Michigan State University, 1968). CHAPTER II GENERATION OF THE DATA BASE From a potential population of interest of all accounting educators in the United States, ten AACSB schools were defined as separate systems and selected for inclusion in the study. A census of all full-time, permanent, ac- counting faculty members at these schools was conducted to obtain the data from which 20 dependent, and 42 independent, variables were generated. Although the overall response rate for the study was in excess of 90 per cent, concentra- tions of nonrespondents prohibited the calculation of depen- dent variable measures at two schools. Ninety-seven indi- viduals from the remaining eight schools form the reSpondent data set used in subsequent analyses. The first section of this chapter specifies the population and sample, and is followed by sections on the data-gathering procedures used, selection and Operational- ization of the dependent and independent variables, response bias testing, and the data modification procedures. Population and Sample The ultimate population of relevance to the research question addressed by this study consists of all teachers 22 C‘) 23 of accounting at the college level. Inasmuch as the meth- ods used in this research to measure opinion leadership and the linking communication function require virtually a 100 per cent sample and response rate from the defined system, the overall population was broken into smaller systems‘- departments--so that control procedures which would permit a realistic chance of achieving the high required response rates could be employed. For the purpose of this research, a department was defined as all full-time, permanent, accounting faculty members at an institution of higher learning, who had been in residence at least one full term during the academic year in which the data was gathered-~1974-7S. This defini- tion excludes: 1. Part-time faculty members such as practitioners teaching an accounting course, and individuals whose pri- mary responsibilities were those of an administrative posi- tion other than department head or chairman; 2. Non~permanent individuals such as visiting faculty from another school, and graduate students who held the rank of instructor or equivalent; 3. Faculty who held a full-time, permanent position at their institution, but who had been gone all academic year. Ten departments of accounting, chosen from the membership of the American Association of Collegiate Schools of Business (AACSB) were selected for inclusion in the In de in 24 study and form the defined population. The AACSB school group includes many large and/or state universities, and is considered a significant population with respect to two dimensions which have relevance for this research. First, the number of students in the accounting programs at many of the AACSB institutions is substantial. Since students are at least one, if not the primary, group who would bene- fit from improved instructional methods, selection of these schools promises large numbers of potential beneficiaries.1 Second, interviews with a number of publisher representa- tives, conducted when this study was in the research design stage, indicated that a large school often serves as a model-~opinion 1eader—-for smaller schools in the nearby geographic vicinity, with respect to factors such as course content and selection of textbooks. This appears to be especially prevalent in states with large branch, or state, systems. Although the ten schools selected were not chosen at random, they are considered representative of the AACSB population in this research.2 In order that the reader may, if he so desires, infer the results of this study to a population of interest such as all AACSB institutions, summary descriptive information concerning department size, highest academic degree, professional certification, aca- demic rank distribution, tenure status, total years teach- ing and years at present institution for the faculty at the ten schools selected, is presented in Table 1. In addition, 25 Table 1. Summary of Descriptive Information Pertaining to Departments Which Received Questionnaires Number of faculty Number of faculty Percent of total Number of faculty Percent of total Number of faculty Percent of total Number of faculty Percent of total Number of faculty Percent of total Number of faculty Percent of total Smallest 5 Masters 24 19.05t Certified 96 76.193 DEPARTMENT SIZE Total Largest Mean Departments 19 12.6 10 HIGHEST ACADEMIC DEGREE Doctorate 102 80.953 PROFESSIONAL CERTIFICATION Not Certified 30 28.31‘ ACADEMIC RANK DISTRIBUTION Instructor, Assistant Associate Full EEEEUTCT Professor Professor Professor 3 SS 26 42 2.38l 43.65t 20.63t 33.33% TENURE STATUS Tenureg Non-tenured 68 58 53.97l 46.031 TOTAL YEARS TEACHING 0-4 5—9 10-14 15‘19 20* 30 17 9 31 23.81% 30.95‘ 13.49l 7.14t 24.60% YEARS AT PRESENT INSTITUTION 0-4 5-9 10-14 l5-19 20+ 59 14 7 l7 46.83t 23.02‘ 11.11t 5.56% 13.498 Total Faculty 126 Total Faculty 126 1003 519.131 126 100% Total Faculty 126 100% Total Faculty 126 100% Total Faculty 126 100% Total Faculty 126 100‘ 26 it might be noted that the ten departments are geographi- cally dispersed over most of the continental United States, and are evenly split between schools on a quarter system, and schools on a semester or trimester system. Three major types of statistical techniques are em- ployed in this research, with different units of analysis examined depending upon the procedure used. Variable means for respondent and nonerespondent groups were tested for differences using t-tests; 60 z-score variables for each of 97 individuals were factor analyzed by variable type; factor scores for each individual, generated from the factor analy- ses, were used as a data base for multiple regression pro- cedures. Where appropriate, tests of statistical signifi- cance have been presented as an aid in interpreting the results, and to supply an additional informational dimension for the reader who wishes to infer the results to a pOpula- tion of interest. The reader should, of course, be aware that since the individual respondents analyzed in this re- search constitute a population-~not a random sample-~then, for some of the procedures used, any actual difference is a "statistically significant" difference. Whether such differ- ences represent meaningful differences is a matter of judg- ment; as is the interpretation of the size of correlation coefficients, factor loadings and adjusted R squares.3 Data Gathering Procedures The data analyzed in this research was gathered in two phases--two of the 10 schools were chosen for the 27 initial distribution of the instruments in fall of the 1974-75 academic year; the remaining eight schools were censused in late spring of the same academic year. In both instances, an individual known to the faculty at each school distributed the questionnaires, assured respondents of anonymity, and requested the c00peration of the individ- uals in his department. Since only minor editorial changes were made in the questionnaire sets used for each distribu- tion, and since the differing times of collection were not considered a significant difference, the data sets from the two distributions were combined for the analysis in this research. The procedures used for each questionnaire dis- tribution are detailed in the following subsections. Initial Distribution Two schools, whose faculty were known personally by the researcher, were selected for the initial distribution of the data instruments. Distribution of the data-gathering materials, which included a cover letter, communication questionnaire and personal contact listing,4 was made by the researcher, who also assured the respondents of anonymity. Personal interviews were conducted with most of the faculty members at these two schools after the questionnaires had been returned, in order to determine whether the respondents experienced difficulties in filling out the instruments, whether there were semantic difficulties with any questions, and to obtain an estimate of the average time required to 28 complete all materials. No unforeseen difficulties were encountered,5 and only very minor changes-—spelling and punctuation-~needed to be made in the instruments. The average time required by these respondents for completing both the communication questionnaire and personal contact listing was half an hour. There were also strong indications from the inter- views that an implicit, perceived "norm” exists for the amount of communication that a faculty member should have with his colleagues on professional and teaching‘related matters.6 Many respondents, both in the initial and second distribution groups, expressed surprise at the relatively low--as perceived by the respondents-~frequency levels of communication with their fellow faculty members that they reported in their own questionnaire answers. These feelings were universal enough to have generated conversations on this tOpic, after most of the data-gathering had been com- pleted, between groups of faculty members at most of the ten schools included in the study. Second Distribution As previously mentioned, the ten schools selected from the AACSB population are geographically dispersed over most of the continental United States. Since it was not economically feasible to obtain the data by personal inter- views with the faculty at the remaining eight schools, there- by necessitating use of the United States Postal Service, 29 the following procedures were used in an attempt to provide the study with source credibility at each school. First, the c00peration of an individual faculty mem- ber, who agreed to handle the distribution of the question- naire and to request the participation of his fellow facul- ty members, was obtained in advance. These individuals also served as information sources after the actual distri- butions had been made--for questions from their fellow fac- ulty concerning the nature and purpose of the research, and for the researcher with respect to problems encountered in gathering the data at each school. A second procedure used was to make sure, in advance, that the department chairman knew of the research, knew that his faculty were being asked to participate and would, as a minimum, not discourage participation. This was accom- plished by an initial letter briefly explaining the nature and the purpose of the research, followed by a telephone call in which any questions by the chairman concerning the study were answered, and in which his c00peration, in the form of a memo to his faculty or mention of the study in a faculty meeting, was solicited. The package of materials distributed to each faculty member at each school consisted of: 1. Cover letter for the data instruments; 2. Communication questionnaire, which was the data source for all independent variable measures and the opinion leadership dependent variables; 30 3. Personal contact listing, the data source for the network centrality dependent variable measures; 4. A return envelope with individually typed to and from address labels, and which bore a forty cent stamp; and 5. The envelope containing the above materials, bearing an individually typed label addressed to each facul- ty member. All printed materials used in these data—gathering procedures were personalized to the maximum extent possible, and were professionally printed. For example, the cover letters for each of the individual questionnaires were in- dividually typed, using an IBM MT/ST typewriter, on Michigan State University letterhead; had the name which the author of the cover letter would usually have used in addressing each respondent included in the salutation; and were indi- vidually signed in ink. Samples of the letters to department chairmen, cover letters, communication questionnaires, and personal contact listings are included in the Appendix. Operationalization of the dependent and independent variable measures generated from the combined data sets of the initial and second distributions are detailed in the following two sections. Dependent Variable Measures Twenty dependent variable measures were selected for analysis in this research-~six opinion leadership indexes 31 that measure reported advice-seeking behavior, and fourteen network centrality indexes that measure the extent to which individuals provide a linking dimension in the flow of in- formation throughout their system. Three of the six opinion leadership indexes pertain to advice sought with respect to new teaching methods; the remaining opinion leadership indexes are defined with res- pect to advice sought regarding overall teaching effective- ness and improvement. The fourteen network centrality in- dexes are split along similar lines--7 variables measure teaching innovation communication; 7 variables measure com- munication on many teaching-related matters. The concepts underlying, and the method of calculating, each index are presented in the following subsections. Opinion Leadership Indexes Opinion leadership has often been measured using a sociometric choice question of the following general form: "Whom would you ask for information or advice concerning Topic X?" Variants of this question include asking the question with respect to past, rather than future, behavior; and the specification of a limited number of choices "whom you would be most likely to" or "whom you have sought out most often." Responses would be solicited from as many members of the defined system as possible, resulting in 7 The choice nominations from most members of the system. data from this type of question can be conveniently repre- sented in the form of either a sociogram or matrix. For 32 example, suppose the data set listed in Figure 1 represents the choice nominations from a defined five member system: Individual 1 chooses individuals 2 and 3, in that order. Individual 2 chooses individuals 3 and 4, in that order. Individual 3 chooses individual 2. Individual 4 chooses individual 3. Individual 5 chooses individual 1. Figure 1. Opinion Leadership Choice Listing Data Set A sociogram is an illustration of the number and direction of reported sociometric choice nominations, where each individual in the defined system is represented by a circle and each choice is represented by an arrow. A direc- ted arrow pointing toward one circle--individual A--from another circle--individual B‘-represents individual A having been chosen by individual B. See the left half of Figure 2. Non-reciprocated Reciprocated /_ 1 i® .. 7C9 Figure 2. Non‘reciprocated and Reciprocated Dyads If reciprocal choices have been made by individuals A and B--each has chosen the other-~the arrow between the two in- dividuals will point in both directions, as in the right half of Figure 2. The following sociogram represents the data set listed in Figure 1: (‘3 Eh 33 Figure 3. Sociogram of Figure 1 Choice Data that is especially convenient for computational purposes is the use of a square matrix, whose rows represent the re- spondents-~choosers--and whose columns represent their Choices. A cell entry of l in the matrix indicates the ex- istence of a choice by the row individual of the column in- dividual; a cell entry of 0 indicates the lack of such a Choice. The matrix in Table 2 is a representation of the choice data from Figure 1. Either method of representing the data can be use- fUl for the analysis of opinion leadership. For example, counting the number of directed arrows toward each indi- vidual in the sociogram will inform the researcher as to which individuals are chosen most often for advice concern- ing the topic of the question. A reference to Figure 3 indicates that individual 3 has been chosen most often-- three times--by the other system members, and individual 2 has been chosen next most often. The same information can 34 Table 2. Binary Matrix of Figure 1 Choice Data Individual Choice Number l 2 3 4 S l - l I l O 0 2 O - l l 0 Individual Respondent 3 O l - 0 O lumber 4 0 0 l - i 0 5 1 [—3 I O O I r —; J .__ 1 2 3 1 0 be read from the column totals of the matrix in Table 2. Thus, the number of choices received by each individual in the system is a basic measure of the extent to which other members in the system report either having sought, or are Willing to seek, the advice of that individual concerning the question topic. This basic measure, consisting of the number of CbOice nominations received, can then be converted to a Size-free, continuous variable with a potential range of zero to one by dividing by the total possible number of ChOices that could be received. In formula form, the resultant measure is: 3S Unweighted a a = number of choices Opinion received Leadership n = number of individuals Index n ' 1 in the system Individuals 2 and 3 in the preceding five member data set would have unweighted opinion leadership scores of 0.500 and 0.750, respectively, indicating they are chosen by 50 per cent and 75 per cent of the other members in their system. A slight variation of the above index can be achieved by assigning inverse weights according to the or- der in which an individual's choices are listed. Thus, if up to two choices were specified in the question, the indi- vidual chosen-elisted--first by a respondent would receive a score of two, and the individual chosen second would re- ceive a score of one. Data in this form can be analyzed in either a sociogram or matrix form, the easier of which is usually the matrix representation. The only adjustment required consists of replacing cell entries of l with the apprOpriate assigned weight. A matrix of this type, pre- pared for the data from Figure l, is illustrated in Table 3. The column totals of this matrix yield the sums ofthe weights corresponding to the choices each individual has received and are, in themselves, a second basic measure of Opinion leadership. This measure can be converted to a size-free, continuous index with a potential range of zero to one by dividing the weight score sum for any individual 36 Table 3. Weight Matrix of Figure 1 Choice Data Individual Choice Number 1 2 3 4 S 1 - 2 1 o In! 2 0 - Z 1 0 Individual Respondent 3 O 2 - O 0 Number 4 O O 2 - 0 S 2 0 0 0 - j 2 4 5 l 0 by the maximum that could be achieved. In formula form, this index would be: h = column total from Weighted b matrix c = number of choices Opinion _ Leadership - c (n _ 1) in question Index n = number of individuals in the system The third measure of opinion leadership employed in this research is based on the concept of centrality--the degree to which an individual is linked to the other members of his system. When operationalized with respect to Opinion leadership choice data, this concept becomes the degree to which an individual functions as a real, or potential, in- fluence center or focal point, in the advice-seeking commu- nication patterns within his system. By incorporating the 37 idea of advice-seeking links between system members, the third measure allows for paths of potential influence in the advice—seeking behavior of individuals in the network. For example, in the simple sociogram illustrated in Figure 4, individuals B and C would each have Opinion leadership ® #69 {Q Figure 4. Three Member Chain Sociogram index scores of 0.50, weighted or unweighted, and indi- vidual A would have index scores of 0.00. Yet, if you had to choose the one individual in the system who would, every- thing else being equal, have the greatest potential influ- ence in this three member system, you would choose indi- vidual C. Why? Because if individual C can influence individual B, who can, in turn, influence individual A, then individual C can also potentially influence individual A.8 An alternate way of stating this consists of describing the advice-seeking relationships in terms of directed paths up to two steps in length between each of the system members: There are no directed paths to individual A. There is a one-step directed path from individual A to individual B. There is a one-step directed path from individual B to individual C. There is a two-step directed path from individual A through individual B to individual C. There are no two-step directed paths to individual B. The method of calculating the index from this data proceeds as follows. The maximum possible path length-- allowing no redundant links or steps-~from one individual 38 to another in a system of n individuals is (n-l) steps in length. Since the shorter the path, everything else being equal, the greater the potential influence,9 the shortest directed path from each individual to each other individual-- if a directed path existle--is identified. These shortest paths are then inversely weighted, beginning with a weight of (n-l) for a one-step path, (n-Z) for a two-step path, and so on. For example, individual 4 in Figure 3 is connected by two-step directed paths from individuals 1 and 3. Since there are five members in this system, each of these two- step paths would be weighted with a value of 3. The weights corresponding to the shortest directed paths toward one individual from all other system members are then summed and divided by (n-l)2, the maximum score that could be 11 attained. These calculations, for the three member sys- tem illustrated in Table 4, are illustrated below: Table 4. Directed Centrality Opinion Leadership Index Calculations for Figure 4 Choice Data __ Shortest Path Path Weights Weight Sum Individual l-step Z—step l-step Z-step Sum : (n-l)2 A O 0 0 0 0 0 B l 0 2 0 2 0.500 c 1 1 2 1 3 0.750 39 The resultant measure,ttnmumlthe directed central— ity opinion leadership index, is a size-free, continuous variable with a potential range from zero to one. Notice that this index has rank ordered individuals A, B and C in the order of their relative potential influence, whereas both of the previous measures failed to discriminate between individuals B and C. Individual A has a directed index val— ue of 0.00, indicating that he is not sought for advice by any other members of his system. Individual C has the high- est directed centrality index value--O.7S--indicating very high, but not the maximum possible, potential influence. If there had also been a one~step directed path from indi- vidual to individual C, then the index value for individual C would be 1.00, indicating that he is the locus of direct- ed one-step paths from all other members of his system. All three opinion leadership measures just cited are used in this research and were calculated with reference to each of two topic areas-~new teaching methods and general teaching. The specific questions used in the questionnaire to obtain this data were the following: 4.1 Do you discuss ways to improve the learning experi- ence of your students with any full-time, permanent accounting faculty members in your department? Yes . No . (IF NO: Please continue with question 4.2)——TF YES: 4.1.1 Please list the names of the three individuals you seek out most often for information and/or advice. 40 4.2 Do you discuss new teachin methods and materials in accounting education (e.g., programmed textbook, teaching by television, preparing transparencies) with any full-time, permanent accounting faculty 4.2.1 Please list the names of the three indie _. _ The terminology "ways to improve the learning experience of your students," used in question 4.1, was selected as repre- sentative of the multitude of possible topics that could be considered related to teaching improvement and overall teaching effectiveness. The t0pic of question 4.2-~new teaching methods and materials--was intended to be a subset Of the general teaching dimension of question 4.1. The six opinion leadership indexes calculated for each individual are summarized in Figure 5: Variable Teaching Topic Area Index Type De51gnatlgfl . Weighted ”1 Ways to improve Unweighted DZ learning experience Directed centrality D3 - D4 Wei ted New teaching methods Unw§?ghted D5 and materials Directed centrality D6 Figure 5. Opinion Leadership Variable Designations 41 Network Centrality Indexes Functional communication roles have usually been identified by analyzing data obtained from questions of the following general type: "Which of your fellow employees do you communicate with about Topic X?” More complex data bases can be generated by asking respondents to indicate the approximate frequency of contact, to identify the usual mode of communication, to indicate the average direction- ality of contact, to assess the general importance of the contact, and by specifying multiple topics of communica- tion.12 Whereas the focus of Opinion leadership is on directed paths of communication, the focus of communication network analysis in this research is on bi-directed, or non- directed, paths. In other words, the existence of a defined communication link between two individuals implies the pos. sible transfer of information from either individual to the other. In the nine member system illustrated in Figure 6, e a <9 <9 46? o o o a Figure 6. Network Analysis Sociogram with Liaison 42 individual E occupies a key role in the transfer of infor- mation throughout this nine member system, by virtue of being the only communication link between the two groups of individuals in the network-~individuals A, B, C and D form one group; individuals F, G, H and I form the second group. The communication role of individual E in this system has been termed that of a "liaison”--an inter-group linking individual. The other important linking role is that of a "bridge”--an individual who, although the member of a defined group, also functions as a communication link to another group. For example, in the eight member system illustrated in Figure 7, both individuals D and F occupy J e @ e @ Figure 7. Network Analysis Sociogram with Bridges bridge roles. Individual D is a member of the group com— posed of individuals A, B, C and D; individual D also has a direct link to individual F, who is a member of the group composed of individuals F, G, H and 1. Note, in the socio- grams in Figures 6 and 7, that the heads of the arrows used to represent links between individuals in the system have been removed-~denoting the absence of Specified 43 directionality-~and that the definitions of liaison and bridge roles assume that a transfer of information could take place in either direction. Thus, for example, a mes- sage could be transmitted from individual C to individual I, or vice versa; in either case, the message would be trans- mitted along a path which includes the link between indi— viduals D and F. A rank ordering of each system individual, accord— ing to the extent that each individual serves a linking function between other system members, can be achieved by calculating a non-directed centrality index. The proce- dures for calculating a non-directed index are similar to those used in the calculation of the directed opinion leadership index discussed previously. First, the shortest path from each individual in the system to each other in- dividual in the system is identified. In calculating the index score for a specified individual, his shortest one- step paths, two-step paths, and so on, are identified, inversely weighted and summed. This total is then divided by the maximum possible score that could be attained-.(n-l)2 for an individual belonging to a system with n individuals. The calculation of non-directed centrality index scores for the individuals in the system illustrated in Figure 6 is presented in Table 5. Notice, in Table 5, that individual E, the liaison, has the highest index score-~0.891. Individuals D and F each have the next highest index score of 0.875, because a 44 Table 5. Non-directed Centrality Index Calculations for Figure 6 Choice Data Shortest Paths Individual l-step Z-step 3-step 4-step 5-step A 3 l 1 2 l B 3 l l 2 l C 3 l l 2 l D 4 1 2 l O E 2 5 l 0 0 F 3 2 3 0 0 G 2 2 l 3 0 H 2 l l l 3 I 2 2 l 3 0 . . Weighted Shortest Step Paths Weight Spm IndiV1dual lestep 2—step 3-step 4-step 5-step Sum % (n-l) A 24 7 10 4 51 0.797 B 24 7 10 4 51 0.797 C 24 7 10 4 51 0.797 D 32 7 12 0 56 0.875 E 16 35 6 0 57 0.891 F 24 14 18 0 56 0.875 G l6 l4 6 15 0 51 0.797 H 16 7 6 5 12 46 0.719 I 16 14 6 15 0 51 0.797 45 message would have to be transmitted through the samenumber 13--to reach all other connected mem- of total steps-~16 steps bers of the system, regardless of whether the message origi- nated with individual D or individual F. If individual E was the initiator of the message, however, all other connected members of the system could be reached in a total oflfisteps. Hence, individual E is slightly more central to his system as a whole than individuals D and F, and has a higher non- directed centrality index value. The arithmetic steps involved in the calculation of the non—directed centrality index are identical to those re- quired for the calculation of the directed centrality opin- ion leadership index. A major difference in theinterpreta- tion of these two indexes hinges on the differing definitions of a communication link upon which each index is based. The non-directional flow of information that is assumed with the non-directed index requires a reciprocity decision to benmde by the researcher--is it sufficient, in order to define a communication link between two individuals, if only one of the two individuals mentions the other as a contact? Alter- natively, should each individual be required to mention the other as a contact in order for a link to be defined? A de- cision by the researcher to accept the first alternative re- quires the addition of contacts to those reported in the orig- inal data; a decision to accept the latter alternative--re- quire reciprocity-~requires the deletion of contacts from those reported in the original data. The more conservative 46 of the two approaches, that of requiring reciprocity, was used in calculating the non-directed centrality indexes re- ported in this study. The number of contacts added or de- leted would be the same, inasmuch as the researcher haeither completing a link for which half already exists, or deleting the existing half of the same link. Calculating the total number of added or deleted contacts, as a percentage of the total reported contacts, provides a measure indicatingcfither how closely the reported relationships in the data corre- spond to the non-directional relationships that are assumed, and/or the existence of measurement error.14 Respondents in this research were asked to indicate, from a listing of all individuals in their department,those persons with whom they communicated on any of four topic ar- eas--teaching production, teaching innovation, teaching main- . . . 15 . . tenance and profeSSional communication. The descriptions contained in the questionnaire for these four topic areas are presented below: 1. Professional Communication: includes all teach- ing, research and service related communication. 2. Teaching Production: discussions concerning, and the preparation of, course materials, lec— tures, cases, quizzes, examinations; time spent in the classroom. 3. Teaching Innovatigg: discussion of, and the develOpment and use of, new teaching methods and techniques; discussions concerning substantial revisions of course format, materials, content. 4. Teaching Maintenance: conducting office hours; grading student work; assigning grades; student and peer teaching evaluations and feedback. Please note that the four categories abpve are Egg mutually exclusive. Categories 2, Sand 4-- Teaching Ci 47 Production", "Teaching Innovation" and "Teaching Maintenance"--are mutually exclusive and together include all teaching-related communication. These three categories fOrm a subset of Category l--"Profes- sional Communication"--which, as defined in this study, includes all teaching, research and service-related communication. The topic area of primary concern in this researchis the teaching innovation category. Non-directedcentrality in- dexes, with respect to teaching innovation communication, were calculated based on the reported communication network data from each department concerning this topic. hiaddition, non-directed centrality indexes were calculated for a compos- ite of the three teaching topic areas. This composite teaching topic category will hereafter be referred to as "combined teaching;" a communication contact for thecombined teaching network was defined as the existence of a reported contact for 33y of the three separate teaching categories. For example, if individual A listed individual B as a commu- nication contact for teaching production and/or teaching in- novation and/or teaching maintenance, this was considered to be a reported contact withrespect to combined teaching. Responses to the professional communication tOpic category were used solely as a partial check in determining Whether respondents understood the directions supplied in the questionnaire pertaining to the network analysis sections. For example, since the three teaching topic areas were de- fined as subsets of the professional communication category, then if a respondent identified an individual as a contact for any of the three teaching topic areas, he should also 48 have listed contact with that individual in the professional communication category. The reverse, however, is not neces- sarily true, since a reported contact in the professional communication category could have been with reference to re- search or service-related topics. Respondents were also asked to indicate the approx- imate frequency of communication with each listed contact according to the following six point scale: = at least once a day = 2 or 3 times per week 8 about once per week 2 or 3 times per month about once per month about once per term HNtflbU'IO‘ II II When defining a communication link for the combined teaching category, the highest frequency level listed for any of the three teaching topic areas was chosen as the frequency level of the combined teaching link. For example, if individual A reported contact with individual B in the teachingtnoduction category at a frequency level of 4, in the teaching innova- tion category at a frequency level of 2,2uuireported no com- munication with individual B in the teaching maintenance category, then the frequency level designating the contact With individual B in the combined teaching category would be 4. Thus, the frequency levels used for combined teaching represent a lower bound, and conservative, estimate of the frequency of teaching-related communication.16 Non-directed centrality indexes, with reference to both teaching innovation and combined teaching, were 49 calculated at six different sets of frequency levels—~once per term or more; once per month or more; twocn'three times per month, or more; once per week or more; twocn'three times per week, or more; and once a day or more. THunn sixindexes were calculated for each individual for each of the two con- . l7 . . . tent areas. These twelve indexes are summarized in Figure 8. Variable Content Area ngquency Levels Designatiop Once per term or more D7 Once per month or more D8 Teaching 2-3 times per month or more D9 Innovation Once per week or more D10 2-3 times per week or more Dll Once a day or more D12 Once per term or more D14 Once per month or more D15 Combined 2-3 times per month or more 016 Teaching Once per week or more D17 2-3 times per week or more D18 Once a day or more 019 Figure 8. Unweighted Non-directed Centrality Variable Designations In addition, a weighted index was calculated for each individual for each content area by multiplying the six indexes for each content area by inverse weights correspond- ing to the ratios between the different frequency levels represented. The weights used in these calculations are listed in Table 6; a sample calculation follows the table. 50 Table 6. Non—directed Centrality Index Frequency Set Weights Index Semester Quarter Frequency System System Set Weight Weight Once per term, or more 1.000 1.00 Once per month, or more 3.750 2.50 2-3 times per month, or more 9.375 6.25 Once per week, or more 15.000 10.00 2-3 times per week, or more 37.500 25.00 Once a day, or more 75.000 50.00 As an example, the following steps were used to calculate the weighted combined teaching centrality index measure for each individual from School X, a school on a quarter term system. The individual's D14 value was mul- tiplied by a weight of l; the individual's D15 value was multiplied by 2.5; the 016 value was multiplied by 6.25; and so on. The six weighted values were then summed and divided by the sum of the quarter term weights--94.75. The variable designations for the two weighted indexes are listed in Figure 9. Sl Variable Content Area Index Type Designation Teaching H., Innovation htlghtcd U13 Combined . Teaching Weighted 020 Figure 9. Weighted Non-directed Centrality Variable Designations The 20 measures of opinion leadership and network centrality discussed in this section comprise what can be thought of as the dependent, or criterion, variables ana- lyzed in this research. The 42 independent, or predictor, variables are discussed in the following section. Independent Variable Measures Rogers and Shoemaker cite a number of generaliza- tions, culled from the results of many prior diffusion studies, concerning attributes of opinion leaders. To summarize most of these, opinion leaders have higher social status, are more cosmopolite, have greater social partici- pation, have greater change agent contact, and have greater exposure to mass media than their followers. In addition, in modern systems, opinion leaders are more innovative and technically competent than their followers.18 The indepen— dent variables selected for analysis in this study were chosen by applying these generalizations concerning opinion leadership to the social system of higher education in accounting. The resulting 42 independent variables have 52 been categorized as 8 biographic variables, 22 interper- sonal communication variables, and 12 mass media commun- ication variables. Interpersonal communication channels are defined by Rogers and Shoemaker as "those that involve a face-to- "19 face exchange between two or more individuals, and are operationalized in this research in terms of convention attendance, contact with other faculty and contact with publisher representatives. Mass media communication chan- nels are "all those means of transmitting messages that in- volve a mass medium, such as radio, television, film, news- papers, magazines, and the like, which enable a source of one or a few individuals to reach an audience of many."20 The mass media channel variables that are operationalized in this research refer to selected accounting and non- accounting publications. Thus, the interpersonal and mass media communication variable sets measure the perceived frequency of use, and importance, of alternative informa- tion sources concerning new teaching methods. The biOgraph— ic variables are detailed in the following subsection. Biographic Variables The social status of a faculty member is undoubted- ly a function of many different individual and system level variables. For example, factors which denote social status at a large, research oriented institution may have little, or even negative, status implications at a junior or community college; and vice versa. Four variables were 53 selected as possible representations of faculty status at an intitution--highest academic degree held, academic rank, total years teaching and years at the institution. Data for these variables was obtained from the following questions: 1.2 1.4 1.5 1.6 What is the highest academic degree you have received? Bachelor's Master's Doctorate What is your present academic rank? Professor Associate Professor Assistant Professor Instructor or lecturer Approximately how many total years have you been teaching? less than 1 year 1 year, but less than 2 2 years, but less than 5 5 years, but less than 10 10 years, but less than 15 15 years, but less than 20 20 years or more Ill Have you taught at more than one institution within the last ten academic years? Yes . No . (IF NO: Please continue with question 1.7) IF YES: 1.6.1 Please list the institutions at which you have taught, within the last ten academic years, prior to latest employment at your present school. Name of Institution Academic Year(s) Employed One facet of the technical competence of a faculty member is his up-to-dateness and familiarity with new 54 developments having an impact on his academic field. One such development in recent years has been the computer. Measures of the extent of use and familiarity with compu- ters and computer programs were obtained from the follow- ing questions: 3.1 Have you used computer facilities in courses you have taught, academic research or related activities within the last five academic years? Yes . No . (IF NO: Please continue with question 4.0) IF YES: 3.1.1 In which activity or activities have you used these facilities? Courses taught Research Other (please specify): 3.2 Did you write or personally debug any of the programs you used in these activities? Yes . No . (IF NO: Please continue with question 4.0) IF YES: 3.2.1 Approximately how frequently did you write or personally debug the programs you used in connection with these activities? always often sometimes seldom A computer utilization score was obtained by simply counting the number of different types of use mentioned in response to question 3.1; thus, the scale for this variable was zero to three. A frequency of program preparation score was obtained using response data from question 3.2, by weighting an "always" answer as 3, an "often" answer as 2, a "sometimes" answer as l, and a "seldom" answer as 0. This four point scale, as well as the other scales used in this research to measure degrees of frequency and importance with respect to the communication variable sets, were de- veloped by Bass, Cascio and O'Connor.21 55 Innovativeness has been defined by Rogers and Shoe- maker as "the degree to which an individual is relatively earlier in adopting new ideas than the other members of his ”22 where relatively earlier refers to the actual 23 system, time of adoption. Inasmuch as a multiple measure of in— novativeness-—determining the relative earliness of adopting a number of innovations, rather than just a single inno- vation-~15 likely preferable to a single measure, research- ers have often calculated innovativeness scales from time of adoption data pertaining to more than one innovation.24 From interviews with publisher representatives and selected accounting faculty members, as well as from a review of the published education-related literature in recent accounting journals, seven innovations were selected for inclusion in the questionnaire--programmed instruction, modules, view- graph, slides and filmstrips, television, motion pictures 25 and simulation. Time of adoption data for these seven in- novations was obtained using the following questions: 2.1 Have you used programmed instruction or modular course content in any courses you have taught within the last five academic years? Yes____. No . IF NO: Please continue with question 2.2; IF YES: 2.1.1 Please examine the following list and ask yourself: first, have you used it; second, in which years did you use it; and third, was it prepared commercially (C), non-commercially by other persons (0), or did you prepare it yourself (5). For each time you have used an item, enter the appropriate preparation code in the year column corresponding to that use. 56 Prior to Current 8 1970-71 1970-71 1971-72 1972-73 1973-74 Method Programmed Instruc— tion written material teaching machine computer-assisted Modules 2. 2 Prior to 1970-71 Have you used a Viewgraph, slide tran_sparencies or filmstrips in any course you have taught within the last five academic years? Yes . No (IF NO: Please continue with question 2. 35 IF YES: 2.2.1 Please examine the following list and ask yourself: first, have you used it; second, in which years did you use it; and third, was it prepared commercially (C), non-commercially by other persons (0), or did you prepare it yourself (S). For each time you have used an item, enter the appropriate preparation code in the year column corresponding to that use. Current 8 1970-71 1971-72 1972-73 1973—74 Method Viewgraph individual transparencies 2 .3 continuous roll Slides and film- strips without taped sound synchro- nization with taped sound synchronization Have you used television or motion pictures in any course you have taught within the last five academ- ic years? Yes . No . (IF NO: Please con- tinue with question 2.4) IF YES: 2.3.1 Please examine the following list and ask yourself: first, have you used it; second, in which years did you use it; and third, was it prepared commercially (C), non‘commercially by other persons (0), or did you prepare it yourself (8). For each time you have used an item, enter the appropriate preparation code in the year column corresponding to that use. 57 Prior to Current 8 1970-71 1970-71 1971-72 1972-73 1973-74 Method Television live lectures, with feedback live lectures, without feedback pre-recorded audio-visual tapes 2.4 Have you used simulationgprojects in any course you have taught within the last five academic years? Yes . No . (IF NO: Please continue with question 3.0) IF YES: 2.4.1 Please examine the following list and ask yourself: first, have you used it; second, in which years did you use it; and third, was it prepared commercially (C), non- commercially by other persons (0), or did you pre- pare it yourself (5). For each time you have used an item, enter the appropriate preparation code in the year column corresponding to that use. Prior to Current 6 1970-71 .1970-71 1971-72 1972-73 1973-74 Method Simulation business games financial state- ment statistical sampling systems design budgeting and/or control ”T behavioral Two innovativeness-related indexes were employed in this research. The first index was obtained by simply count- ing the number of innovations used-~of the original list of seven--by each individual respondent. The second index was calculated by noting the first indicated use, or lack there- 0f, of each innovation, and assigning a score for each re- sponse using one of the following weights: 58 Table 7. Weights Used in Calculating Innovative- ness Index Academdc Year of Assigned First Reported Use Weight Prior to 1970—71 3 1971-72, 1972—73 2 1973-74, 1974-75 1 Not used 0 The sum of the weights assigned for each innovation for each individual constitutes the innovativeness index.26 Z—scores were then calculated by adjusting individual scores for their respective department's mean and standard deviation. A summary of the eight biographic independent vari- ables measured in this research is presented in Figure 10: Variable Variable Name Designation Highest academic degree 11 Academdc rank 12 Years at present institution 13 Total years teaching 14 Computer utilization 15 Frequency of program preparation 16 Innovativeness 17 Number of innovations used 18 Figure 10. Biographic Variable Designations 59 Interpersonal Communication Variables Cosmopoliteness is "the degree to which an indiv- idual is oriented outside his immediate social system."27 In diffusion studies of rural and peasant societies, cosmop- oliteness has frequently been operationalized in terms of the number of trips by a farmer/villager to urban centers 28 An analogous measure with respect to or other villages. accounting educators is attendance at regional and national conventions. Since the programs of many conventions include formal presentations pertaining to educational topics, the questions used in this research pertaining to convention activity were subdivided into attendance at educational presentations and informal discussions with other faculty, a measure of social participation. Additional measures of the degree of an individual's external orientation to his immediate social system--defined herein as his department-~include the extent of interaction with non-accounting faculty, both in business and non— business fields, at his own school; and the extent of con- tact with faculty at other schools. Since the concern of this research is with facilitating the adoption of newer instructional methods, these measures of external orienta— tion were operationalized in terms of their perceived use and importance as sources of information with respect to new teaching methods and materials. ()0 Finally, an additional interpersonal source of in- formation regarding innovations is the change agent. A change agent is "a professional who influences innovation decisions in a direction deemed desirable by a change agency.”29 One example of a change agent in the context of higher education in accounting is the traveling publisher representative, who attempts to secure the adoption of pro- ucts such as textbooks, filmstrips and simulations marketed by the change agency--the particular publishing house with which the agent is affiliated. The following questions were used to Obtain data regarding the perceived extent of involvement with, and importance of, convention activity, contact with other facul- ty, and contact with publisher representatives: 4.3 Which of the following types of interpersonal contact are sources of information for you with respect to new teaching methods and materials that could be,§_ .05 level. 165 Reduced versions of both equations, however, were significant at p i .01. The full regression equation cited in Table 45 between combined teaching centrality at low and middle fre- quency levels (DFACZ), and the interpersonal factor score variables (IFACS through IFACIO), included all six of the independent variable measures. The coefficients of all in- dependent variables were positive, except for the coeffi- cient of ZFAC9--the perceived frequency and importance of attendance at convention educational presentations--which exhibited a negative relationship. Approximately 11 per cent of the variance is combined teaching centrality at low and middle freqdency levels (DFACZ) was explained by the six independent factor score variables, whereas approxi- mately 12 per cent of the variance in DFACZ was explained by a reduced equation containing all the independent factor score variables except for IFAC10--contact with publisher representatives. The five variable equation was found to be significant at p i .01 in the overall F-test, and will be used as evidence of the primary underlying relationships. Thus, network centrality concerning general teaching-related matters, at low and middle frequency levels of communication, was found to be positively correlated with the perceived frequency and importance of contact with other accounting faculty members; contact with non—accounting faculty members; and with engaging in informal discussions with other faculty members at conventions. A negative relationship was found Table 45. 166 Interpersonal Communication Independent Variable Factors Regressed With Combined Teaching Dependent Variable Factors Dependent Variable: Dependent Variable: Middle Frequency Levels Multiple R 0. 41 R Square 0.16 Adjusted R Square 0.12 Calculated F 3.57 Significance p5.01 Independent Variables: Frequency of contact with non-accounting faculty Frequency and importance of contact with other accounting faculty Frequency and importance of informal discussions with other faculty at conventions Frequency and importance of attendance at educational presentations at conventions Importance of contact with non-accounting faculty Middle FrequencyiLevels Multiple R 0.36 R Square 0.13 Adjusted R Square 0.08 Calculated F 2.73 Significance p§.03 Independent Variables: Frequency and importance of informal discussions with other faculty at conventions Frequency of contact with non-accounting faculty Importance of contact with non-accounting faculty Frequency and importance of contact with other accounting faculty Frequency and importance of contact with pub- lisher representatives J. 0.25 0.20 0.19 -0.13 0.11 0.25 0.23 0.07 0.07 -0.05 Combined Teaching Network Centrali§y_at Low and Beta 0.25 0.20 0.19 -0.13 0.11 Teaching_innovation Network Centrality at Low and Beta 0.25 0.23 0.07 0.07 -0.05 167 to exist with the perceived frequency and importance of attendance at convention educational presentations. Somewhat similarly, the full regression equation, significant at p i .05, between the interpersonal communi- cation factor score variables and teaching innovation network centrality at low and middle frequency levels (DFACS), included all six independent factor score measures. The highest adjusted R2--slightly under 10 per cent--was achieved with the inclusion of just two independent factor score variables--the perceived frequency and importance of participating in informal discussions at conventions (IFAC6), and the perceived frequency of contact with non-accounting faculty (IFAC8). A reduced regression equation containing 1FAC5--the perceived importance of contact with non-accounting faculty--in addition to 1FAC6 and IFAC8, was significant at p i .01; approximately 9 1/2 per cent of the total variabil- ity in DFACS was explained by this regression equation. All variable coefficients were positive except for the coeffi- cient associated with IFAC10--the frequency and importance of contact with publisher representatives-which was negative. Thus, the primary relationships exhibited between the inter- personal communication variables and teaching innovation cen- trality, at low and middle frequency levels, appeared to be between network centrality and the perceived frequency and importance of two interpersonal sources of information-- informal discussions at national and regional conventions, and contact with non-accounting faculty. Additional 168 relationships with teaching innovation network centrality included a positive relationship with the frequency and importance of contact with other accounting faculty, and a negative relationship concerning contact with publisher representatives. As was the case with the results of the regression equations with the opinion leadership measures, the results of the regression equations with reSpect to combined teach- ing innovation centrality confirm and eXpand upon the re- sults of the z-score Pearson correlation analyses. Although only two of the six regressions produced regression equations significant at p i .05, the results of the Pearson correla- tion section identified only one relationship between an interpersonal variable and any of the opinion leadership measures, and no relationships that held specifically for high network centrality levels. As previously mentioned in Chapter 11, very little communication contact was defined at the highest frequency levels; thus, the wealth of rela- tionships found with respect to centrality at the low and middle frequency levels applies to most of the 97 individuah; in the analysis. Mass Media Variables As was the case with the interpersonal factor score variables, only two of the six regressions between the mass media factor score variables and the dependent factor score variables were significant at p i .05. Regrettably, neither of these statistically significant regression equations 169 explained variance in the teaching innovation factor score variable set. The mass media factor score variables were iden- tified as the frequency of use of accounting journals (IFACID the perceived importance of accounting journals as an infor- mation source (IFACIZ), and the perceived frequency of use and importance of Dissertation Abstracts and Collegiate News and Views (IFAC13 and IFAC14, respectively). As may be seen in Table 46, two of these variables-~IFAC11 and IFACl3-r were found to be related to combined teaching centrality at low and middle frequency levels; slightly over 4 1/2 per cent of the variance in DFACZ was eXplained by these posi- tive relationships. Thus, these results indicate that in- dividuals with relatively high combined teaching centrality scores use the accounting journals as an information source regarding new teaching methods more frequently than their colleagues, and have a higher frequency of use, and perceived importance, of Dissertation Abstracts as a source of information. The final regression equation significant at p i .05 explained slightly over 4 per cent of the variance in gen- eral teaching network centrality at high frequency levels. The independent variables included in the regression equathni were IFACll and IFAC12--the frequency Of use and perceived importance of the accounting journals. Thus, individuals more central to their general teaching communication networks at high frequency levels perceive the accounting journals Table 46. Mass Media Communication Independent Variable Factors Regressed With Combined Teaching Dependent Variable Factors Dependent variable: Combined Teaching Network Centralipy at Low and MiddlefiFrequencyiLevels Multiple R R Square Adjusted R Square Calculated F Significance Independent Variables: Frequency and importance of Dissertation Abstracts as an information source Frequency of use of the accounting journals as an information source Dependent Variable: Combined TeachingNetwork Centraliiy at High Frequency Levels Multiple R 0. 25 R Square 0.06 Adjusted R Square 0.04 Calculated F 3.11 Significance pf.05 Independent Variables: Importance of the accounting journals as an information source Frequency of use of the accounting journals as an information source Beta 0.18 0.18 Beta 0.23 0.09 171 as a more important, and frequently used, source of infor- mation for themselves than do the other members of their departments. The results reported in this subsection are not identical to, but do confirm, the results of the Pearson correlation section of this chapter. Although 2 of the 3 Pearson correlations between combined teaching Opinion leadership and the frequency of use of the Education Research and Academic Notes section of The Accounting Review (137) were found to be significant, the correlation between the frequency of use of the accounting journals (IFACll) and combined teaching opinion leadership (DFACl) was not sig- nificant at p i .05. Apparently, when combined with the other accounting journal sources in IFACll, the strength of the relationship between the Education Research and Academic Notes section of The Accounting Review as an information source, and the combined teaching opinion leadership vari- ables, was diluted. Thus, the Education Research and Academic Notes section of The Accounting Review appears to be the only accounting journal source more frequently used as a source of information, with respect to new teaching methods, by general teaching opinion leaders than by their colleagues. The accounting journal sources, as a group, are a more frequently used source of information by individuals with high combined teaching centrality indexes--at all fre- quency levels--and are considered a more important source 172 of information by those individuals central to their net- works at higher frequency levels. In addition, Dissertation Abstracts is both more frequently used, and is considered more important, by individuals central to their network at the low and middle frequency levels. The results of both the Pearson correlation and multiple regression analyses may be summarized as follows. The Education Research and Academic Notes section of The Accounting Review is the only accounting journal source more frequently used by both general teaching opinion leaders, and by individuals relatively more central to their general teaching communication networks. The frequency and importance of Dissertation Abstracts is postively associated with high and middle frequency level combined teaching cen- trality. Given the younger, junior faculty profile devel- Oped in the previous analyses--representing individuals with high combined teaching centrality-~the relationship between network centrality, and the frequency of use and importance of Dissertation Abstracts, might have been expected. Final- ly, the accounting journal sources of information are more frequently used by individuals central to their networks at low and middle frequency levels, and are perceived as more important by the key individuals in networks defined at high frequency levels. 173 FOOTNOTES TO CHAPTER III 1The work by MacDonald and Schwartz examines rela- tionships between liaisons and non-liaisons with respect to demographic and communication variables. The role of liai- son has not been explicitly defined in the present research, but was used as an explanatory concept in the discussion of centrality measures. In communications research, the con- cepts of liaisonness and centrality are related, but not identical, and no commonly accepted liaisonness index has yet been formulated. The work 0 Guimaraes employs communi- cation integration--a measure O the overall connectedness of a system, and a measure which is a system analog to in- dividual centrality measures--as a dependent variable. However, his analysis examines relationships between systems, rather than employing the focus of the present research-- relative individual differences. See MacDonald, "Communi- cation Roles and Communication Content." Schwartz, "Liaison Communication Roles;" and Lytton L. Guimaraes, "Communica- tion Integration in Modern and Traditional Social Systems: A Comparative Analysis Across Twenty Communities of Minas Gerais, Brazil" (unpublished Ph.D. Dissertation, Michigan State University, 1972). 2See virtually any basic statistical text, such as Gene V. Glass and Julian C. Stanley, Statistical Methods in Education and Psychology (Englewood Cliffs,NewJersey: rentice- a , nc., 0), pp. 109-27. 3See Nie, et al., SPSS, p. 281. A two-tail test is employed in this research partly because Of the lack of evidence for predicting the direction of the relationships between the independent variables and network centrality, and partly because of the difficulties involved in interpret- ing the meaningfulness of the size of the correlation coefficients. For example, a correlation coefficient of only .168 is sufficient for statistical significance at the p = .05 level using a one-tail test with n - 2 = 95 degrees of freedom, whereas the same coefficient--.l68--is signifi- cant at only the p - .10 level using a two—tail test. Since the tests presented here are, in fact, simply an aid in the interpretation of the results, rather than being tests of formal statistical hypotheses, use of the two-tail test might be thought of simply as a more conservative approach in interpreting the size of the coefficient. Even when a formal statistical test is used, however, a statistically significant difference does not necessarily imply a meaningful difference. Although a correlation coefficient of .200 is statistically significant at the 174 p -.05 level using a two-tail test and 95 degrees of free- dom, whether the .200 represents a meaningful difference is a matter of judgment. The usual procedure in such a case is to compare the size of the coefficients with the results of similar previous research. As previously mentioned, how- ever, such data are not, to this writer's knowledge, avail- able with respect to the network centrality measures. In addition, suitable data for comparison purposes are unavail- able with respect to opinion leadership in the institutional- ized setting of higher education. It should be recognized, of course, that any corre- lation coefficient differnt from 0 is, in fact, statisti- cally significant in this research, inasmuch as the present analysis examines a population. The formal use of the sig- nificance tests reported here is based on the assumption that the individuals who were analyzed constitute a random sample from a larger population of research significance. 4This result could be considered supportive of the research results reported by Farace and Danowski regarding the perceptions of liaisons and non-liaisons with respect to the perceived number of communication contacts, etc. See Farace and Danowski, "Networks in Organizations." SFred N. Kerlinger, Foundations of Behavioral Research (2d ed., New York: ‘HOIf, Rinéhart afidCWinston, Inc., 1973), p. 659. 6See William w. Cooley and Paul R. Lohnes, Multivariate Data Analysis (New York: John Wiley 8 Sons, Inc., 1971), p. 129. 7Harry H. Harman, Modern Factor Anal sis (2d ed., Chicago: The University 0 icago ress, , pp. 14-15. 8For example, Harman cites the time required with a desk calculator, for the calculation of just the first factor weights in a twenty-four variable analysis, to be more than seventy hours. Ibid., p. 156. 9Kerlinger, Behavioral Research, pp. 667-68. Al- though Kerlinger is speaking here of the principal factor model, rather than the principal component model per se, the geometric analogy would be applicable to all factor analytic models. 10The concept of rotation will be discussed at a later point in this section. 11Harman, Modern Factor Anaiysis, p. 15. 175 12Maurice M. Tatsuoka, Multivariate Anaiysis: Techniques for Educational and PsyéhOTO icaI ReseafEh (New‘YOrk: JehnTWiIeyE Sons, Inc., 19717: pp. I46-48. 131m order to eliminate the difficulties involved in factoring variables with different size scales or ranges, most modern factor analytic solutions are obtained after first standardizing the variables. With n variables, then, the total variability of these n variables will be n times 1 equals n. Thus, the total variability-~n--divided by the number of variables--n--represents the average contribution of any single variable toward the total variability of all variables in the set. 14Raymond B. Cattell, "The Scree Test for the Number of Factors," Multivariate Behavioral Research, Vol. 1 (April, 1966), pp. 245-76. 15R. J. Rummel, Applied Factor Analyeis (Evanston, Illinois: Northwestern UniverSityPress,11970), pp. 364-65. Rummel mentions the discontinuity test in conjunction with the common factor model, an alternative approach to the principal component model. However, since the decision re- garding the number of factors to be retained must be made be- fore rotation regardless of the model employed, the test for discontinuity could have potentialtitility with either model. 16See for example, Tatsuoka, Multivariate Analysis, pp. 146-48; and Chapter 15 of Rummel, AppliedFactor Anaiysis, pp. 349-67. 17 See Harman, Modern Factor Anaiyeis, pp. 97-99. 18Robert Libby, "Prediction Achievement and the Use of Simulated Decision Makers In Information Evaluation" (unpublished Ph.D. Dissertation, University of Illinois, 1974), p. 62. As noted by Libby, a complete discussion of this procedure is contained in Harman, Modern Factor Analysfln pp. 304-13. 19 Kerlinger, Behavioral Research, p. 671. 20See Ibid., p. 673, among others. 21The factor scores used in this research were calcu- lated by the SPSS Factor routine. Details of the procedure employed may be found in the Nie, et a1. SPSS manual: SPSS, pp. 487-90. By a "true factor score" is meant a method of calculating the new variable value using all the original variables to some degree, depending upon their loadings on the specific factor for which the score is being calculated. This procedure may be contrasted with approaches in which 176 only a single variable, or subset of variables, is used in the computational procedure. The use of a single variable, called the basic variable approach, has the advantage of preserving experimental reality, but requires that the variable be loaded very highly on the factor and allows relatively highly intercorrelated variables to be chosen to represent the different factors. This introduces the pos- sibility of multicollinearity problems if further analysis such as multiple regression is to be used. 22As mentioned previously, there have been virtually no substantive applications of the diffusion research or network analysis methodologies in the context of innovation in higher education. The variables used in this research were selected based on preliminary interviews, a review of related research, and the operationalization of constructs from the diffusion and network analysis research traditions. Since this dissertation is primarily an exploratory effort, it was decided to use a method of factor representation aimed at identifying significant dimensions in the total variability of the variables used--generation of true factor scores--rather than procedures such as the basic variable method. 23"Tended to indicate" in this context refers to the difficulties involved in actually applying the scree test. Since the method involves the subjective determination of when a graphed curve starts to flatten out, the method is imprecise in situations where the curve does not have marked discontinuities. 24It may be of interest to the reader, after examin- ing the three factor solution presented in Table 42, to know that the four factor solution followed a pattern similar to the three factor solution. Specifically, whereas the three factor solution will be shown to yield factors repre- senting opinion leadership, low centrality levels and high centrality levels; the four factor solution yielded factors representing opinion leadership, as well as low, middle and high centrality frequency levels. The major points dis- cussed in Chapter III of this dissertation regarding the three factor solution would also be applicable to the four factor solution. 25See Chapter 5 of Rogers with Shoemaker, Communica- tion of Innovations, pp. 174-96, for a discussion of’fhe adopter categorization scheme based on innovativeness. 177 26The scree test was even more difficult for this researcher to apply to the combined teaching dependent variables than for the teaching innovation dependent vari- ables. Results of the scree test again inidcated a four factor solution as being the most appropriate, whereas application of the Kaiser-Guttman rule and Thurstone's structural clarity criteria suggested a three factor solution. 27Tatsuoka, Multivariate Analysis, p. 26. 28For a discussion of the relationship between "b" and "beta," see the excellent basic reference to regression by John Neter and William Wasserman, Applied Linear Statis- tical Models: Regression, Analysis offivariance and‘Experi- mentaTTDesi ns (Homewood, IIlinois: Richard‘D. Irwin, Inc., , pp. -68. 29Nie, et al., spss, p. 345. 3oibid.,p. 346. 31See, for example, Ibid., pp. 334-40, as well as Neter and Wasserman, Applied Linear Models, p. 228. In the tables of this section, the calculated F statistic for each significant regression equation is given; in addition, and of more potential utility to the reader, the calculated significance levels for these F tests are reported. Actual significance levels for these F-tests were calculated using a computer program contributed by Professor Andrew Snyir, of the Pennsylvania State University, to whom appreciation is expressed. 32Maurice M. Tatsuoka, Validation Studies: The Use of Multiple Regression Equations, SelectedTopics in Advanced Statistics: Ah Elementary Approach, Number 5 (Champaign, Illinois: The Institute for Personality and Ability Testing, 1969), p. 11. 33Ibid., pp. 11-12. See also, for example, Neter and Wasserman, Applied Linear Models, p. 229. It might be noted here that the‘"adjustedR4"‘reported in version 6.0 of SPSS does not use the formulas cited by ei her of the above sources in calculating the adjusted R . The formula used by SPSSzpresents a slightly more liberal--closer to the origi- nal R --adjusted R2 than most published sources. The adjusted R2 statistics presented in the tables in this section use the formulas specified by Tatsuoka, Neter and Wasserman, and others. In a recent newsletter, SPSS has announced that it intends to revise the formula used in their calculations to conform with the more accepted version. 178 34Further, there is theoretical justification for expecting non-linear relationships between opinion leader- ship, and at least some of the independent variables, to exist. See Rogers with Shoemaker, Communication of Innova- tions, p. 190. However, identifying the best fitfing, and most useful, polynomial functions as expressions of the relationship between the dependent and independent variable factors, and variables, in a very complex task that is con- sidered outside the scope of this exploratory research. This writer has, in fact, started examiming these relation- ships and found much greater adjusted R s--e.g. up to 20% with just one of the independent standardized variables-- than are reported in this research. However, the complexi- ties of identifying the best, most general, and most useful transformations make this further investigation a worthy research project in its own right. 35As a means of assessing how much is, in fact, lost by splitting the independent variable sets and using single dependent variable factors, it may be of interest to the reader to know that the first significant canonical variate alone, in a canonical correlation of the 14 independent variable factors and 6 dependent variable factors, yielded a canonical correlation coefficient Of .639, significant at less than the 2 per cent level, which explained 40.8 per cent of the total variance. Further, it should be remembered that up to 30% of the variability within each independent variable set was lost by extracting only the significant uncorrelated factors; itis likely that a larger portion of the total variability would be retained with more complex factor models, such as those suitable for oblique rotation procedures, where the resulting factors are allowed to be correlated. CHAPTER IV SUMMARY AND CONCLUSIONS Summary Chapter IV of this research is devoted to an expo- sition of the overall results of the previous data analyses; a conclusions section, in which the research results are applied in the context of the existing problem area; a section mentioning some of the limitations of the current research; and a brief final section Offering suggestions for the direction of further research in the problem area of achieving increased implementation of existing, or future, innovative teaching methodology within accounting higher education. Methodology Forty-two independent, and twenty dependent, vari- ables were operationalized in Chapter II. The independent variables were categorized as 8 biographic, 22 interpersonal communication, and 12 mass media communication variables; each variable was standardized within each school resulting in a relative measure of the differences between the 97 individuals, from 8 schools, who formed the respondent set analyzed in this research. The 20 dependent variable 179 180 measures, which were also standardized within each school, consisted of 6 measures of opinion leadership and 14 mea- sures of network centrality. Half of the opinion leadership variables, and half of the network centrality variables, pertained to communication regarding new teaching methods and materials; the remaining halves of the opinion leader- ship and network centrality variable sets pertained to all teaching-related communication. For the convenience of the reader, a listing of the variable name and designation of each of the 62 z-score variables is presented in Figure 13. Initially, the existence of linear relationships between all independent and dependent z-score Variables was estimated by Pearson product-moment correlation coeffi- cients. The significant relationships which were identified are listed in Table 47. The relationships within the variable sets was then explored by conducting a principal components factor analy- sis for each of the following variable sets--biographic independent; interpersonal communication independent; mass media communication independent; teaching innovation, or teaching methods, dependent; and combined teaching dependent. A determination of the number of significant dimensions within the variability of each of these five variable sets was made by determining the number of significant factors. Four significant factors were extracted from the 8 biographic variable set; 6 factors were retained from the 22 interpersonal communication and variable set; and 4 significant Designation 11 12 13 I4 15 16 I7 18 181 Figure 13 BIOGRAPHIC INDEPENDENT VARLABLES Variable Name Highest academic degree Academic rank Years at present institution Total years teaching Computer utilization Frequency of program preparation Innovativeness Number of innovations used INTERPERSONAL COMMUNICATION INDEPENDENT VARIABLES Designation 19 110 111 112 113 114 115 116 117 Variable Name Frequency of attendance at educational presentations at national conventions Frequency of participating in informal discussions with other faculty at national conventions Frequency of attendance at educational presentations at regional conventions Frequency of participating in informal discussions with other faculty at regional conventions Importance of attendance at educational presentations at national conventions Importance of participating in informal discussions with other faculty at national conventions Importance of attendance at educational presentations at regional conventions Importance of participating in informal discussions with other faculty at regional conventions Frequency of participating in discussions with publisher representatives 182 Figure 13.-~Continued Designation Variable Name Importance of participating in discussions with 118 publisher representatives 119 Frequency of participating in discussions with other accounting faculty at own school Frequency of participating in discussions with 120 non-accounting business faculty at own school 121 Frequency of participating in discussions with non-business faculty at own school 122 Importance of participating in discussions with other accounting faculty at own school 123 Importance of participating in discussions with non-accounting business faculty at own school 124 Importance of participating in discussions with non-business faculty at own school 125 Frequency of participating in discussions with other accounting faculty at other schools 176 Frequency of participating in discussions with ‘ non-accounting business faculty at other schools 127 Frequency of participating in discussions with non-business faculty at other schools 128 Importance of participating in discussions with other accounting faculty at other schools 129 Importance of participating in discussions with non-accounting business faculty at other schools 130 Importance of participating in discussions with non-business faculty at other schools MASS MEDIA CO‘vMUNICATION INDEPENDENT VARIABLES Designation variable Name 131 Frequency Of use of Collegiate News 8 Views as an information source 132 Frequency of use of Dissertation Abstracts as an information source Oesignation I33 I34 I35 136 137 I38 I39 I40 I41 I42 Designation DI DZ D3 D4 D5 D6 183 Figure 13.--Continued Variable Name Importance of Collegiate News 8 Views as an infor- mation source Importance of Dissertation Abstracts as an infor- mation source Frequency of use of the Book Review section of The Accounting Review as an information source Frequency of use of the Education and Professional Training section of the Journal of Accountancy as an information source Frequency of use of the Education Research and Academic Notes section of The Accounting Review as an information source Frequency of use of the Committee Reports Supplement to The Accounting Review as an information source Importance of the Book Review section of The Accounting Review as an information source Importance of the Education and Professional Training section of the Journal of Accountancy as an infor- mation source Importance of the Education Research and Academic Notes section of The Accounting Review as an infor- mation source Importance of the Conmittee Reports Supplement to The Accounting Review as an information source OPINION LEADERSHIP DEPENDENT VARIABLES Variable Name Unweighted general teaching Opinion leadership index Weighted general teaching opinion leadership index Directed centrality general teaching opinion leadership index Unweighted teaching methods opinion leadership index Weighted teaching methods opinion leadership index Directed centrality teaching methods opinion leadership index Designation D7 D8 D9 010 D11 D16 017 018 019 DZO 184 Figure 13.--Continued NETWORK CENTRALITY DEPENDENT VARIABLES Variable Name Teaching innovation centrality at once per term or more Teaching innovation centrality at once per month or more Teaching innovation centrality at 2-3 times per month or more Teaching innovation centrality at once per week or more Teaching innovation centrality at 2-3 times per week or more Teaching innovation centrality at once a day or more Weighted teaching innovation centrality index General teaching centrality at once per term or more General teaching centrality at once per month or more General teaching centrality at 2-3 times per month or more General teaching centrality at once per week or more General teaching centrality at 2-3 times per week or more General teaching centrality at once a day or more Weighted general teaching centrality index Figure 13. Complete Listing of Standardized Variable Names and Designations 185 Table 47. Sumnary of Significant Relationships Between Independent and Dependent Variables Biographic Variables _Pearson Correlation Multiple Regression Dependent Variables Inde- Direction Inde- Direction pendent of Rela- pendent of Rela- Variable tionship Variable tionship 01, 02, D3 12 + IFACZ + DFACl 17 + IFACl + IFAC4 + 04, 05, 06 17 + IFACZ + 0FAC4 18 + IFACl + 014, 015, 016, 017 IFACl - DFACZ 12 - IFAC2 - I3 - IFAC4 + 018, 019 14 - IFAC3 + DFAC3 ' 08, 09, 010 DFACS 12 - IFACl " r 13 - IFAC3 + 011, 012 I4 ’ DFAC6 J 186 Table 47.-~Continued Interpersonal Communication variables Pearson Correlation MulpipleRe ession Dependent Inde- DirectiOn Inde- irection variables pendent of Rela- pendent of Rela- Variable tionship variable tionship 01, 02, D3 - - DFACl D4, 05, D6 117 + - DFAC4 110 + a 112 + 014, 015, 016, 017 114 + IFAC8 + DFACZ 116 + IFAC7 + r 119 + IFAC6 + 018, 019 125 + IFAC9 - DFAC3 J 128 + IFACS + 120 + 121 + 110 + IFAC6 + 08, 09, 010 112 + IFAC8 + DFACS 114 + IFACS + 116 + IFAC7 + D11, D12 119 + IFACIO - DFAC6 120 + 123 + 187 Table 47.--Continued Mass Media Communication Variables Dependent Pearson Correlation Multiple Regression Variables Inde- Direction Inde- Direction pendent Of Rela- pendent of Rela- Variable tionship Variable tionship 01, 02, D3 137 + - DFACl 04, 05, D6 - _ DFAC4 014, 015, 016, 017 132 + IFAC13 + 0FAC2 141 + IFACll + f 137 + 018, 019 136 + IFACIZ + DFAC3 j 138 + IFACll + 08, 09, 010 _ _ DFACS 011, 012 _ _ DFAC6 188 factors were extracted from the 12 mass media communication inde- pendent variable set. Three significant factors were extracted from each of the 12 z-score, dependent variable sets pertain- ing to teaching innovation and general teaching-related com- munication. Varimax rotation was applied in order to clarify the structure of each significant factor; each final factor was then identified by noting which of the original z-score variables correlated most highly with that factor. A listing of the name and designation of each of the 20 significant fac- tors is contained in Figure 14. Factor scores were calculated, for each of the 97 individuals for each of the 20 signifi- cant factors, thereby creating 20 new factor score variables representing the significant components of the variability within the z-score variable sets. The relationship between the independent variable factor score sets and each significant dimension in the variability of the dependent variable sets was examined using multiple linear regression procedures. The set of significant factors generated from each z-score, independent variable set was regressed with each Of the 6 dependent variable factors; 8 of the 18 separate regressions resulted in regression equations found to be significant at the p 1 .05 level. A summary of the relationships between inde- pendent and dependent variable factors, as contained within these regression equations, is given in Table 47. The overall summary of the results which follows has attempted to combine the most important, and consistent, ( 189 Figure 14 BIOGRAPHIC VARIABLE FACTORS Designation Factor Name Primary Variables IFACl Institutional seniority 12, 13, I4 IFACZ Innovativeness I7, 18 IFAC3 Computer familiarization 15, I6 IFAC4 Highest degree held 11 INTERFERSONAL CO‘vNUNICATION VARIABLE FACTORS Designation Factor Name_ Primary Variables IFACS Importance of contact with non-accounting faculty 123' 124: 129. 130 Frequency and importance of IFAC6 informal discussions with 110, 112, 114, 116 other faculty at conventions Frequency and importance of IFAC7 contact with other accounting faculty 119, 122, 125, 128 . Frequency of contact with IFAC8 non-accounting faculty 120’ 121’ 126' 127 Frequency and importance of IFAC9 attendance at educational 19, 111, 113, 115 presentations at conventions Frequency and importance of IFACIO contact with publisher 117, 118 representatives 190 Figure 14.--Continued MASS MEDIA COMMUNICATION VARIABLE FACTORS Designation Factor Name Primary Variables Frequency of use of the IFACll accounting journals as 135, I36, I37, 138 an information source Importance of the accounting IFACIZ journals as an information 140, 141, 142 S OUTCC Frequency and importance of 1FAC13 Dissertation Abstracts as 132, 134 an information source Frequency and importance of IFAC14 Collegiate News 8 Views as 131, 133 an information source CDMBINED TEACHING DEPENDENT VARIABLE FACTORS Designation Factor Name Primary Variables DFACl Combined teaching Opinion 01, 02, D3 leadership Combined teaching network DFACZ centrality at low and 014, 015, 016, 017 middle frequency levels Combined teaching network DFAC3 centrality at high 018, 019 frequency levels 191 Figure l4.--Continued TEACHING INNOVATION DEPENDENT VARIABLE FACTORS Designation Factor Name PrimaryVariables . . Teaching innovation Opinion DFAC4 leadership 04, 05, 06 Teaching innovation network 0FAC5 centrality at low and D8, 09, 010 middle frequency levels Teaching innovation network DFAC6 centrality at high 011, 012 frequency levels Figure 14. Complete Listing of Factor Score Variable Names, Designations and Primary Variables 192 research findings from these separate analyses into a unified whole. Opinion Leadership First, it can be said that opinion leadership, as traditionally measured in diffusion of innovations research, exists within higher education in accounting. A profile of opinion leaders as being individuals both relatively more innovative, and more senior in their organizations, than their fellow accounting faculty members was developed from the results of the Pearson correlation and multiple regres- sion analyses. Second, the information sources most frequently used, and considered important, by opinion leaders were, for the most part, the same as for their colleagues. The only inter- personal communication source more frequently used by opin- ion leaders, than by their colleagues, was contact with publisher representatives; the only mass media source used more frequently by opinion leaders was the Education Research and Academic Notes section of The Accounting Review. Finally, there is substantial evidence to support the contention that the role of being an opinion leader is quite distinct from the role of functioning as an important link in the day-to-day communication activities within an accounting department concerning teaching-related matters. First, the results of the factor analyses of the dependent variable sets indicate that the only overlap between opinion 193 leadership and network centrality occurs at very low fre- quency levels of communication. Hence, it appears that although individuals may be sought out for information or advice regarding many types of teaching-related matters, the opinion leaders are not the same individuals who com- prise the core of the network participants in their depart- ment at even moderate frequency levels of communication. Further, the only overlap between opinion leadership and network centrality, regarding new teaching methods or teach- ing innovations, exists at the lowest frequency level on the measurement scale used in the personal contract listing-- once per term. Second, as will be seen in the following subsection, the characteristics of opinion leaders are in marked contrast with the characteristics of individuals with relatively high centrality indexes. Network Centrality In contrast to the characteristics of those persons who function as opinion leaders in their systems, individuals who play central roles in their departmental communication networks concerning teaching matters are relatively junior in the organization--in terms of academic rank, total years teaching and years at the institution. Further, such in- dividuals have more familiarity with computers and, on the average, hold somewhat higher degrees. The most important interpersonal communication sources for individuals with high centrality measures are £91 aT. u.i .. « in »10130 v3i121ano: 110‘? 7r , »r v'" ,“ZHSH .flOIJflJIflUflIOS 3° 5‘ 10 8103 9d! 1.“ 413m 39V, 3.7” ' I anoufli ant 'C 1 Tuam91ueael 3;;1 (sq QDHO .. ,Wwil)92dua wuznu: 06118. '.:s131 diiw forjanui Odfl L"" ‘ - . “ “fins; vnfq ofll . ‘2’ '1“""*-" - . ..' . inZSTHOD aflrowiafl iififix itio: v*“*“ .i: v: "JVTESIUEQTO ad! It ,‘fli dang . .7 "1 '1 -:-2.- . , - A u .' ,isn.io. .... s ,- a? is aisev has In _i. *L'9-sbflfl RTOJuqmoa nJiw (iirailimsi atom ovsd I . ‘ ‘ ,_ is *1 194 informal discussions with other faculty while at national and regional conventions, and contact with other account- ing and non-accounting faculty at their own schools. In addition, there is evidence from the multiple regression analyses that, while contact with faculty at other schools is relatively more important to individuals with high net- work centrality indexes; contact with publisher representa- tives, and educational presentations at conventions, are relatively less important as sources of information for these individuals than for the average respondent. Finally individuals who are central to their depart- mental communication networks, with respect tO general teaching-related communication, perceive the accounting journals-~particularly The Accounting Review--and Disserta- tion Abstracts as being relatively more important sources of information with respect to new teaching methods and materials. None of the mass media sources, however, were considered relatively more important by individuals with high centrality measures pertaining to teaching innovation, than by the average respondent. Thus, it may first be concluded that communication networks may be defined, with respect to the communication that occurs between members of accounting departments, con- cerning teaching innovation and general teaching-related topics. Second, the characteristics of individuals with relatively high centrality indexes suggest that the linking $91 ane‘Tnx :~ . . uiiuuni Tudiu “It! amok; , .enoiaaovutam '“L'i)fi bflc . t: wilu;ui gniaauoaalb «i a r Dive 21 010d3~§-I .5 RIP“ 9.14w .JIdJ C 16' 1 on VISVIII ,- uirileHOQ ”imbe bflfl .8~ ‘ . s19v13810(»~ :ibni 520104 ll u.fd ' ‘ 11m 183a!- p gLEdDBBJ s -2Jen1uoi 3| '2uA n01: , amtélflf 30 EILITGJBN I :siehfianoa A'JWTHaE dgid v; uni yd med: . ; an ‘~ in.iv 7 'i “r' ;. : “ J ' ': ,wufiT Al , I 30133“in;mmw; ', c- *.-' :“w ,::w7 ”a ed vim ziTOU“!»C ’T _,. , . I U e23? ,aansmiisqeb gnLInUOJuL lo oiedmox neawied aruaadfkiggif‘“_ ‘ I ' ' one: Intensg bne noiiavonnl gnldalqtflfSI '9 ”a a.1- , . L12 hl’ ‘: 195 function, in the transmission of information regarding teaching-related topics, is performed primarily by younger, junior faculty members. Further, the characteristics of these individuals differ from the characteristics of indiv- iduals who function as opinion leaders. Third, individuals with high centrality indexes-- whether with respect to teaching innovation or general teaching-related matters-~both access, and consider more important, the available interpersonal communication channeh; as sources of information. In addition, individuals with high centrality indexes pertaining to general teaching- related topics both use, and consider more important than does the average reSpondent, the published information sources--specifically, the accounting journals and Disserta- tion Abstracts. Conclusions Conclusion 1. Contact with other accounting educa- tors, and The Accounting_Review, are the Only interpersonal communication and mass media information sources, respec- tively, to be considered even moderately important as.sources of information regarding new teaching methods by the average respondent. The non-standardized mean importance scores fer all respondents, on a scale from 0 to 4, for the group of 11 interpersonal communication and 6 mass media communication variables, are listed in Table 48. In order that the 301 ,M:. . i'rmitini To neieainensis ms-t it; . ,2iiqoi 503.3 , ..: a zisdmefl 2308‘ .gilih aiaubtvl' .. : ._r. notional adm- 1211;? ,bridT ...~ .f153701ll ;u.ruoe 83 1’3 ”'"JD “Sid 'vt $933191 :d? BQOb -:soruoe a: ' . dA noi: .‘H'I. .970: ‘ii '., vi 1 ringinummoa ~- Jzafarr . ’ . _ , .:; at ca .Usvl: TSEEEK; “L? E“ ‘ VJ n .-:q:;“ r;7 <.iu'~357 noiism1olni 19 ‘ .msbaoqm- } “41% 54‘} .,> ”r. b.1y , 11° '03 3°1°33 93HBI’0qwi nssm Lesrbrsbneie- non odT .efing.r ‘ ‘l “Mgwroi 0030.011 almanac. 196 Table 48. Importance of Interpersonal and Mass Media Information Sources for the Average Respondent Non-Standardized Variable Mean 0232:1323: 4.000 Extremely Discussions with Accounting . Colleagues from your School 2’776 Con51derably Education Research and Academic 2 275 Notes of the Accounting Review ° Discussions with Accounting 2 260 Faculty at Other Schools ' Comittee Reports Supplement 2 215 to the Accounting Review ° Book Review Section of the 2 105 Accounting Review ° 2.000 Moderate Mean responses for the other 12 non-standardized interpersonal commmication, and mass media commmication, variables were all below 2.000 OQI IP‘JI'TVJIIII €333 21v' and" U18 Innonaq‘roial 30 ., ‘rtaimaqasp. 5331mm ed) to} I" r'nlu'v'fiA IIJIW m1 ' 100'; ram} 2 . , {thaw-.33 not: v I. rug". 3!: 3f” 10 u 1216123113810 It :5 inland. " JEIJIW "mi 9d} m . 1:251. .1008 .11-)“14033A ' «fight-i y {I ' f. .L j l! ‘ - n‘fi’ . 1“! o 41‘“ ‘ . . .i ,. ' ." ',.milr1\nl‘.’_ 197 reader may interpret these means, importance descriptors, derived from the Bass, Cascio and O'Connor listings,1 for various mean levels are also presented in the table. An examination of these non-standardized mean importance ratings, for each of the interpersonal and mass media commu- nication sources, indicates that The AccountinguReview is the only mass media source to be considered even moderately important by the average respondent. Of even more interest is the fact that the only interpersonal communication sources to be rated at least moderately important by the average respondent were contact with other accounting facul- ty members, both at an individual's own school and at other schools. To this researcher, these are extremely important results. First, the fact that The Accounting_Review, and contact with other accounting faculty members, are the only information sources considered even moderately important by the average respondent suggests that the dissemination of information and influence regarding new teaching methods and materials is likely to be spread through these channels. Second, the primary importance of the interpersonal channel-- contact with other accounting faculty members at an indi- vidual's own school--is empirical justification for this re- search, which has sought to identify and analyze character- istics of the individuals who play key roles in the inter- personal communication channel--the opinion leaders, and the individuals central to the communication networks Y‘s! .n 1:“.710qmi ,huBOfl 089d: 3.?J 1"0 has niiasD .eanl ,_ ,. n «p:. ers aIevsI 75:« vi. need? 30 “01" . ;, wwg“ -Mi 23 dune 103 1-123 ad: a}... '5 hi 29310601 ,.w‘ryu 4 _: 53519Vl .awodmsm Y3 .efoodaa .LTIUZBT .r' I 25131103 nee noitsmioini ‘ . V ‘I ,, . ncxulqo lo nosiandi' R .a.fi vileaiqvj o 'wc, 118 To qldat'l, .d‘ ,{ioodsr 8 0d, ,. ‘a ii ' “izaisbnol hoist-‘ ) , flail. "11.11.“: '0 NW zisdo 9(4) , .ifl .2quo1; .uv 1m xobnl .‘2h ad) \ :n,yv nummov ;' . _ p 9'19" r "yfl'fUflIfflOD :: ;;ob Yd -' ., .1 1"51'1‘3151’13 ‘x'x‘uv 15 u - Joys:,jnummoa oon::s. w“ , g 4 . ’ 7 ' “x -.;3nanpuii 310. -xe hidfin giro, .- ‘,H; ~.n .._‘ 3' ,:,-_' .ytuupQTi 3. bobuioxs an blusm fliuiififl- .1» 'K'mflhfi‘ vi? To Jaon 953‘? ; ‘, f! .ZIOVSI (anoupuii Tewoi have In hsuilob aixowJOfl .‘ii:’ in: a) 3 . a. 5 ' .LVN; .~ “hive-119a no: main 2310301 oaoflfigjf. 200 to this researcher, that department chairmen would necessari- ly be the best targets when designing a diffusion strategy. First, a change agency--used here as representing any indi- vidual, group or organization attempting to secure the adop- tion of a teaching innovation--with limited resources, might very well choose to designate some maximum percentage of the individuals within a given system as targets for their promotion strategy. The use of a maximum percentage such as 10 per cent, would result in the selection of from 1 to 3 individuals for most departments of accounting in the United States. Although almost all department chairmen were in the upper third of their department, with respect to relative opinion leadership rankings, only 2 of the 8 chairmen, in the departments analyzed in this research, would be selected as targets using a 10 per cent criterion. Furthermore, it is likely that the relatively high average opinion leadership ranking of the department chairman is at least partly a function of the fact that they are chairmen, and thus likely-viewed as influentials by younger, junior faculty. It does not necessarily follow that department chairmen would be viewed as influentials with respect to teaching matters by relatively senior faculty. If the decision-making process regarding a particu- lar innovation is largely authoritarian in nature--the department chairman either makes, or heavily influences, the decision--then department chairmen would be key individuals. lariV= .‘m.‘ ‘ ' - . 2 s h;7voqsi nemiisda "#1131"! ““7 I V g 7' . -‘. ""1' W ‘ 11 'v'JI ‘(XSV 3‘ 5*‘35’311’3‘9 7'3 . .«xi'. :4; s - ‘1111'11;:§’i" itOiSSjiflUMGD " "4r . ~ .i' ' MIN-911m .. .. «.sm timeuperi ato- 59501319 3d blucw flQMTIBH; saga :flsmrisdg 9d, 30 3805 Qbflfi,‘j .45.» v.7 .wloodae 8 or”: iii»: cidaiebasl not“! I ”’0 mad nanrtafl: 'u k. . ll JFIIISI‘IJ O“; J , . . ...uH .aquoin uasom xobni firm, snub 9d! 'uiJLDInUMWOD , .,;':;' 3310;: 919i! M's ';ii_"5r:iinumloa ‘ ’i- ’“ . ' ~ _ “1:;Aa :rafli-sqeb vd 'XB hIUDW 111(1):?! ;.- _ ZJQD (ig:*_ “rm-'15., 70 aI'JYJI "(3fi9Up91* 3‘ .alovel yonsupori rswol nsvs in beniieb fi’ni a: snobive- ~1192 mesa 353:. 2:10:01 200 to this researcher, that department chairmen would necessari- ly be the best targets when designing a diffusion strategy. First, a change agency--used here as representing any indi- vidual, group or organization attempting to secure the adop- tion of a teaching innovation--with limited resources, might very well choose to designate some maximum percentage of the individuals within a given system as targets for their promotion strategy. The use of a maximum percentage such as 10 per cent, would result in the selection of from 1 to 3 individuals for most departments of accounting in the United States. Although almost all department chairmen were in the upper third of their department, with respect to relative opinion leadership rankings, only 2 of the 8 chairmen, in the departments analyzed in this research, would be selected as targets using a 10 per cent criterion. Furthermore, it is likely that the relatively high average opinion leadership ranking of the department chairman is at least partly a function of the fact that they are chairmen, and thus likely viewed as influentials by younger, junior faculty. It does not necessarily follow that department chairmen would be viewed as influentials with respect to teaching matters by relatively senior faculty. If the decision-making process regarding a particu- lar innovation is largely authoritarian in nature--the department chairman either makes, or heavily influences, the decision--then department chairmen would be key individuals. 00$ ‘.‘i I1JmJinqsb tad? .1.‘ ‘ if ..;n "~ » unlugt2"fi nadw 23331.3'3“..7TH ' i is 57:3 r In -v3noga Dflfl‘dyw‘” vu.’ IIHEHTO 1o quata‘}. 1. guidanel A 3-9' ! ahead: [19"- ..1 A <;hubivibfli‘0‘1: ‘JBII? Hello-déi‘, ,.i'v :sq 01 0‘ ‘4' -JinR 8 r;? Lelia“ 1 view ifrlfl 01 v, .ranwrada . ..r~ -4 riuow i »g ‘mrodiiu? . . , ' . ep;i H'EDIQD i, "i .313" JZBQI ”t: .'t a “ - , ', - '.‘ ‘i zed! bfil tusnitaich =;‘- , .... , ,4 ,,M; 3‘ .yjfuaii O.‘ 135”?“ A -'5 : ' ’i' . 1‘.|'.J'—." I ,1, 1. “y."— ~‘ '1' bf ,rr‘i'd flM‘Ii‘db .H‘iuoui torn > ri3v115331 Y9 fizgjjfim guidaltl_g; . g". ' ' - . ‘ .. 'v‘N" n.4- 71-134 3 80151889; aesaorq gn1Xsm-10iaisib 9d: 11 'N.'3;--y - ”,g.;3!fil¥§i at astraziiodsua 1193181 at . fa." 201 An example of this type of decision might be a decision re- garding use of an innovation requiring substantial depart- mental commitments or resources, such as the use of instruc- tional television.2 However, to the extent that the decision regarding use of the innovation could be made by the indi- vidual faculty member and not be upon the direction of the chairman, the informal channels of influence represented by opinion leadership would be important. Many, perhaps most, of the available teaching innovations would be in this category-~innovative textbooks; the use of visuals such as slides and filmstrips; innovative organization of course material, such as in modules; the use of cases, simulations, and so forth. By virtue of their position, department chairmen may function as gate keepers in their systems, and thereby be able to increase or prevent, at least to some ex- tent, the adoption of certain teaching innovations within their systems.3 Second, at the persuasion stage of the innovation- decision process, the interpersonal channel of communication becomes relatively more important.4 Thus, the low frequency levels of communication reported by most chairmen might tend to make chairmen relatively poor candidates for assisting at the persuasion stage. Department chairmen might, however, be ideal candidates for assisting at the awareness stage of the innovation-decision process--the simple spreading of information regarding an innovation--by virtue of their 10f ,1. I. . ,5 .4 ‘ 15.7: 31119.; «b '10 “ff! CM” ;',. ufl' JOI'SVOflfll a. Iii ,« ., up n—e't 10 amen“,- ‘. s. ., , »v (an S.noiclvvl¢3,d lo sen gaj yxluosi J .<.: em .ann ‘ nub-’11 Hakim" ~.» age 9d! 1’ ‘ -,~; y10393¢0 nu aebill a ‘ ,J;j1918l :‘, r-,; r~a brta m Jami} aria an vdsxsd: 4;? .Inei 4‘.@¢ 1?653 ;§*Hi_.‘,afl.. ; T 7, ,3q HofaiDSh '-'Z).’l.”,‘!w?", .- .411" ‘ W. . ‘ minis” eemoaod bflSJ 352.? flam1.:t .h c u'f""J ';f H" uhmeJ 30 819'.1 33 33132132: 113 233;Lib;e -;p, yzgt'ssjo; remwisdv sill$dggl' .vgtzqv’Vod «Siaim nomijsdn inaersqsfl .52218 nolanuaro . ‘ . I (gagstflvs ed: 13 antsafaes to} eoxgbuflflqgflixf‘ . . a :3. 202 accessibility to their colleagues. Thus, the fourth conclu- sion is as follows. Conclusion 4. It appears likely, based on the re- sults of this research, that when the decision regarding adoption of an innovation can be made by individual faculty members, that the primary role of the department chairman is as a facilitator at the awareness stage, rather than as an influential at the persuasion stage, of the innovation- decision process. Conclusion 5. The only interpersonal communication or mass media communication source more frequently used by both opinion leaders and individuals with high network cen- trality,with respect to communication concerning teaching- related.topics, than by the average respondent, is The Accounting Review;:h1particular, the Education Research and Academic Notes Section of The Accounting Review. Thus, The AccountingReview is not only an important source of information regarding new teaching methods for the average respondent, as was cited previously; The Accounting Review is also the only mass media source of information used more frequently by both opinion leaders and individuals with high network centrality measures. One can only wonder why the American Accounting Association chose to reject the strong recommendation of one of its committees--that the association publish a journal devoted to research in account- ing education.5 It appears likely that such a publication .Lnoieulfi ” 5 ‘ 39¢“. . :53??? aid! 1.” ‘7 ; -H: Ted: .at~~ '- XkaIiaai . i 2:. i meal“- 7_, {T3181} tussle? 11:932A ;.msbsofi if .n‘ . 10 oaruoa _q. g3gfi v . -' ‘ , z” vxl'r' 93819?! boas; nob.» .~ . . g , ._ gaje 2i uoivg! ZESLD’,‘ h1_ . .. ., . J -A l 1: s . . , a ‘ ‘r . '{liuoupoti‘ 0'5ng ; ‘->_. . xcbaou (Inc as: gnu .rsrn‘.~w ;:i;urjqes {Ionssn All“? .1 r 3.11m 91 used: npixsiaoaen snirnuoaoA Haiti W...‘ 7' » - .’ y 5 .k . _ _-,l . ‘r_ u . .: - ‘3. a ? ~. ' _ __ ' v . _ .A a. ' ‘ . . . , ', “'1- ,' " ' ' ~ .1." .. 203 outlet would serve as a forum for both opinion leaders and individuals with high centrality index measures. In addi- tion, with the source credibility of both the American Accounting Association and the opinion leaders behind it, it seems very possible that such a publication would be viewed as important by the average accounting educator. It is this researcher's opinion that such a journal would have a good chance of establishing a reasonable level of prestige and reward for research pertaining to accounting education. The very lack of such an effort, and the "back-of-the-bus" location of the Education Research and Academic Notes section in The Accounting Review, by the organization rep- resenting ‘the teaching arm of the accounting profession, serves to reinforce the lack of prestige and potential re- ward for research efforts in this direction. Barring a change in policy by the American Accounting Association, one may only hope that the route the American Accounting Asso- ciation chose to follow--their Education Series collection-- achieves a higher frequency of use and perceived importance than the results of this research tend to indicate. Conclusion 6. Whereas, in general, Opinion leaders with respect to teaching-related matters neither use the avail- able interpersonal communication forces of information more than do their peers, nor consider them as more important; individuals with high centrality measures, with respect to communication concerning teaching-related topics, make more 7" O 1‘;4 1:3 vn:roi s an O -‘ .; ~“::1:u93 Hall 1“.“ '1‘ H w‘xw 5J1uoa 0‘3 ncr15i3088A [I ‘ o. tea» 3 dizzoq troy: ,. «d; {d :an:1-- ‘1 ‘13n31592L1 '9 uansfla ,. ~ ‘.‘ bTBWST 52‘ ‘bi {10V on? u . * notjsaol »;KT ~~ noi139¢ v.;*n9291 'H 21 .J’ EOWOR 5 b13w figfisdD uyhfi ”lno YBG 127$, n0135i3 ~jh L asveiiii 3 1 .Ifiiuarr! ofl: fllfl3 ' a. 7 E15552; ”*A‘iqt .i.' ' I' . . ’ s wohcuianoa st£i5¥$rbd3 sen Todiiun aratsnm 377:;a.»¢nid:a93 o: 339q¢§31,;fi ' .. '3¢'sh , Itiifiljoial io esoaol auxisaiuummoa Ianoq a§-" :W it “were... ,7, . , . 7.... .“ ‘1‘ , I ‘ ' " ‘ I. A l 204 frequent use of these sources-*particularly participating in informal discussions with other accounting faculty while at national and regional conventions, and contact with non- accounting faculty members--and consider them as more impor- tant, than do their colleagues. Thus, the individuals who are central to the communi- cation networks within their departments are also relatively more active than their colleagues in interpersonal channels while at conventions, and with respect to contact with non- accounting faculty. Those persons with high network central- ity are likely to be the individuals who first became aware of new teaching methods used by non-accounting faculty acquaintances, and are also the individuals who are in a position to disseminate this information, both within their own departments and to accounting faculty at other schools. The capability of serving these linking functions, in conjunction with the profile of individuals with high relative network centrality as being junior faculty members, suggests to this researcher the importance of attempting to direct the efforts of junior faculty toward accounting edu- cation topics and research. The recommendation made previously--the establish- ment of a journal of accounting education--would be a signif- icant step in this direction. In this researcher's opinion, providing incentives to graduate students at the disseruujon stage, and to junior faculty at the post-doctoral stage, 293100? 380d! 3.! an valuaal gab, 1n alkdl: jJnuoaat 913 {it won 10 $5933 0.3;?Oq um nifid! .elooflaa ;,Lnoa ni uuiSBIST 2319330! ad: 139115 alqu3 £01533 'm-rxouox adT isnxuot s 30 ith}, 4": b game 5 3d bIUOtI"n(3ijtju b3 2r'ijnuc;_g ,CWnu-z an: M .noiiasxib aid: at g. 205 would also seem particularly promising. This writer is ex- tremely pleased to note the recent announcement by the Touche Ross Foundation of a five-year, million dollar re* search program primarily for accounting education and multi- disciplinary research efforts.6 The availability of adequate research funding, in conjunction with a suitable publication outlet for the results-~that would serve to provide profes- sional recognition to the researcher and to disseminate research results to the profession-~would be very powerful incentives, hitherto not in existence, for doctoral students and junior faculty to direct their research efforts towards . . . 7 problems in accounting education. .!:;rmi1q mnrgORQ’ u 1.4 ugevsr vxsnilq .gr‘ibnu'? d3?- 4 951‘ to? 1917 .. Isudl’ . r 12139291 " 'u/f noon! Jw[ has Ido1q 206 Limitations Perhaps the most significant limitation of this research consists of the assumption of a linear model as representative of the underlying relationships between vari- ables. Each of the types of analysis presented in Chapter III--Pearson correlation, principal components factor analy- sis and multiple linear regression--are based on a linear model or function. As has been previously mentioned, there is evidence from prior diffusion research in other fields,8 which suggests the existence of a non-linear relationship between opinion leadership and other variables used in this dissertation. However, this writer is unaware of prior research that provides a basis for estimating the linearity, or lack thereof, of the relationship between the network centrality dependent variables operationalized in this re- search and the independent variable measures. The assumption of a linear model was made, and is considered appropriate in this research by this writer, for the following reasons. First, the present research is ex- ploratory in nature and the statistical techniques selected have been used simply to provide descriptive measures of linear relationships in the data; these statistical tech- niques have been used neither for formal hypothesis testing, nor for prediction purposes. Second, unless the two variables exhibit a perfect linear relationship, a curvilinear function can, potentially, always be found which will better fit the data. The ' L2 ._. nuS " A: ~_: 3 £133.41 J .37 n"ra 320m 6d! C. ff. :‘js ‘7 WM 30 8311.“: .‘ '~L, Vfi’ 70 OVI’UI; .,~ $17 30 donfl ~qu31um but '1“".1.‘.} 10 I.” t Jonabivo at 0 fine daidv ,. noowiad ‘7? 719221b '1139291 E1 .gsi 10 fillfhilfls3 '3! 1'1Ubiafl09 "“_. , ‘, A. . L al.,:lga, -_'g :: ._ -H‘ w; ‘V’.RLC133791 18.“!1 tgllieai aizadioqvfl {5:101 19: 19A;;nn beau need svsd =«fifii;i ; .asaoqruq noiaaiborq.iCj4 .5; I!“ all; news .1, V '4 . t.. ‘l... :‘ .\. In. "L “g“ “‘ ‘ 207 selection of suitable transformation functions for the independent variables in this research, or of a general transformation function for the opinion leadership dependent variable(s), is a difficult task and a worthy research pro- ject by itself. Even after "better-fitting" models have been identified, the question of whether the higher order models are more useful than the simple linear model remains to be answered. Finally, from a practical perspective, computer pro- grams for statistical techniques that assume linear models are by far the most widely used and available. Next, the statistical techniques employed in this research assume bivariate, or multivariate, normal distribu- tions. Thus, a second limitation of the present research is that if violations of these assumptions are present in the data, the statistical analyses may have yielded spurious results. Third, as has been mentioned many times previously in this research, the departments chosen for distribution of the data-gathering instruments were not a random sample from a defined population. Thus, the results presented in this research may be generalized, in the sense of statistical in- ference, only to the schools and individuals analyzed. Se- lected characteristics of the ten departments in which the data was gathered are presented in Chapter II, in order to assist the reader who wishes to infer the results of this research to a specific population of interest. $03 in. n " ‘ruinmroian513 oldflm “ J ‘13991 11d: of zoldnlzq‘ . ' . , ;‘-~ ag~ x01 noisonui 1.31 . ,-.:.,t Huatiiih s :1 .(31 lg.« 1‘ T’lL nuvfl .110133 ..~,._ 3‘7 .beiiijnoblu- new 818 a!“ 'w-n swans Dd .- ‘L’xseze'x “1:1 .mxoi! 1% 15d: .. .m‘ ,5385 :31u291 ‘ . a > {'75, 3'! 2i{(1 “1 'Vu i 5;;n: ‘ . -'-n3$’ -p?5b 3d3 :Efil: - TL: - , . .5 ‘ , . ‘ > ; ',.",L5'.-§‘-'q L—‘vfliasb ‘. .3 ' f .f+ 4. A . “1 ~5-11c116;< . . 2 ly'v-,_::;:3;: Oil VM dD'Xl’m _‘.93 -b'3S‘(L5fls alsubr ubxt‘r inn? : .c3t. '3? ad I 03 ‘(Iflo 9%} 9‘! taken at canaaaraqobr rte: “d1 10 aoiiai1eiafitld3 r f" 208 Final Note The current research represents, to the best of this writer's knowledge, a pioneering effort within the context of higher education in accounting. As such, it has not benefited from the previous efforts of a developed research tradition with a similar frame of reference; as a result, the possibilities for further research are corres- pondingly abundant. This research has focused solely on relative indi- vidual differences between individuals in accounting depart- ments at selected AACSB schools. No attempt has been made to assess dyadic, group or higher level metrics; in addition, many other types of networks could be defined. It is this researcher's opinion that the use of techniques such as net- work analysis, which retain the structure of the relation- ships between individuals, allows a more powerful and poten- tially fruitful analysis than weaker procedures applied on a grander scale. It is this writer's hope that the results presented here have provided a start, however tentative, toward the development of a research tradition or methodology capable of addressing problems that should be of concern to all accounting educators-~those within accounting education. 90$ gjpx_lsuk1 .fv« ,a. »1 fl31£3801 18011IQ_ ',~.~.c a .eabelwoai~§f “ai'noubo rod‘ld'it i ,1 “or! perilouod ' seq 9d: .11’ 'Lera Yfgntul "L Iaubiy ‘T‘ ‘r 23fl9' 4.9235 01 L was: I . . t ‘ . f v‘YBOBO'X ;. isms iXOV -. .; aqlda €“ VKLBil ‘75"? :ahnsrg B )0? f; . . . 1: uvsfl Stad ; 314.303.? -“'_",' g: _ .r , . i‘ , A .._.‘ t I T.” '115mq019V.b 1 [IL 01 fi‘dthQ E~ or L:;.,~ ;;“' am *icrg aniEBSTbhlf" 9' | . . ;.n01363ube gui7u00333 mid¢;w 340d7"i70383Ub9 ant 209 FOOTNOTES TO CHAPTER IV 1Bass, Cascio and O'Connor, "Expressions of Frequency and Amount". 2It might very well be, however, that use of a medium such as instructional television would be a collec- tive decision of all the faculty in a department. If this were the case, informal channels of influence would also be a factor. 3Rogers with Shoemaker, Communication of Innovations, p. 30. Their potential function as gatekeepers, or facili- tators,is the primary reason that department chairmen were consulted prior to the distribution of the survey instru- ments at each school. 4Rogers with Shoemaker, Communication of Innovations, p. 255. SSee Committee on Multi-Media Instruction in Accounting, "Report of the Committee," p. 134. See also the forward by Harold Langenderfer contained in Edwards, Accounting_Education, p. ix. 6Touche Ross 6 Co., "The Touche Ross Program to Support Accounting Education;" brochure distributed in fall, 1976. 7This writer personally believes that one without the other--funding without a publication source, or vice- versa-—would be a step, but only a step in the right direc- tion. Substantial dollar funding for education research has been available for years from organizations such as the Alfred P. Sloan Foundation, but has, to the best of this writer's knowledge, been used very little by accounting academicians. Just as important is the fact that the reward systems at most major institutions heavily stress publication records, even going so far as giving different point alloca- tions for publications in different "classes" of journals. With an article in the Education Research section of The Accounting Review as the most prestigious publication outlet available within major accounting journals for research in accounting education, it is not surprising that most doctor- al students opt for a dissertation topic which offers better possibilities for recognition. 8Rogers and Shoemaker, Communication of Innovations, p. 190. '21‘.! ..) '3. !'_'JTO“TM , w '0 hns 013253 .8. ' ."1nuoaA .'” Q Hg 779V Jd‘ifl ’1: ._. r..'391;»ai an don: u. ":0 nokalath. :;¢.r3n: .9259 9d? 'sLTn hsaluaaoa {4.2 ”n zine- V,jNUOJDA _. 'Y U‘Il'l 0} Al'tr In3JA ; _ r" '194qu2 ' _ g; .1151 ‘ . i.w» 7L :meJZXZ "517-1 »- ._' _ 4 _ 3! ”‘1". {[1039} LJ "I" ‘ i V. .JL- ‘CII 3:10.13 3i: 7 “"- . t . _ . ‘,,.J,jr; na 01:" Ieliuo no::;.ij.w. ,3 _ _ , ;, ; J, .5, Luiznuoaah . “I flax-EQLHH". 7:")— 41:.” .' IO! “' r” ,‘ _V -'r-', . ,V,:.:.V_."“.‘.AL.- '10330b Icom 35d? gnr.;1q¢u€ f‘; a: f; .waIEDUPfi gflljzu. 'lttlied 279310 dgifiw Wifigf n“. , ‘qo eJflsbugg .5 ~ , ;Jufl‘w;_ D S noixlugoss: ~03 aoltilki SELECTED BIBLIOGRAPHY SELECTED BIBLIOGRAPHY Sources Related to AccountinggEducation American Accounting Association Committee on Accounting Edu- cation and American Institute of Certified Public Accountants Computer Education Subcommittee. "Inclusion of EDP in an Undergraduate Auditing Curriculum: Some Possible Approaches." The Account- ing Review, Vol. XLIX (October, 1974), pp. 859-64} American Accounting Association Committee to Prepare A Revised Accounting Teachers' Guide. A Guide to Accounting Instruction: Concepts and—Practices. Zafed. Cincinnati, Ohio: SouthIWesternPublishing Co., 1968. Arens, Alvin A.; May, Robert G.; and Dominiak, Geraldine. "A Simulated Case for Audit Education." The AccountingReview, Vol. XLV (July, 1970), pp. 573-78. Askins, Billy E. "Determining the Effectiveness of Programmed Instruction-—A Training Course Example." The AccountingReview, Vol. XLV (January, 1970), pp. 159-63. Benjamin, James J., and Ricketts, Donald E. "A Profit Planning Project in the Management Accounting Course." The Accounting Review, Vol. XLVIII (October, 1973), pp. 791-97. Butts, Franklin Eugene, and Prickett, Gary L. "The Effect of Audio-Tutorial and Programmed Instruction. . " Laboratories on Achievement in Accounting PrinCiples. Unpublished Ed.D. dissertation, Colorado State University, 1969. Caldwell Jimm Carl. "An Inquiry Into Business Gaming as a Pedagogical Technique in Accounting Educat10n.‘ Unpublished Ph.D. dissertation, UniverSity of Alabama, 1970. Cloud, Charles Dou las. "An Experimental Study Comparing the Effectivengss of Programmed Instruction and the Conventional Method of Teaching First-Semester Principles of Accounting." Unpublished D.B.A. dissertation, Arizona State UniverSity, 1971. 210 ‘ , l.7.f;‘-9_x. 25313108 . gulinuoooA nl9= i1;-. has noiis) ' ,.nnxnuoaoA . cILLulDCI" - u.ii~u) I ‘.' LI} gang .h..<;¢ neDfTOIA 2 , q ‘I (Hi ‘_. _.l ”‘7""W V" H ' r , , / I' (:1I3‘TA . A; , j ,i ,83308 ' 3",1' ' '3’) . , ‘ .v . . .‘. :3.’.=o1.".l 555‘: 7+ -r." :% . = "1 . C ls :5.LUQRU ’IiffOV2flU . s.an inimea 2:3'1 at him. ,-..HH. : .ryg; JamiL ,Ilowblla. ”'“°¢3~9“55 flfliifi2031A iv'cmy-ndvuf I stgogsboq .53: *0 131379V1HU .001151rszzra .Q.dq hofieiiduqnu A #7; .0101 ,s-IJOIA '5'“ 4 ,‘J‘I'a‘y .‘ '4; 7"“! “A" ‘¢.I‘uw .1}, . ._.-,.Wf¢“;~ “g,;5° 3'N.f , 211 Cole, Frederick Miller. "A Study of Comprehension Levels of College Students Studying Elementary Account- ing Via Rate-Controlled Speech." Unpublished Ed.D. dissertation, University of Florida, 1971. Committee on Multi-Media Instruction in Accounting. "Report of the Committee on Multi-Media Instruction in Accounting." Supplement to Volume XLVII of The Accounting Review, 1972, pp. 110-62. Cushing, Barry E., and Smith, Charles H. "A New Emphasis for Introductory Accounting Instruction." Ihg Accounting Review, Vol. XLVII (July, 1972), pp. 599—601. Daily, Victoria Lee DeFore. "The Effect of Programmed Instruction in the Teaching of Principles of Accounting." Unpublished Ed.D. dissertation, Colorado State University, 1969. Dock, V. Thomas; Guy, Dan M.; and Williams, Doyle Z. "Integrating the Computer in the Classroom: An Approach in Auditing." The Accounting Review, Vol. XLIX (January, 1974]] pp. 149-53. Edwards, James Don, ed. Accounting Education: Problems and Prospects. Education Series Number 1. _N.p.: American Accounting Association, 1974. Flanagan, Stephen Michael. "The Effectiveness of Random Access Tapes in the Instruction of Elementary Accounting." Unpublished Ed.D. dissertation, University of Northern Colorado, 1970. Glein, Irvin N., and Wallace, John B., Jr. "Probabilifljcally Answered Examinations: A Field Test." The Accounting Review, Vol. XLIX (April, 197TTT pp. 363:66. Glover, Mildred Williams. ”An Experiment in the Use of Programmed Instruction in Elementary College Accounting." Unpublished Ed.D. dissertation, University of Georgia, 1970. Granof, Michael H. "Conference Telephone Calls: A Means to Bridge the Academic-~‘Real World' Gap." Th3 Accountin Review, Vol. XLVIII (July, 1973), pp. GlZ-lfi. Hong, Sunion Theodore. "An Empirical Study of the Effec- tiveness of Pregrammed Instruction and Computer- Assisted Instruction in Elementary Accounting." Unpublished Ph.D. dissertation, New York University, 1972. IIS -1'“" ‘o vf-uia A 10111! ,- 1.,2‘ ;- f.-1 3n1{iu ajuobuia 0:01, .1 ' ' n ,L liiloiinoD 533 . *xiravi-U .noiiszioackb .U} ‘93; .ihsM-t’luu I. 3i‘:...J v1! 10 JTOQ'I" , 1 ‘.;nijnU033A at " ' r:.'nu033A 0d? h-i_dl..-__~. .fi V7738 (in? to) . .inuoy3A "" t: ( ,Iii .xclid 1' ‘ , . ,‘1‘ .- a'_}}LM vs . gt .7 r' ,‘wkigo:q ~1 ~'* '. . azo’cg *: f.,3a-ro‘llrlU aneofl A : .Yls; anodqeieT agraisluou” .i issdJIM OAT .qnu ‘blioW :tmv;;3A 5N: ughiifl OJ “.(iifil ,quLj IIIvJX .LoV .wsivofl niiauoa r . ' .5001 .V .1300 kiuwhi fisgansid _n;913 ,1evch ”a .t-:.‘ 2’ 212 Humphrey, Joseph Lee. "An Inquiry Into Programmed Instruc- tion as A Pedagogical Technique in Accounting Education." Unpublished D.B.A. dissertation, Texas Tech University, 1971. Kinney, William R., Jr. "The Use of the Time-Shared Interactive Computer in Audit Education." The Accounting Review, Vol. XLIX (July, 1974), pp. 390-94. Li, David H. "Audit Aid: Generalized Computer-Audit Pro- gram as an Instructional Device." The Accountihg Review, Vol. XLV (October, 1970), pp. 774-78. McCosh, Andrew M. "The Case Method of Accounting Instruction and Microwave Television." The AccountingReview, Vol. XLVII (January, 1972), pp. 161-64. Markell, William, and Pemberton, Wilfred A. "Programmed Instruction in Elementary Accounting--Is It Successful?" The Accountin Review, Vol. XLVII (April, 1972), pp. 331-34. Onah, Julius Onvorah. "An Experimental Study Using the Audio-Visual Tutorial System to Teach Principles of Accounting to Community College Students." Unpublished Ph.D. dissertation, Michigan State University, 1971. Orefice, Dominick Salvatore. "An Experiment to Determine the Effectiveness of Programmed Instruction in Elementary Accounting." Unpublished Ed.D. disser- tation, Rutgers University, The State University of New Jersey, 1971. Sale, J. Timothy. "Using Computerized Budget Simulation Models As A Teaching Device." The Accountin Review, Vol. XLVII (October, 1972}, pp. 8 - 9. Smith, Jay M.; Taylor, Dale; and Western, Harold. "Experi- ment in Modularized Learning for Intermediate Accounting." The AccountinghReview, Vol. XLIX (April, 1974), pp. 385-90. Streuling, G. Fred, and Holstrum, Gary L. "Teaching Machines Versus Lectures in Accounting Education: An Experiment." The Accounting_Review, Vol. XLVII (October, 1972), pp. 806-10. Touche Ross 8 Co. "The Touche Ross Program to Support Accounting Education." Brochure distributed by the firm in fall, 1976. SIS .9, , - , I .' .F 7.11%” n1 m‘." eoJ'd iwli . ,w,g i )T 5ii3038b09 A h._ . [V w ., .‘Jcrldr:an ".no!’ ~ :_ I' a ta~cvinU doe? I‘ . l ' 4 \ -i ” .11 .a-. no? svisostt3li _.3 of v_ni3nuoaaA -¢H "EMA" .3 bf 15 as “:18 - is 4' $9,111.28. -' -.4~:br:A ..‘I'. am I; .10"! EIEIIVDY 1A 1’! f "1 i: EU (2 "at" . (“4111 1' . 4 , 9 b c M “954.15"! ‘. 1' 2' 'ffV'Il' ,.M ‘(13Y } I;"'- ' 1 I: -V n; Jnsm Alli . , I . ' " xiwuooaA Iii gain: 591‘” ..I V325.) .r:...r.?.;.-_r2'-’, bras 391'! :nngsauba zflljflUCJSi e1 zsruioeJ auazov zenida o‘OV uvelvefl garinnoaoA miT “.Jnemiro x8 IA ,.s usiiltl"' 11911:“ .9313910 .01- cos .qq .(SYQI .19 one)» dzoOaI. L .9138 .dstaa 213 Walgenbach, Paul H., and Frank, Werner G. "A Simulation Model for Applying Audit Sampling Techniques." The AccountinggReview, Vol. XLVI (July, 1971), pp. 583-83. Will, Milton Mike. "The Effect of Free Operant Learning on Achievement in the Principles of Accounting Course." Unpublished Ph.D. dissertation, University of North Dakota, 1970. Williams, Doyle Z. A Statistical Survey of Accounting Education 1967769. “New‘York: American Institute of Certified Public Accountants, 1969. Zieha, Eugene L. "Computer-Generated Accounting Assignmentsfl The Accounting_Review, Vol. XLIX (July, 1974), pp. 600-02. Sources Related to Communications Research Amend, Edwin H. "Liaison Communication Roles of Profession- als in a Research Dissemination Organization." Unpublished Ph.D. dissertation, Michigan State University, 1971. Carlson, Richard 0. Adoption of Educational Innovations. Eugene, Oregon: *The Center‘for the‘Advanced Study of Educational Administration, 1965. Evans, Richard 1. Resistance to Innovation In Higher Education. sen Francisco: Jossey-Bass Publishers, Inc., 1970. ; Smith, Ronald G.; and Colville, William K. The University Faculty and Educational Television?— HostilityLResistance andChange. Houston,_T3Xas: UzTiversity of Houston, 1562. Farace, Richard V. "Instructions for Design and Use of Network Analysis Instrument." Mimeographed copy of unpublished paper, Department of Communication, Michigan State University, 1974. , and Danowski, James A. "Analyzing Human Communica- tion Networks in Organizations: Applications to Management Problems." Mimeographed copy of paper presented at the International Communication Asso- ciation meeting, March, 1973. IuUmSVBidOA an un ".oztqu , 3;_ » [I djxoW 30 ‘ ulvflfl .th -.‘ ,1»:Lgoub8 -nf‘1o i i I 5,, , (“Iii .1119“ unnuIIBD ,zzxfi'la : , V -' '1 A/ sin ,~3d;:fl .easroi «J {(3 5'"- ‘ll \1 - . , . I .AIC.L70’..)I;'J. x. ‘ . ' "- :"I‘Jqfl" '4‘" 7 J 3"- »Aer ‘. . >. .Y’l .scidoim ~uainumm03 nun H yni' 03 anoilhiiiqqA TdnoijnsgfingG at PiYOWJOK flOiJ '9filq 3° YQOD bsdangoemiv “.zmnldoxfi sasmsgualfl ..,;;,:-IgaA aoiasolaualov IBflOiJOfl1SJflI 0d: 38 basasalifi .EIQK .dorsN .gniieo- notl_, 35 .. - - rnv; ‘ g.- , ,io -flnuh” .h aims; .zlsvonsu has i “31' ‘ . J" '- 'h 4‘ ’ I 214 , and Johnson, Jerome David. "Comparative Analysis of Human Communication Networks In Selected Formal Organizations." Mimeographed c0py of paper pre- sented at the International Communication Associa- tion meeting in New Orleans, April, 1974. ; Richards, William D.; Monge, Peter R.; and Jacobson, Eugene. "Analysis of Human Communication Networks In Large Social Systems.” Unpublished paper, Department of Communication, Michigan State Univer- sity, 1973. Guimaraes, Lytton L. "Communication Integration in Modern and Traditional Social Systems: A Comparative Analysis Across Twenty Communities of Minas Gerais, Brazil." Unpublished Ph.D. dissertation, Michigan State University, 1972. Havelock, Ronald G. A Guide to Innovation in Education. Ann Arbor, Michigan: Institute fer Social Research, 1970. . The Change Agent's Guide to Innovation in Education. Englewood Cliffs, New Jersey: Educational TeChnolo- gy Publications, 1973. , and Havelock, Mary G. Traininggfor Change Agents. Ann Arbor, Michigan: Institutegfor Social Research, 1973. MacDonald, Donald. "Communication Roles and Communication Content In a Bureaucratic Setting." Unpublished Ph.D. dissertation, Michigan State University, 1970. Monge, Peter R., and Lindsay, George H. "The Study of Communication Networks and Communication Structure in Large Organizations." Mimeographed copy of paper presented at the International Communication Association meeting in New Orleans, April, 1974. Richards, William D. Jr. "An Improved Conceptually-Based Method for Analysis of Communication Network Structure of Large Complex Organizations." Mimeo- graphed; East Lansing, Michigan: Department of Communication, Michigan State University, 1971. "Network Analysis in Large Complex Systems: Theoretical Basis." Mimeographed copy of paper presented at the International Communication Asso- ciation meeting in New Orleans, April, 1974. an} . 1' ,nilsel “.33 55v .pb1sflotl . 4?;nfi' .5noanl 4’ 3218-1 fl! , 'i;m:18q50 .Vg ,AWnISVBH LQ'LT .. V 3 ' 4 ».~i;W .abtsdbin A' \,;l‘., ‘ : 3”,; ' , ‘3;,~;,:-'x ,' Ton-33M -uome “.»«; ‘nglr .-' . .. _,' I‘vnusuwia ‘ lo lfsmiwnzqw .ung..vlh 3:; At} Yin! 2b$dq5‘3 4”“ «Hm-with «m? -;~;._..:.- ,ueijmimoa - 'A 3.8.3.12 rolqaoa ogre! ui ”12.1un) Atoviéfl" 1.1.1 )0 1‘0: b‘dqctsoouin ".2123: Isaifo '“4. . v.9 . _ 03 inactialtorai ad: in be _;4l 3' > '¢ 4 a : paw :IOJ . A 31?'-. *' ' 1.". (F - ' -~ 0 215 . ”Network Analysis in Large Complex Systems: Techniques and Methods‘-Tools." Mimeographed copy of paper presented at the International Communica- tion Association meeting in New Orleans, April, 1974. . "Network Analysis in Large Complex Systems: Metrics." Mimeographed copy of paper presented at the International Communication Association meeting in New Orleans, April, 1974. Rogers, E(verett) M. Diffusion of Innovation. New York: The Free Press, 1962. , with Shoemaker, F. Floyd. Communication of Innova- tions, A Cross-Cultural Approa'h. 2d ed) INEw York: ThewFree Press, 1971. , with Svenning, Lynne. Modernization Among Peasants: The Impact of Communication. New York: Holt, Rinehart andfiWinston, inc., 1969. Ross, Donald H. Administration for Adaptability: A Source Book Drawing Together the REsuIts ofiMore‘Than ISO* Studies Related to the Question of_Why and How SEhools Improve. New York: Metropolitan SCEOol Study Council, 1958. Schwartz, Donald F. "Liaison Communication Roles in a Formal Organization." Unpublished Ph.D. disserta- tion, Michigan State University, 1968. Shoemaker, F. Floyd. "System Variables and Educational Innovativeness in Thai Government Secondary Schools!‘ Unpublished Ph.D. dissertation, Michigan State University, 1971. Weiss, Robert Stuart. ”Processes of Organization." Unpublished Ph.D. dissertation, University of Michigan, 1954. Statistical and Other Sources American Association of Collegiate Schools of Business. FacultyPersonnel. Edited by Cyril C. Ling. 10th ed? St. Louis, Missouri: American Association of Collegiate Schools of Business, 1970. Bass, Bernard M.; Cascio, Wayne F.; and O'Connor, Edward J. "Magnitude Estimations of Expressions of Frequency and Amount." Journal of Applied Psychology, 1974, Vol. 59, No. 3, pp. 31§3320. J 31? 1Q;L‘ ui PizwlsaA |I:._‘l- gtg‘nJDM h” " t n' up :2 fsynsaatq 1 ~99in 1.0138130‘ r, - 2.2VI: nA 110w! ' .vcemiu ".eaxssdi’: «u LJEGTBIII 0‘8 2' ' ,.in:;”-'.‘I') W." a". ' '?:519v)fi 1‘371. adT J7 djiw 2 aafloi'j "1 "(ll {ITW 12':fi ' Hi; ‘0“ 22°” 4? ‘1; .':“.(5wd')2 ~ ,‘;-=»§Lmsod3 v 511' " 73 329" H ,zaiSW . r (1.;qu , :1. ;,.£!:‘H.{ ,‘ .aasniam’i ‘11) «13153:: £301.3nid .3 ix v "d ) .iarj 30 notzsiaoaaA afiqixsmfi. :12uoa21. ,ai .OYQI .econicui to aloodoa 'V 3i 3Iuoa§ 1‘. :2 .50 1‘92 eistgtllin ‘I’Wf. ‘4' u‘- ' ,-p‘a ‘ 3! 216 Cattell, Raymond B. "The Scree Test for the Number of Factors." Multivariate Behavioral Research, Vol. 1 (April, l§66), pp. 245-76. Commission on Instructional Technology. To Improve Learning: Volume 1. Edited by Sidney G. Tickton. 2 vols. New YorE: R. R. Bowker Company, 1970. Cooley, William W., and Lohnes, Paul R. Multivariate Data Analysis. New York: John Wiley 8 Sons, Inc., 1971. Edwards, James Don; Hermanson, Roger H., and Salmonson, R.IR Accounting: A Programmed Text. 2 vols. Homewood, Illinois: Richard D. Irwin, Inc., 1967. Garner, Paul. Some Reflection on Research_by Doctoral Candidates—in Accounting. _University, Alabama: Center ibr‘Business andiEconomic Research, Univer- sity of Alabama, 1973. Glass, Gene V., and Stanley, Julian C. Statistical Methods In Education and Psychology. Engiewood Cliffs, New Jersey: Prentice-Hail, Inc., 1970. Harman, Harry H. Modern Factor Analysis. 2d ed. Chicago: The University oi Chicago Press, 1967. Hasselback, James R. Accounting Facultng1974-7S. Gaines- ville, Florida: By the Author, 1974. Kerlinger, Fred N. Foundations of Behavioral Research. 2d ed. New York: Holt, Rihéhart and Winston, Inc., 1973. Libby, Robert. "Prediction Achievement and the Use of Simulated Decision Makers in Information Bvaluation!‘ Unpublished Ph.D. dissertation, University of Illinois, 1974. Neter, John and Wasserman, William. Applied Linear Statistical Models: Regression, Anafisis of Variance and Experimentai Designs. H5mewood7_lllinois: Richard D.Irwin, Inc., 1974. Nie, Norman H.; Hull, Hadlai; Jenkins, Jean 0.; Steinbrenner, Karin; and Bent, Dale H. SPSS: Statistical Package for the Social Sciences. Zd‘edl ‘New Ybrk: McCraw- Hill Book Company, 1975. Oppenheim, A. N. Questionnaire Design and Attitude Measure- ment. New York: Basic Books, Inc., 1966. ol: 319i-.~I,é odT" .‘r; 13:1. [J.‘L-CFSVLT'w "o ":1- ‘ .1” - Law. ”1111‘; . m; . iunnttourlenx no. ‘TEiOYA ‘ff‘? (VJ)... I w m.1111u . "" I“ . 1-‘3SI‘M I“. ‘1 z-amsL .abij 47‘111(‘:1J,'. -"'.){|1-11T 2.. H .190103' 38313 nastH :Jiaazsfl 'I .pflfl1113x l _ :'Ur‘>i ,‘(ddiJ quu ‘ » 2 . “gall, .1539“ .JfUJrn! .22; ' 2. ,. _. . .7 .232 ". '1 ' , 315 ' Flaffl l w ‘ r .7; ruin? t3md0193? gu) mag“ H-chhifl. ;,.j[[.g,l_! ghgsii ;,H 113.110“ _ a). ‘ JA§ - ixainsi? 331%} .n also ,:503 7.x ‘ y - “ .«e'znaeiuég~ {31302 ad: - 2‘51 ,Ynsqnoj o ‘_ N,A."‘ 217 Rummel, R. J. Applied Factor Analysis. Evanston, Illinois: Northwestern University Press, 1970. Tatsuoka, Maurice M. Multivariate Analysis: Techniques for Educational and Psychological Research. New YorR: John Wiley 5 Sons, Inc., 1971. . Validation Studies: The Use of Multiple Regression Eqpations. Selected Topics iniAdvanced Statistics: An Elementary Approach, Number 5. Champaign, Illinois: The Institute for Personality and Ability Testing, 1969. n a .{x . A ~ . ~_!:,:1Aw101‘)6=[ 5011*”. . "‘JF Px“1?8w133 .l .77.— .- ., rI'w' m! "#592 w . T HES .2 'nionIIII gfliIBBT APPENDIX 218 APPENDIX DATA-GATHERING INSTRUMENTS Initial letter to department chairmen Cover letter to individual faculty members Communication questionnaire Personal contact listing ME 219 MICHIGAN STATE UNIVERSITY GRADUATE SCHOOL or m aommnou w'r umsmc - ”uncut: . “I. WARM” 0! WIN O "NWAL ADMINISTRATION May 6, 1975 Professor Head, Department of Accounting College of Business Administration Dear : I am writing to solicit your cooperation on behalf of Vince McCormack, Department of Accounting and Management Information Systems, Pennsylvania State University, who is completing a doctoral degree in accounting at Michigan State. Vince is conducting a study concerning selected aspects of the communica- tion patterns of accounting faculty members, and is seeking the participation of your faculty in his study. In formulating the research design, Vince has consulted extensively with faculty members from Communications Departments, both at the University of Michigan and Michigan State. My colleagues and I believe that Vince has come upon a novel approach to investigating an issue of real concern to accounting educators. The major goal of the study is to facilitate the transmission of informa- tion concerning new teaching technologies to accounting faculties. One result of the analysis will be a "mapping" of the communication network in your department. The method 6f analysis used to construct such a mapping requires a 1001 sample of the faculty in your department and virtually a 1002 response rate. It is hoped that your approval, in the form of a request to your faculty to participate, would help to ensure this degree of c00peration. In addition, Vince has already contacted of your staff, who has agreed to handle the distribution of the data-gathering instruments. The average time required to complete all materials, based on the results of the pretest anaysis, is half an hour per respondent. Distribution of the questionnaires would take place in approximately ten days; the completed forms would be mailed directly to Vince at Penn State. I can assure you that no one other than the researcher will see any of the completed questionnaires, and that no individual will be identified by name with any of his or her responses. I can futher assure you that no individual department will be identified by name with the collective responses of its faculty. 1' managed .u ”new In sgslm : no! v r u I 4» nmmssfl l.'.‘aYn'hflI ...:'l "e; no): T1101 In I -' ‘. :‘JuufiOD " :n (Hod , U 5. h 0 CL Inn '4 '1? I‘ K 4 [101) » mix is H ‘1'“ 4:191) - .. . mi 5 . ' "IIY . .2 ‘1111'103 , ,: mtg. nI wen". and :1va i -- ‘ "Wk 9!" ‘Vl’v'q 9:11 . ~;'1,,'.n:~~;.1)au'p ..,: "~03: mi MUM! .Wasae we; I 1 ‘ . pup Lain-Mm: 043 V “‘3‘ ‘ , -' - ' .- .W1 in van {I}!!! I“. ‘9'” 5‘ ‘4‘ "‘1 7"" "' ’ -- ' ' - . . I u.‘ {lug “.2er .mqnb Imubflw .1511“)! I21 h 220 May 6, 1975 Page 2 Your cooperation would be very much appreciated, and Vince would be happy to supply your faculty with an abstract of the results of the study. In order to answer any questions you might have concerning the study and to expedite getting the project underway at your school, either Vince or I will be calling you in a few days. Thank you. Sincerely, I ner M. 3 es Professor n Chairman CHJ/lmr 03g '11. "‘12.: .',:3‘ ‘1': ‘I‘Iir‘ 221 MICHIGAN STATE UNIVERSITY cannons. sum or mum ADHINIS'I'IA'HON usr ransom - men-saw - um uraammr or mm a mamas. aowmnsramvou May 13. 197 5 Professor Department of Accounting College of Business Administration Dear : I am writing to solicit your cooperation on behalf of Vince McCormack, Depart- ment of Accounting and Management Information Systems, Pennsylvania State University, who is completing a doctoral degree in accounting at Michigan State. Vince is conducting a study concerning selected aspects of the communication patterns of accounting faculty members, and is seeking your participation in his study. In formulating the research design, Vince has consulted extensively with faculty members from Communications Departments, both at the University of Michigan and Michigan State. My colleagues and I believe that Vince has come upon a novel approach to investigating an issue of real concern to accounting educators. The major goal of the study is to facilitate the transmission of information 'concerning new teaching technologies to accounting faculties. One result of the analysis will be a "mapping" of the communication network in your department. The method of analysis used to construct such a mapping requires a 100! sample of the faculty in your department and virtually a 1001 response rate. Your response is essential to the completion of this research. The average time required to complete all materials, based on the results of the pretest analysis, is-half an hour. Please mail your completed forms directly to Vince at Penn State using the envelope provided. I can assure you that no one other than the researcher will see any of the completed questionnaires, and that no individual will be identified by name with any of his or her responses. I can further assure you that no individual department will be identified by name with the collective responses of its faculty. Your cooperation will be very much appreciated, and Vince would be happy to send you an abstract of the results of the study. Thank you. Sincerely, \K Br ’/ jigtdddr‘hk 'nes ofessor sf dChairman GMJ/nm Enclosures In..g,a;.‘A 3° . a a up)“ '0 qr: ’1 'V 11:! I .J..ng;,‘mr 1° J ‘1 "flint-3:. ,3 “I 535:7 .'\ ?\i nnij:N ”(Ma "r nan-5.} .TL‘;A,LJ’:4 It“. ..,. - r1 [1:00qu1 “4‘7""! ed: "New. ., GUY 4'12““3 N! not 5.. 222 ACCOUNTING FACULTY MEMBERS COMMUNICATION QUESTIONNAIRE (XIIIAHCMJION STUDY INSTRUCTIONS. Many of the questions In this part of the questionnaire can be answered with a check In front of the appropriate answer category. Throughout this questionnaire, guidelines are given In capital letters to summarize the content of each section. When questions can be skipped, the number of the next question to be answered is given. I.O BIOGRAPHICAL INFORMATION. l.i As stated In the cover letter, no one other than the researcher will see any of the completed questionnaires, and no individual will be Identified by name with any of his or her responses. Further, no department will be Identified by name with the responses of any or all of Its faculty. I do ask for your name because I am charting the communication "map” of your department; however, all names will baiinmedlately transferred Into code numbers upon receipt of your completed Instruments, and the original questionnaires will be destroyed. Your name: I.2 What Is the highest academic degree you have received? Bachelor's Master's Doctorate l.2.l in what discilene was it awarded? i.) Have you received any type of professional certification? Yes . No . (IF NO: Please continue with question i.4) iF YES: l.3.I Which type(sl have you received? C.F.A. C.M.A. C.P.A. Other (please specify): i.4 What is your present academic rank? Professor Associate professor Assistant professor instructor or lecturer l.4.l Are you tenured In this rank? Yes . SSS *jrh‘t .j “ ‘ ‘(i‘iC-iTCIIUO 3"? ‘0 fl , _ -. r . .5 ‘5‘ ’u the-ii ml in“. in i, v. ' .~". -' seniiahluc .C‘ no‘wx. ‘s’ih molt”. M . new”: (i rem“ ”. «if ni lieflfl ‘ Q ~r-.:.I«'2mo: erit to W A wiw siren V‘ f ‘ w" .o boi‘rltMI ‘ ~ to: ice Ob I . n '2' ‘I no ‘woy . marsh ohm ‘«~.- i"(11?m° ' l " 'Uiv' 3i tad! OJ. I . .. v ‘1:- 'Il.-' _ v 4 ,_ I. .w .3“. me“. #53“. “- 3 '2.) 1_:*,-:,-.t;ni . obi . :9‘1’ Shun eitt ni benunet MN I 223 l.5 Approximately how many total years have you been teaching? less than I year i year, but less than 2 2 years, but less than 5 5 years, but less than IO l0 years, but less than l5 l5 years, but less than 20 20 years or more l.6 Have you taught at more than one institution within the last ten academic years? Yes No . (iF N0: Please continue with question i. 7) IF YES: i.3.| Please list the Institutions at which you have taught, within the last ten academic years, prior to latest employment at your present school. Name of Institution Academic Year(s) Employed I.7 Within the last five years, have you served as faculty advisor or coordinator for any student committees, clubs or fraternities; honorsNo or Internship programs; or other major student activities? Yes (IF NO: Please continue with question i. 8) IF YES: I. 7. I Please scan the list below and check those that are applicable. Accounting Club Beta Alpha Psi Beta Gamma Sigma Honors program InternShip program student committees student consulting services (9.9., tax service) Other (please specify): I.8 Within the last five years, have you served on any professional committees at the national or state level (AAA, AICPA, NAA, etc.) whose charge was concerned with accounting education? Yes No . (IF NO: Please continue with question 2.0) iF YES: i. 8. I Please II st the committee(s), its (their) professional affiliation and level, and the year(s) In which you served. Committee Affiliation and Level YearIs) Lacey latot vnsm "I l merit alol iud .waev I .ml no ,3110‘1‘3 :ud ,r1sev C ; .cnaev 0| '_'.7 '2' ‘.-i'(' 5' I ‘ A 'I, I" . a uni —‘£z€016 TV 1|) ., or" )‘5‘.’ .ilflTIH 8o. .pwoiten ant to M. In‘ 5091971103 iW fill" .' Tim ~.--,1(-2|-;,-.r)etl . new as W 2.0 2.I 2.2 224 THE NEXT FEW QUESTIONS ARE CONCERNED WITH SPECIFIC iNSTRUCTIONAL METHODOLOGY YOU MAY, 0R MAY NOT, HAVE FOUND WORTHWHILE TO USE lN COURSES YOU HAVE TAUGHT. The method of answering each question is the same. You are asked to determine: a. if you have used the Item within the last five academic years, b. If so, In which academic year or years you used It, and If so, was the Item prepared commercially (C), prepared non- commercially by a person or persons other than yourself (0), or prepared by yourself (S). For those Items you have used, If you chose to use any, enter the appropriate preparation code or codes in the year column or columns corresponding to your use of each Item. For example, If you previewed "Deep Throat" in one of your classes two years ago, you would answer: C. l972—73 Method C Motion pictures Have you used pgogrammed Instruction or modular course content In any courses you have taught within the last five academic years? Yes . No . IF NO: Please continue with question 2.2) iF YES: 2.I.i Please examine the following list and ask yourself: first, have you used It; second, in which years did you use it; and third, was It prepared commercially (C), non-commercially by other persons (0), or did you prepare It yourself (S). For each time you have used an Item, enter the appropriate preparation code in the year column corresponding to that use. Prior to Current & I970-7l i970-7I I97i-72 l972-73 I973-74 Method Programmed Instruction written material (e.g. text) teaching machine computer-assisted Modules Have you used a viewgraph, slide transparencies or filmstrips In any course No you have taught within the last five academic years? Yes . (IF NO: Please continue with question 2.3) IF YES: 2.2.i Please examine the following list and ask yourself: first, have you used it; second, In which years did you use It; and third, was It prepared commercially (C), non-commercially by other persons (0), or did you prepare It yourself (S). For each time you have used an Item, enter the appropriate preparation code in the year column corresponding to that use. Prior 'I'O Currenf & I970-7i l970-7l l97l-72 I972-73 i973-74 Method Viewgraph .._..._.__ _ _ Individual transparencies __ continuous roll Slides and filmstrips without taped sound synchronization with taped sound synchronization ‘35 a 1.1-. -‘|rl ‘ay‘jm.fl IRA MI " . i ,q' p" "rmj MI. I“ i; W at." m: '- "DOG 001* .7, ‘4 l"5 N" . nu',‘ 'iQil‘IU whatnfl‘m ,‘.‘fi"' “0, -. '1 al 5 0M 7 ' 3 "fiidl :15 ... ,.r‘.'!0.‘“flm H I.“ , a r ~~,(99 103 . . ..,., 1.35”! I)!” III . o" 10”“ :‘ V ' 7 A ‘ 7—‘ <17 '."J- ,.L-c-‘x‘vl “£31 7D eelsnesecensv. lap-S iv 'iw’ l‘i. IUIII ilm eucunl tncc meg-m mam“: " ' mu. 2115 2. 3 Have you used television or motion pictures in any course you have taught within the IasT five academic years? Yes___ . No____ . (lF N0: Please continue with question 2. 4) IF YES: 2.3. I Please_ examine the following list and ask yourself: first, have you used it; second, In which years did you use It; and third, was It prepared commercially (C), non-comerclally by other persons (0), or did you prepare lt yourself (S). For each time you have used an Item, enter the appropriate preparation code In the year column corresponding to that use. Prior to Current & I970-7l l970-7l l97l-72 I972-73 i973—74 Method Television live lectures, with feedback live lectures, without feedback pro-recorded audio-visual tapes Motion pictures with sound track without sound track 2.4 Have you used simulation projects In any course you have taught within the 3.0 30' last five academic yearSi' Yes No . (IF NO: Please continue with question 3.0) IF YES. 2. 4.i Please examine the following list and ask yourself: first, have you used It; second, In which years did you use It; and third, was it prepared commercially (C), non-commercially by other persons (0), or did you prepare it yourself (S). For each time you have used an item, enter the appropriate preparation code In the year column corresponding to that use. Prior to . Current A l970-7l l970-7l l97i-72 l972-73 i973—74 Method Simulation business games financial statement statistical sampling systems design budgeting and/or control behavioral THE NEXT TWO QUESTIONS DEAL WITH YOUR USE OF COMPUTER FACILITIES lN TEACHING, ACADEMIC RESEARCH AND RELATED ACTIVITIES. Have you used computer facilities In courses you have taught, academic research or related activities within the last five academic years? Yes (lF NO: Please continue with question 4. 0) IF YES: 3. I.I In which actlv Iw or activities have you used these facilities? Courses taught Research Other (please specify): . .......— v...— V.‘ avOH _ ,.i teal 7 : :‘I';‘ M’Il ‘I “ '. Bill ~: en :‘ti ‘I-f?1 DIMEGAQA crane-1M . been I r I' Hi ,‘C";“IOJ ’ wrlfiq i' C‘VI h" 'ni 7'7: masu uOV 9"" .xzisx boifiIIW ‘0 1"»;- :¢;;_.Iq If?) ‘I. .’m' axnd eeltivltai 19 up 1’ 2:3».- no.) (13156389 :(vtiaeca eaeelq) wodto i.l 3.2 ‘00 4.I 4.2 4.3 1E26 Did you write or personally debug any of the programs you used In these activities? Yes . No . (IF NO: Please continue with question 4.0) IF YES: 3.2.I Approximater how frequently did you write or personally debug the programs you used in connection with these activities? always often ' sometimes seldom THE LAST FEW QUESTIONS IN THIS PART OF THE QUESTIONNAIRE ARE CONCERNED WITH THE SOURCES THAT ARE IMPORTANT TO YOU FOR BECOMING AWARE OF NEW IDEAS, AND PROVIDING INFORMATION NECESSARY FOR EVALUATING NEW IDEAS AND METHODS, IN ACCOUNTING EDUCATION. Do you discuss ways to Improve the learning experience of your students with any full-time, permanenITaccounthg faculty members In your department? Yes No . (IF NO: Please continue with question 4. 2) IF YES: 4. I. I PIease IIs T the names of the three Individuals you seek out most often for information and/or advice. Do you discuss new teaching_methods and materials In accounting education (e.g., programmed textbook, teaching by televIsIon, preparing transparencies) with any full-time, permanent accounting faculty members In your department? Yes . No . (IF NO: Please continue with question 4. 3) IF YES: 42 I Please IIst tthe names of the three Individuals you seek out most often for Information and/or advice. Which of the following types of Interpersonal contact are sources of information for you with respect to new teaching methods and materials that could be, or are being, applied In accounting education? Please assign one of the following frequency codes and one of the following Importance codes 125 each Item listed. Frequency Codes Importance Codes I I always engage in I I extremely important 2 I very often engage In 2 I quite Important 3 I engage In fairly many times 3 I moderately Important 4 I occasionally engage In 4 I somewhat Important 5 I never engage In 5 I not Important Frequency Iggortance ACTIYIEX when attending national conventions/conferences presentations on education-related topics Informal discussions with other faculty when attending regional conventions/conferences presentations on education-related topics Informal discussions with other faculty H H “fin-a... 38$ ‘:',i . III';!'I\3" W MWitnmnoo lamina" gnitsnet a menu It m no amintneenq “Iv dismantle Inn—Iain! m 1;: ncivml an ' .I‘IOITAZIJ‘B alum 3‘! £81) dIb M“ I' wir—llut m 1101 I " l- V‘N'IUKIIE ‘0 '01 'IJf’clflI 10" an” In (If: ‘1’: am to ISM“ {J I .'Iv may 10‘ .m'wd . raupo‘fi 'I‘I;IT'V. "01:: zvane “ I . , unto now I f ' ‘ ni egsgne 'l ”Almanacs” ' I W 1"” ' ' HI 32.623211 w ' 227 4.4 Which of the following types of interpersonal contact are sources of Information for you with respect to new teaching methods and materials that could be, or are. being, applied In accounting education? Please assign one of the following frequency codes and one of the following Importance codesngg'each Item listed. Frequency Codes Importance Codes I I always engage In I I extrme inportant 2 I very often engage In 2 I quite Important 3 I engage In fairly many times 3 I moderately Important 4 I occasionally engage In 4 I somewhat important 5 I never engage In 5 I not Important Frequency Importance Activity discussions with publisher representatives discussions with faculty from your Institution with accounting colleagues with faculty from non-accounting business fields with faculty from non-business fields discussions with faculty from other institutions with accounting colleagues with faculty from non-accounting business fields with faculty from non-business fields Which of the following publications are sources of information for you with respect to new teaching methods and materials that could be, or are being, applied in accounting education? Please assign one of the following frequency codes and one of the following Importance codes 12£_each item listed below. _§:equengy Codes Importance Codes always read or scan I I extremely lmortant very often read or scan 2 I quite Important I moderately Important somewhat Important not important have no knowledge of this source read or scan fairly many times occasionally read or scan never read or scan have no knowledge of this source ue Importance Publication Audiovisual Instruction Book Review section, The Accounting Review Collegiate News and Views Dissertation Abstracts Education and Professional Training, Journal of Accountancy Educational Product Report Education Recaps Education Research and Academic Notes, The Accounting Review Research Reporter . Supplement to the Accounting Review: Committee Reports - Other (please specify): - . GUIbUN-I I I I I I I OIU'I#U I I I IIIIHIHHE HIIHHH TSS I .6, _-'- - '.. ;- q1rtn? to zeqvt Qfl'lfl‘jfl. >. ‘ ~ 3-.. :1-0 "1:52-35? '0" ‘ ‘ '3‘. «373.! ‘f ‘. ”KP-SUE“) WIT-“Dom E... I ‘ ~54 : A ed? to one m “ I l - 5., “I ii ‘4' j ‘ ‘V :,‘1‘ I A 1‘ 'e‘i \ . : “Ifii‘ Z“ ‘I ‘. al-5‘ . - A ‘ '3:;‘ LIT-col. "31::me 2:307 .921: 1"»; *v . some .' \grr.’ v'w‘s“ Fr: , t":dem:s«1-. ' --1.' - v’v- A .Ig. ~ :" »1~_Ir .noo‘) .wlw "'"fls- {:‘ .35: r I: J. 5d tr'e EH .1031” T5" “flit ,‘ : I II. i «tram 790 Car. { ‘U S = I you a mum fzeoi 1’! I C ' it. In“ . .’ Mm 160 ezno foods K S‘. ,Ieew 1eq emit E 10 S I C v “t , ‘ . met no eono fund: I l xm 19c; eono too“ '3 Frequency Codes 6 . at least once a day 5 t 2 or 3 tlmas per week 4 0 about once par week 3 I 2 or 3 tlnas per month 2 0 about one. per aonth 230 Inltlatlon Codes 3 = I usually lnltlata contact 2 8 We both lnltlata. about equally l ' He or she u5ually Ybur Name l 0 about one: par tern lnltlatos (XlIlNICATION TOPIC AREAS PROFESSIONAL TEACHlNG TEACHING TEAGHNG NM OF INDIVIDUAL CONNICATICN PRODUCT“)! llNOVATlCN MAINTENANCE Fra— lnltl- quency atlon Fra- lnltl- quency atlon Fra— lnltl- quency atlon Fra- lnltl- quency atlon +- F—T— 08 S 4:; ...:,if_s_i-Hnl ,. . ‘, .‘-r~'~:L““{ Nb.” ,u‘win “an $01 -I‘! «an. at -v . «mam 10¢ cult! ) "‘11111111]thMWS