INFORMATION TO USERS

This was produced from a copy of a document sent to us for microfilming. While the most advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependent upon the quality of the material submitted.

The following explanation of techniques is provided to help you understand markings or notations which may appear on this reproduction.

1.The sign or "target" for pages apparently lacking from the document photographed is "Missing Page{s)”. If it was possible to obtain the missing page(s) or section, they are spliced into the film along with adjacent pages. This may have necessitated cutting through an image and duplicating adjacent pages to assure you of complete continuity.

2. When an image on the film is obliterated with a round black mark it is an indication that the film inspector noticed either blurred copy because of movement during exposure, or duplicate copy. Unless we meant to delete copyrighted materials that should not have been filmed, you will find a good image of the page in the adjacent frame. If copyrighted materials were deleted you will find a target note listing the pages in the adjacent frame.

3. When a map, drawing or chart, etc., is part of the material being photo­ graphed the photographer has followed a definite method in "sectioning" the material. It is customary to begin filming at the upper left hand corner of a large sheet and to continue from left to right in equal sections with small overlaps. If necessary, sectioning is continued again-beginning below the first row and continuing on until complete.

4. For any illustrations that cannot be reproduced satisfactorily by xerography, photographic prints can be purchased at additional cost and tipped into your xerographic copy. Requests can be made to our Dissertations Customer Services Department.

5. Some pages in any document may have indistinct print. In all cases we have filmed the best available copy.

University Microfilms International 300 N. ZEEB RD.. ANN ARBOR, Ml 48106 8128966

Bo l t o n , W il l ia m T h e o d o r e , Jr .

THE PERCEPTION AND POTENTIAL ADOPTION OF CHANNEL 2000: IMPLICATIONS FOR DIFFUSION THEORY AND VIDEOTEX TECHNOLOGY

The Ohio Slate University PH.D. 1981

University Microfilms International 300 N. Zeeb Road, Ann Arbor, M I 48106

Copyright 1981 by Bolton, William Theodore, Jr. All Rights Reserved PLEASE NOTE:

In all cases this material has been filmed in the best possible way from the available copy. Problems encountered with this document have been identified here with a check mark V .

1. Glossy photographs or pages______

2. Colored illustrations, paper or print_____

3. Photographs with dark background_____

4. Illustrations are poor copy ______

5. Pages with black marks, not original copy ______

6. Print shows through as there is text on both sides of page______

7. Indistinct, broken or small print on several pages ^

8. Print exceeds margin requirements _____

9. Tightly bound copy with print lost in spine ______

10. Computer printout pages with indistinct print

11. Page(s)______lacking when material received, and not available from school or author.

12. Page(s)______seem to be missing in numbering only as text follows.

13. Two pages numbered______. Text follows.

14. Curling and wrinkled pages______

15. Other______

University Microfilms International THE PERCEPTION AND POTENTIAL ADOPTION OF CHANNEL 2000:

IMPLICATIONS FOR DIFFUSION THEORY AND VIDEOTEX TECHNOLOGY

DISSERTATION

Presented in Partial Fulfillment of the Requirements for

the Degree Doctor of Philosophy in the Graduate

School of The Ohio State University

By

William Theodore Bolton, B.A., M.A.

*****

The Ohio State University

1981

Reading Committee Approved By:

Joseph Foley Thomas McCain W. Wayne Talarzyk Advisor 0 COPYRIGHT

WILLIAM THEODORE BOLTON

1981 THE PERCEPTION AND POTENTIAL ADOPTION OF CHANNEL 2000:

IMPLICATIONS FOR DIFFUSION THEORY AND VIDEOTEX TECHNOLOGY

By

William Theodore Bolton, Ph.D.

The Ohio State University, 1981

Professor Thomas McCain, Advisor

The dissemination of computerized information into the home, via interactive home information systems, has been characterized as a revolutionary technological innovation which will affect the way people work, and how they conduct their daily lives. This research examined one form of this technology during a three month field test of a proto­ type viewdata system called Channel 2000.

Channel 2000 as an electronic informational device, delivered library and banking services, in addition to various types of community information, to 71 randomly selected Columbus, Ohio households. Users had free and unlimited access to these electronic services throughout the field test.

Diffusion theory served as the theoretical model which guided the questionnaire development and data interpretation procedures used in this research. According to diffusion theory, the perceived attributes of an innovation can to a certain extent, determine whether an innova­ tion such as Channel 2000, is either accepted or rejected. In an effort to assess the degree to which these perceptions, and other life style, personality, and socio-demographic measures could discriminate between iii potential adopters and non-adopters of Channel 2000, a series of three stepwise two-group discriminant analyses were completed.

.. Three significant discriminant functions were derived using each of these measures before, during, and after the test. The innovation perceptions consistently demonstrated the most discriminatory strength among all other measures. The perceptions themselves also indicated that although Channel 2000 was perceived as being not complex, users felt that it offered them few advantages, and that it was not very compatible with their current experiences, values, or needs.

The results of this study suggested that innovation perceptions may be a very useful discriminatory tool, as well as a helpful device in the development of future home information systems. Other recommenda­ tions for future research, and the implications for diffusion theory and videotex technology which resulted from this research, are presented within the discussion of the findings. All questionnaire items and frequency tabulations are also included as Appendices. ACKNOWLEDGMENTS

The Channel 2000 project represented the combined efforts of many individuals. The author is deeply indebted to the following people for their dedication, assistance, and encouragement throughout this research effort: Tom McCain, Tom Harnish, Wayne Talarzyk, Joseph

Foley, Neal Kaske, Dave Penniman, Mark Bendig, Dennis Schmidt, Mark

Ackerman, David Hsiao, D. C. Palmer, Phil Coit, Rich Skopin, Ric Limes,

Jim Ginter, Clark Leavitt, Chris O ’Dea, Hugh Clark, Decision Research

Corporation, and Jaimee Bolton.

v VITA

November 6, 1951 ...... Born - Syracuse, New York

1974 ...... B.A., Psychology, Ohio Wesleyan Univer­ sity, Delaware, Ohio

1975-197-6 ...... Account Executive, WDVR, Philadelphia, Pennsylvania

1977 ...... M.A., Communications, Temple University, Philadelphia, Pennsylvania

1977-1978 ...... Producer/Director, WBNG-TV, Binghamton, New York

1978-1979 ...... Teaching Associate, Department of Commun­ ication, The Ohio State University, Columbus, Ohio

1979-1980 ...... Research Assistant, OCLC, Inc., Columbus, Ohio

PUBLICATIONS

"A Lesson in Interactive Television Programming: The Home Book Club on Qube." Journal of Library Automation, 14:2, 1981.

"Media Use and the Life Span: Notes on Theory and Method." American Behavioral Scientist, 23:1, September/October 1979. Co-authored with John Dimmick and Thomas McCain.

RESEARCH AREAS

Major Areas: Audience Analysis, Media Technology Assessment

Minor Areas: Consumer Behavior, Communication Theory

PROFESSIONAL ORGANIZATIONS

American Association of Public Opinion Research Speech Communication Association American Marketing Association Phi Kappa Phi Honor Society

vi TABLE OF CONTENTS

Page

ABSTRACT ...... ill

ACKNOWLEDGMENTS ...... v

VITA ...... vl

LIST OF T A B L E S ...... xi

LIST OF FIGURES ...... xlii

CHAPTER

I INTRODUCTION ...... 1

The P r o b l e m ...... 4

II A REVIEW OF THE LITERATURE ...... 10

Defining V i d e o t e x ...... 10 Teletext and Viewdata ...... 16 Projections for Videotex Systems ...... 18 Diffusion of Innovations: A Theory of Social Change ...... 22 A Microanalytlc Approach to the Study of Social Change ...... 25 The Diffusion Literature ...... 26 The Attributes of the Adopting U n i t ...... 27

Sociodemographics ...... 29 Life S t y l e ...... 30 Consumer Creativity ...... 34 Interpersonal Communication . .•...... 37 Media Use ...... 39 Product Use ...... 41

The Attributes of the Innovation ...... 43

Innovation Attribute Categories ...... 44 The Perceptions of an I n n o v a t i o n ...... 47 Perceived R i s k ...... 53

A Brief Summary ...... 54 Research Questions ...... 55

vii TABLE OF CONTENTS (Continued)

CHAPTER Page

III THE CHANNEL 2000 SYSTEM ...... 57

Channel 2000 Hardware ...... 57

The Channel 2000 K e y p a d ...... 61

Channel 2000 Services ...... 63

Library Catalog ...... 63 Video Encyclopedia ...... 63 Public Information ...... 65 Community Calendar ...... 65 Math That Counts ...... 65 Early Reader ...... 65 Home Banking ...... 66 t IV METHODOLOGY ...... 67

General Research Design ...... 67 Instrumentation - Independent Variables .... 69 Instrumentation - Dependent Variables ...... 82 Sampling Procedures ...... 84 Channel 2000 Training Procedures ...... 91 Administration of the Questionnaires ...... 91 Analytic Procedures ...... 92

Phase I: Data Reduction ...... 92 Phase II: Independent and Dependent Scale A s s e s s m e n t ...... 93 Phase III: Analysis of Research Questions . 95

V RESULTS ...... 102

S u b j e c t s ...... 102

Variable Frequencies ...... 102

Phase I: Data Reduction ...... 104

Life Style Factor Analysis ...... 104 Activities Factor Analysis ...... Ill

Phase II: Independent and Dependent Scale Analysis ...... 114

Independent Variables ...... 114 Dependent Variables . . . 117

viii TABLE OF CONTENTS (Continued)

CHAPTER Page

V RESULTS (Continued)

Phase III: Analysis of Research Questions . . . 125

Research Questions #1 and #2 125 Research Question #3 131 Research Question //4 131 Preliminary Discriminant Analysis ...... 134 Research Question //5 140 Research Question #6 142

VI DISCUSSION ...... 147

The Perceptions of Channel 2000 ...... 149 Discriminating Between Potential Adopters and Non-Adopters ...... 154 Characterizing the Potential Channel 2000 A d o p t e r ...... 156

Innovation Perceptions ...... 157 Life Style ...... 158 Consumer Creativity ...... 160 Life Style Activities ..... 161 Demographics ...... 163

The Final Discriminant Analysis ...... 164 The Relationships Among Innovation Perceptions . 169 The Relationships Between Innovation Perceptions and Life Style, Activities, and Consumer Creativity ...... 169 Implications for Diffusion Theory ...... 171 Implications for Videotex Technology ...... 175 Limitations ...... 185 Recommendations for Future Research ...... 189 Conclusions ...... 193

FOOTNOTES ...... 196

BIBLIOGRAPHY ...... 197

APPENDIX

A Recent Videotex Projects ...... 204

B Videotex Applications ...... 216

C Channel 2000 Questionnaires ...... 225

ix TABLE OF CONTENTS (Continued)

APPENDIX Page

D Telephone Questionnaire ...... 259

E Screening Questionnaire ...... 262

F Questionnaire Cover Letters ...... 272

G Life Style Intercorrelation Matrix ...... 276

U Activities Intercorrelation Matrix ...... 279

x LIST OF TABLES

TABLE Page

1 Dimensions of the Teletext/Videotex Phenomenon .... 14

2 Summary Table of Innovation Attributes and Their Relationship to Innovative Behavior ...... 46

3 Welcome to Channel 2000 ...... 64

4 Questionnaire Mailout Dates for Times I, II, and III . 70

5 Perceptual Attributes and Corresponding Questionnaire Items ...... 71

6 Life Style Dimensions and Corresponding Questionnaire Items ...... 78

7 Questionnaire Variables by Design Phase ...... 81

8 Opinion Leadership Items and Score Values ...... 87

9 Opinion Leadership Distribution for Original, Selected, and Cooperating Samples ...... 89

10 Frequency Distributions of Channel 2000 Subjects . . . 103

11 Rotated Factor Pattern Matrix for Life Style ...... 105

12 Rotated Factor Pattern Matrix for Activities ...... 112

13 Alpha Reliability Coefficients for Perceptual Dimensions at Three Points in T i m e ...... 115

14 Reliability Coefficients for Life Style and Activity D i m e n s i o n s ...... 116

15 Correlation Matrix of Perceived Use, Price Sensitivity and Purchase Probability at Three Points in T i m e ...... 118

16 Frequency Distribution for Price Sensitivity at Three Points in T i m e ...... 120

17 Frequency Distribution for Purchase Probability at Three Points in Time . '...... 121

18 Frequency Distribution for Perceived Use (Time II) . . 122 xi LIST OF TABLES (Continued)

TABLE Page

19 Frequency Distribution for Three Collapsed Measures of Potential Adoption ...... 123

20 Frequency Distribution for Three Collapsed Measures of Price Sensitivity ...... 126

21 Means and Standard Deviations of Perceptual Scores at Three Points in T i m e ...... 127

22 Analysis of Variance Summaries for the Perceptual Scores at Three Points in Time ...... 129

23 Scheffe Test of Perceptual Means at Three Points in Time ...... 130

24 Correlation Matrix of Perceptual Scores at Time I . . . 132

25 Correlation Matrix of Perceptual Scores at Time II . . 132

26 Correlation Matrix of Perceptual Scores at Time III . . 132

27 Correlation Matrix of Innovation Perceptions by Life Style, Activities and Consumer Creativity at Time III ...... 133

28 Results of the Two Group Discriminant Analysis for Each Construct at Time I ...... 136

29 Results of the Two Group Discriminant Analysis for Each Construct at Time I I ...... 137

30 Results of the Two Group Discriminant Analysis for Each Construct at Time III ...... 138

31 Summaries of the Two Group Discriminant Analysis at Times I, II, and III ...... 141

32 Results of the Classification Analysis at Times I, II, and I I I ...... 143

33 Partial F Significance Levels at Times I, II, and III . 145

34 Structure Coefficients at Times I, II, and III .... 146

xli LIST OF FIGURES

FIGURE Page

1 Concept Sketches of Computerized Home Services .... 6

2 The Elements of Interactive Communication Technology . 12

3 A Partial Model of Social C h a n g e ...... 24

4 The Channel 2000 System ...... 59

5 The Channel 2000 Keypad ...... 60

xiii CHAPTER I

INTRODUCTION

On October 1, 1980, seventy-one randomly selected Columbus, Ohio households began testing a prototype computerized information retrieval service called Channel 2000. For three months, the 141 individuals who comprised this group of test households accessed over 800,000 pages of textual information which was transmitted over telephone lines and dis­ played on the users' existing television sets. Although the test participants were not charged for the use of Channel 2000, they were required to complete three comprehensive questionnaires during the test period. This dissertation details the specifics of this Channel 2000 test, and more importantly, reports on several aspects of these three questionnaire instruments.

Apart from the mere documentation of the Channel 2000 test, and listing of research findings, this dissertation makes use of an existing theoretical framework commonly referred to as diffusion theory. Diffu­ sion theory has been developed to help researchers understand the adoption or rejection of a newly introduced innovation such as Channel

2000. This theory guided the development of the Channel 2000 question­ naire instruments, and the interpretation of the research results. This research effort can therefore be characterized as an application of diffusion theory to an emerging technological innovation called Channel

2000.

1 The research itself was centered around two primary areas of investi­ gation. The first dealt with the perceptions of Channel 2000 as defined by the standard perceptual dimensions of an innovation which have prev­ iously appeared in the diffusion literature. These dimensions included relative advantage, compatibility, trialability, observability, com­ plexity, and perceived risk. Since the test participants completed three

successive questionnaires, it was possible to examine these perceptions at three points in time and assess the degree to which they changed or remained constant. These innovation perceptions have been documented

for the purpose of providing others with an increased understanding of how the test participants felt about Channel 2000 as a new communication

technology, and why Channel 2000 as an innovation was accepted or rejected

by the test participants.

The second area of investigation dealt with an investigation of the

differences which existed between potential adopters and non-adopters of

Channel 2000. In the past, variables which relate to a person's life

style or his personal activities, interests, and opinions on matters

related to a new innovation, have been used to describe those individuals

who are probable innovators, or early adopters. Other theoretically

linked personal characteristics such as opinion leadership and consumer

creativity have also been tied to innovative behavior. Standard socio­

demographic indices have also been a regular part of this type of analysis.

Up to this point, these variables have not been tested for their

ability to discriminate between potential adopters and non-adopters of

a new communication technology such as Channel 2000. In addition,

although the perceptions of an innovation have been suggested as an important determinant of the adoption process, few studies have looked at the discriminative strength of these perceptions in a field setting.

This research investigated the ability of life style, consumer creativity, sociodemographic, and perceptual variables to discriminate between potential adopters and non-adopters of Channel 2000. As part of this analysis, the relative importance of these variables in terms

of their own contribution to this adoption discrimination was also

assessed at three points in time.

The following research questions were designed to reflect the cen­

tral purposes of this research effort:

I. The Perceptions of Channel 2000

1. What are the perceptions of Channel 2000?

2. What are the differences between the perceptions of

Channel 2000 at three points in time?

3. What are the relationships between the perceptions of

Channel 2000 at three different points in time?

4. What are the relationships between innovation perceptions,

life style, and consumer creativity?

II. The Potential Adoption of Channel 2000

5. How well do innovation perceptions, life style, consumer

creativity, and sociodemographic variables discriminate

between potential adopters and non-adopters of Channel

2000 at three points in time?

6. What are the relative importance of innovation perceptions,

life style, consumer creativity, and sociodemographic variables in discriminating between adopters and non­

adopters of Channel 2000 at three points in time?

The Problem

The implications inherent to a computerized, in-home information retrieval system, have led academicians and industry spokesmen into a series of futuristic predictions based on the assumption that these systems have the potential for restructuring the way people conduct their daily lives and function within their work environment. The technology, according to Carey and Quirk (1973, p. 498), has been portrayed

as the ultimate communications machine; it combines the speed and intimacy of dialogue, the memory of history, the variable output of sight and sound, the individuality of total information combined with totally free choice, the political awareness and control of a fully informed and participant electorate, and the analytic skill of advanced mathematics.

Unfortunately, similar descriptions and discussions which revolve around the "disruptive potential" of this computerized information revolution are speculative at best. If one accepts the proposition that home information systems have the capability of changing society and industry alike, it seems critical at this point to understand how people currently perceive this technology, and to determine whether certain kinds of people have the tendency to accept or reject this technological innovation. Obviously this kind of understanding cannot be gained through speculation alone. People must be given the oppor­ tunity to experiment and use this technology within the confines of their own homes. Once this "educational process" has taken place, then it will become possible to understand potential use and acceptance of home information systems from the users' point of view. Although home computers that are currently available have created some initial awareness of computerized home services, awareness of

Channel 2000-like devices among the general population is extremely low.

Researchers who are faced with the problem of assessing public reaction to a Channel 2000 device must rely upon verbal descriptions and/or artist sketches of what the system looks like and how it works. For example, the sketches in Figure 1 were presented to respondents in a national proprietary research study just recently completed. Following a brief explanation of what a home information system is, how it works, and what it looks like, respondents were asked how often they would use the system, for what purposes, and approximately how much they would be willing to pay for the previously described service. Apart from the fact that the reliability and validity of these respondent estimates are subject to a certain amount of error, the findings themselves have not been released to the academic community.

One reported study by Goldman (1980) followed the same basic pro­ cedures outlined above. As part of the USC MEDIA 90 research program, respondents were asked in a series of telephone or personal interviews,

their reactions to a number of individual computerized services. In

these studies (a total of 13 have been conducted since 1973), respond­ ents were asked if they would or would not purchase each videotex serv­

ice for roughly the price of their basic telephone service. Here again,

subjects were essentially asked to predict their own purchasing behavior based on the brief scenarios provided by the interviewer.

Plummer, et al. (1980, p. 35) captured the essence of this research

problem by pointing out that it is extremely difficult to ask people 6

MCktt new wui

g> gl

Figure 1. Concept Sketches uf Computerized Home Services. for an evaluation of a technology that they have never heard of, or seen before. The Channel 2000 test was designed to address this problem by allowing test participants to have "hands on" experience for a reasonable trial period. The Channel 2000 test therefore represents the first publicly documented field test of a home information system where respondents' opinions, attitudes, and perceptions, were based on experi­ ence with the system.

Although diffusion theory has beer, identified as the vehicle to be employed in this research for organizational purposes, it should be pointed out that the theory itself needs additional work in several areas. First, although diffusion theory has been applied to an endless number of innovations, very few of these studies have placed an emphasis on the attributes or characteristics of the innovation as perceived by potential adopters.

The importance of perceptual variables as related to an innovation were discussed by Zaltman and Lin (1971, p. 670):

To the extent that the attributes of innovations can be manipulated, they are important to the change agent as entry points into the social change process. Thus, two steps are involved. The first is to determine current or most probable perceptions of the Innovation. The next step is, where necessary, to alter the attributes of the innovation thereby altering the perception of them. In doing so, the change agent is controlling the most funda­ mental element in the social change process, the social change object itself.

This logic suggests that a detailed examination of Channel 2000 percep­

tions can suggest future modifications for an increased probability of

adoption among the general population. In other words, the modification

of detrimental Channel 2000 attributes according to those perceptual

dimensions which are found to be critical to the adoption process, 8 should facilitate future Channel 2000 acceptance.

This reasoning of course relies upon the assumption that innovation perceptions are critical to the adoption process. Although Rogers and

Shoemaker (1971), Robertson (1971), Ostlund (1969a, 1969b, 1972, 1973,

1974), and McCain and Wall (1977) have provided some evidence in support of this assumption, perceptual variables have not been tested within the context of a communication technology such as Channel 2000. Until the relative importance of these perceptual variables in terms of their ability to discriminate between adopters and non-adopters is established, the question of exactly how important Channel 2000 perceptions are to the adoption decision will remain unanswered.

Similarly, those life style variables that could be theoretically linked to the potential adoption of Channel 2000 have also not been tested for their discriminatory ability. For example, it might be hypothesized that those individuals who have a strong negative opinion of current television programming might be more favorably predisposed towards a home information device. Other life style variables related to an individual's attitudes towards society, family life, or the degree to which an individual feels to be lacking discretionary time and unable to keep up with the news, could all be linked to an in-home electronic information/communication device.

The activities that are a part of someone's life may also be a reflection of the adoption process. For example, frequency of library, computer, banking, encyclopedia, and credit card use could all be related to the adoption of a Channel 2000 service which in fact offered an electronic library and encyclopedia, and also allowed some users to 9

conduct their banking activities from their homes. Furthermore, an

individual's income and educational level could also be an important

factor in the adoption process.

The discriminatory ability of these variables has not yet been

assessed for a communication technology such as Channel 2000. Also,

the relative importance of these variables in this regard has not been

examined at three points in time. Certain variables could in fact be

more useful as discriminators at differing points in the adoption

process. The ability to test these variables in a three month field

test provided a unique opportunity to investigate this area of diffusion

theory, and to also shed some light on those user factors associated

with the potential adoption of a much heralded technology such as

Channel 2000.

4 CHAPTER II

A REVIEW OF THE LITERATURE

This research attempts to address several theoretical issues asso­ ciated with the adoption of a new technology. Because the specific characteristics of any innovation have a very definite impact on the adoption process, it is important to gain a clear understanding of the

technological innovation which this research investigated. Chapter II

therefore provides the reader with an Introduction to interactive home

information systems, and those definitional terms that are currently being used to describe the technology itself.

Chapter II also outlines the theoretical framework that was used

in the data collection and interpretation procedures during the Channel

2000 test. Furthermore, the relevant literature that relates particu­

larly to the understanding of how and why people adopt new technologies,

is presented and organized into two main sections. The first literature

review section reports on the body of research which investigates the

characteristics of the adopting unit. The second section of this liter­

ature review examines research which pertains to the characteristics of

the innovation. The chapter concludes with a brief summary of the

literature presented thus far, which in turn is followed by the principal

research questions posed by this dissertation.

Defining Videotex

Most communication models are graphically described by three

central elements. The sender of a message represents the first element.

The second element consists of the channel in which the message is 10 11

transmitted. The third element is represented by the intended receiver of the message. Obviously there are additional elements inherent to most communication acts, but these three primary components serve as the basic descriptors.

The communication technology which this research investigates can be conceptualized in terms of these same three elements: the sending de­ vice, the channel, and the receiver. Unlike previous broadcast communi­ cation technologies, this newer technology adds complexity to the three element model by introducing the concept of interactive communication.

Interactive electronic communication implies a role reversal for both the

sending and receiving devices according to the demands of the communica­

tion situation. Thus, a typical transaction requires the sending device

to also function as a receiving or display device, once the initial com­ munication request has taken place.

Figure 2 graphically displays the essential elements of interactive

communication technology. A television set (the sender) coupled to a

decoder and keypad, transmits a request via any one of the communication

channels listed, to a centralized computer (the receiver). The computer

in turn responds to the request and relays, or extracts certain informa­

tion for display on the television set (which now functions as a receiv­

ing device).

Apart from the interactivity implied by this three element model,

there are several implicit complications inherent to this technology

which inhibit technocrates and scholars alike from placing distinct

labels or definitions on each unique communication combination of the

three element model. First, it can be noted from the transmission CHANNEL

/ Telephone \ r Broadcast Signals \ Cable Television ) Satellite Transmission

SENDER RECEIVER

Television with Transmission Decoder, Keypad Computers Home Computer

Request/Reception Storage/Retrieval

Figure 2. The Elements of Interactive Communication Technology*

Channel 2000 consists of a television, decoder and keypad, telephone transmission, and a centralized computer. 13 devices listed in Figure 2, that each of these devices is normally de­ fined as a unique communication technology unto itself. Should an inter­ active system that uses the broadcast signal as its channel, be defined in terms that are inappropriate for the same system which uses a satel­ lite transmission device as its channel? If this were the case, then society would be confronted with a multitude of new technological terms that vary according to their own unique combination of the basic three element model.

It is also extremely difficult to label a communication device that in reality represents several technologies that already exist. Eastman

(1980, p. 224) accurately points out that interactive communication de­ vices actually represent a "synthesis of other systems, technologies, and media forces coalescing to form an evolutionary information manipu­ lation vehicle." Definitional efforts should therefore be construed in terms of those factors which stress the systemic differences of the technology itself.

The problem with this approach however is that it becomes short­ sighted in terms of its ability to deal with certain aspects of this communication device beyond the mere technological or mechanical attri­ butes of the system. Plummer, Johansen, Nyhan and Holmlov (1979) have expanded upon the technological distinctions represented in Figure 2

(see Table 1) and included what the authors call "dimensions" of inter­ active electronic communication (see also Johansen, Nyhan, and Plummer,

1980). These dimensions are as follows:

Technology

System Design 14

TABLE 1

DIMENSIONS OF THE TELETEXT/VIDEOTEX PHENOMENON*

DIMENSIONS POTENTIAL ACTORS VARIABLES

Technology manufacturers of: data transmission rate semiconductors graphics resolution computers degree of interaction broadcast equip­ processing capabilities of ment user terminals TV sets telephone equip­ ment

System design service providers method of access to information information providers human factors (e.g., login pro­ engineers cedure, consequences of errors, computer scientists documentation, help features, overall "friendliness")

Content service providers commercially oriented content information providers (e.g., airline schedules, (e.g., newspapers, business yellow pages) publishers) sponsored information (e.g., software houses news, weather, "800" type phone users of the system information) transactions communication capabilities (e.g., electronic mail)

Users general public cohesiveness of group (e.g., the businesses deaf) public agencies organization of group national and region­ degree of access al interest groups information need (e.g., Consumers Union) libraries

Service TV and radio scope of system operation (e.g., providers stations national vs. local) TV networks activities of service providers cable companies (e.g., provide both transmis­ MDS companies sion and content, provide just common carriers transmission) computer networks libraries USPS 15

TABLE 1 (Continued)

DIMENSIONS OF THE TELETEXT/VIDEOTEX PHENOMENON

DIMENSIONS POTENTIAL ACTORS VARIABLES

Economics of system providers . source of income (direct user system billing or indirect advertiser operation support) mechanisms of billing (e.g., decoder rental, telephone add-on charge)

Regulatory and international control of information policy organizations responsibility for information environment FCC pricing NTIA ownership of information Congress technical characteristics industry organiza­ connections to regulated tech­ tions nologies (e.g., attaching local or state devices to TVs) governments who enters market? incentives (e.g., for public service) compatibility of equipment

* From Plummer, R. P., Johansen, R., Nyhan, M. J., and Holralov, P. G., 4004 futures for teletext and videotext in the U. S. IEEE Trans­ actions on Consumer Electronics, 1979, 25:3, 318-325. 16

Content

Users

Service providers

Economics of system operation

Regulatory and policy environment

Each of these dimensions implies an extremely complex interaction among all the variables resulting in what Plummer, et al. have termed a communication "phenomenon”. An examination of the dimensions and vari­ ables listed in Table 1 underlines the complexity and multidimensional nature of this interactive means of communication. The reader is there­ fore cautioned to be cognizant of the fact that future distinctions among technological communication terms may in fact represent various combin­ ations of the seven variable model.

For the present usage, the International Telegraph and Telephone

Consultative Committee (CCITT) has attempted to remedy the complexity of this situation with the use of three specific technological terms. The

Committee has stated that the generic term "videotex" should be used to describe any form of electronic interactive communication which involves the sending or receiving of information or entertainment between a re­ mote computerized communication facility (usually a modified home tele­ vision set), and a larger, centralized.computer (see Noll, 1980, p. 23).

Teletext and Viewdata

The term videotex is further subdivided into two separate, but sometimes overlapping definitional constructs. "Teletext" is being used generically to describe those videotex systems which use "ancillary" broadcast signals to transmit alphanumeric data and simple graphics 17

into the home. The transmitted material, consisting of encoded digital

material, is inserted into the unused portion of the vertical blanking

interval of the television program signal. Users who have a television

set equipped with a key pad and decoder can specifically request a cer-

* tain page of information as it is sequentially cycled through the tele­

text system. This information in turn can be superimposed over the

regular television signal, or it can constitute a separate display.^-

The most publicized teletext systems that are now available to the

public are Britain's Oracle and CEEFAX systems, and France's Antiope

system.

The second subdivision of videotex consists of "viewdata" systems

which is the term used to define the Channel 2000 system. With view­

data, users are directly connected to the database computer via tele­

phone or cable television lines. This enables the user to achieve a

higher level of interactivity since the computer responds to the user's

specific requests for information or service. With teletext the user is

in effect "capturing" information as it is sequentially broadcast, and

a viewdata user is in effect "activating" the computer, according to the 2 information requested. The most widespread viewdata system currently

in use is the British Post Office's Prestel system, and of course,

Channel 2000 is also defined within the context of this research as a

viewdata system. For a listing of current videotex services, and tests

that are being conducted, see Appendix A, and also see Criner (1980),

and Leduc (1980).

An important distinction should be made between videotex and

computer timesharing, for the sake of defitional clarity. The essential 18 distinguishing element of a videotex system is that it offers to its users prespecified and defined sets of information (normally divided into singular "pages" of information) that are not subject to user modification. In other words, in timesharing, a user can add to, re­ structure, or transform the information according to his or her specific needs. A viewdata information base, however, is restricted to a pre­ defined set of information pages which are offered on a mass availability basis due to certain cost efficiency factors (see Simons, 1980). This distinction between videotex and timesharing will most definitely become blurred in the future, as will the conceptual distinctions between tele­ text and viewdata systems. For the purpose of this research, these definitions can only help establish some parameters and guidelines for the discussion to follow.

Projections for Videotex Systems

When the Institute for the Future convened in December of 1980 to discuss the potential applications of a videotex system, they derived a list of approximately 60 specialized videotex services for consideration.

In Appendix B, these application areas are grouped according to the following eleven main headings: community services, advertising and marketing, games and entertainment, home-based services, education, communication, library services, business/office services, banking, health, and electronic publishing. A review of these applications most certainly leads one to believe that videotex has the capability of dramatically changing the way we live our lives, and how we conduct our day-to-day activities either at home, or at work. This disruptive capability of videotex is the essential factor which accounts for the 19 widespread forecasting tendencies described in the introduction of this report. An additional sampling of these forecasts follows.

At the first international videotex conference and exhibition held in London, and entitled Viewdata '80, almost 1,000 delegates took part in a comprehensive three-day program. Eastman (1980) reports that during

Viewdata '80, estimates of 1983 penetration levels for viewdata-equipped

television sets ranged from 10,000 units to 10 million. A Delphi fore­ casting study reported by Vermilyea (1980) required a panel of experts

to respond on a nine-point scale (ranging from very likely to very un­

likely), to a series of technological projections regarding the status of videotex in the year 2000. A consensus was reached on the issue

that it will be "likely" for more than half of U.S. homes with television

to have teletext receiving capability by the year 2000.

A management consulting firm, International Resource Development,

Inc., recently released estimates on the potential videotex market.

According to IRD, 24 million (about 1 in 4) U.S. households are expected

to have installed integrated video terminals and other information

systems capable of accessing videotex data bases, by the year 1990. The

report further states that revenues from these videotex information

services and related transaction services are expected to reach the

$3.5 billion mark by this same year. IRD concludes that $2.7 billion

of this amount will be paid by advertising.

Naturally videotex offers unique advertising opportunities, and the

advertising community has not been silent throughout this recent forecast­

ing period. Senior Vice President of Benton & Bowles, Michael D. Moore,

recently advised members of the American Association of Advertising 20

Agencies that cable will become the dominant broadcast delivery system

(cable has the broadband transmission capability necessary for videotex devices), but will get there slowly, with perhaps a 42% penetration level 4 by 1990. For the advertising community, a wired videotex nation will provide a more cost efficient means of reaching a specific group of potential clients, who in turn have the capability of directly purchasing a product through their videotex/cable system.

The newspaper industry has also responded to videotex as a threaten­

ing technology that cuts directly into the information services currently offered by local newspapers. Aptly put by J. Richard Munro, President/

Chief Executive Officer of Time, Inc. (owner of The Washington Star) ,

"why get your hands dirty" with a newspaper when you can press a

button?^ Others have posed the question, why pay for the whole newspaper when most readers are interested in only part of it?^ The newspaper

industry has generally cast a rather pessimistic eye to the future in

recognition of the impending diffusion of videotex technologies.

Futuristic projections would not be complete without some mention

of Alvin Toffler’s (1980) reaction to videotex in the home. One central

thesis put forth in Toffler's The Third Wave (1980), is the move of

society from one characterized by centralization and mass production,

to one of individualism and de-massification. A singular factor attrib­

uted to this social transition, is the diffusion of personalized infor­

mation and entertainment services into the home. Toffler envisions an

electronic communication center In the home that would enable a growing

percentage of the population to conduct the majority of their activities

(work and leisure) within the confines of their own home. Toffler 21 suggests that life within this "electronic cottage" will dramatically change our society and the socioeconomic factors which govern it.

The extent of these socioeconomic changes predicted by Toffler and others are of course governed by the people who either decide to accept or reject videotex technology itself. This issue of consumer accept­ ance and the identification of consumer need, is one which has been brought up in the literature. For example, videotex has been called

"a technology in search of a function" (Toles, 1981, p. 2), "a product in search of a market" (Eastman, 1980, p. 221), and "a product ..." that "is invented simultaneously with the dysfunction it is designed to cure" (Weizenbaum, 1979, p. 441). Toles, Eastman, and Weizenbaum agree that the real issue is whether there actually exists a need for videotex technology. Forecasters inappropriately assume that since we currently have the videotex technological expertise, consumer need for videotex will supposedly be strong enough to permit the widespread diffusion patterns associated with a disruptive innovation. A prime example of a costly mistake resulting from this line of reasoning is the $200 million Picturephone service offered by Bell Telephone Labora­

tories. Eastman (1980) reports that AT&T projected one million Picture- phones for in-home use by the mid 1980's. Unfortunately the public did not respond with such projected enthusiasm, and the Picturephone has

gone down in AT&T history as one of its major failures.

This same kind of speculative enthusiasm for videotex should be

tempered by a close examination of how useful the videotex service is

to the individual. In a speech to an industrial group in London,

Dr. Walter Ciciora of Zenith Radio Corporation warned: "Failure to 22 accommodate the consumer will likely result in disaster...Teletext and 7 view data must be useful. The more obviously useful the better." The task of how useful a videotex service is to an individual must be system­ atically addressed prior to the futuristic forecasts for videotex tech­ nology. The previously described Picturephone experience has demon­ strated that the availability of a communication technology does not insure hypothesized popular use.

There is however an available approach to this problem that may provide an understanding of why people either accept or reject a tech­ nology such as videotex. Such an understanding can be found through a systematic analysis of the potential videotex adopter, guided by a theoretical framework designed to organize and interpret the research results. What follows is a description of such a theory.

Diffusion of Innovations; A Theory of Social Change

Rogers (1969, p. 3) defines social change as

the process by which altercation occurs in the structure and function of a socidl system...The structure of a social system is provided by the various individual and group statuses of which it is composed. The functioning element within this structure of status is a role, or the actual behavior of the individual in a given status.

As might be expected, social change occurs at both the individual level and at the social system level. Rogers and Shoemaker (1971, pp. 10-11) suggest 'that analysis at the individual level be termed a microanalytlc approach to social change, while a macroanalytic approach refers to a social system level of analysis.

In an effort to graphically represent the dynamics of these change levels, Zaltman and Lin (1971) have proposed a model of social change.

For the purpose of this research, a simplified version of the Zaltman 23 and Lin model is presented in Figure 3. An initial examination of this model reveals three central elements of the social change process.

The first important element of the model is represented by the introduction of an innovation. Rogers and Shoemaker (1971, p. 19) define an innovation as "an idea, practice, or object perceived as new by an individual." Zaltman and Lin (1971) state that the innovation is the basic dynamic core of social change, and Barnett (1953) argues that the innovation is the basis of all personality, social, and cultural change.

A clear understanding of the innovation itself is critical to the social change model. Attention and detail will be given to this subject in a subsequent section.

The second important element of the model reflects the adoption process. Following the introduction of an innovation to an individual or adopting unit, an evaluation process takes place which involves the interaction between the attributes of the innovation, and the attributes of the adopting unit. This interaction results in a decision to either adopt or reject the innovation. Adoption in this sense refers to the acceptance and continued use of the innovation over time (Robertson,

1971, p. 56).

Zaltman and Lin (1971) conceptualize this adoption process as one which represents only one-half of the dynamic core of social change.

Figure 3 has accordingly been adapted to reflect the structure and sub­

division of micro and macro levels of analyses. The second half, or

macroanalytic level of social change, is represented by the third

important element of the model and is called the diffusion process. Attributes of the Innovation Structural Change Introduction of the Innovation Adoption Social Diffusion Process Interaction Process Evaluation Decision

Rejection Functional Attributes Process Change of the Adoption Unit

MICROANALYTIC MACROANALYTIC (Individual Level) (Social System Level)

N3 Figure 3. A Partial Model of Social Change. (Adapted from Zaltman and Stiff, 1973.) Rogers and Shoemaker (1971, p. 12) have defined diffusion as "the process by which innovations spread to the members of a social system."

The central elements in the diffusion process involve the communication of the innovation, through certain channels, over time, among the members of the social system with certain consequences (p. 18). Here the emphasis is on patterned interaction among potential adopters within the specific social system. The variable of time traces the culmina­ tion of adopting units, while an analysis of the communication channels describes interpersonal interactions, and the influence of the mass media. The end result of this diffusion process results in either structural or functional changes in the overall social system itself.

Zaltman and Lin (1971, p. 658) sum this process up by stating:

When an innovation is diffused to and adopted (either symbolically or behaviorally) by a sufficient number of relevant units in a social system so as to register an impact (becoming an integrated part of the normative patterns in the system), it is said that a social change has occurred for the system under discussion.

A Microanalytic Approach to the Study of Social Change

Ideally, the study of social change would entail both micro and macroanalytic procedures that trace patterns of social change from initial innovation introduction, to eventual adoption and diffusion.

Clearly the scope of such an undertaking is enormous, and the factor of limited research time is one which is most difficult to overcome (some innovations may take 10 or more years to diffuse through society before social change can be documented). As is the case with this research investigation of an innovation, most social change studies therefore deal specifically with only certain aspects of the overall model. The developmental status of videotex technology automatically reduces this 26 research to the microanalytic level since videotex services are not com­ mercially available in the United States at the present time. It is therefore virtually impossible to study the total diffusion process if the innovation is not available for adoption.

It is possible, however, to examine the initial stages of the social change process at the microanalytic level for a technological innovation such asChannel 2000. According to the model of socialchange developed thus far, research at this level would look at three basic questions:

1. What are the attributes of the innovation?

2. What are the attributes of the adopting unit?

3. How do these two factors interact to result in an adoption

or rejection decision?

An analysis of the literature that has examined these three areas re­ veals a distinct area of emphasis, and also suggests several potential areas for improvement, and development.

The Diffusion Literature

In a critical summary of the literature, Rogers (1976) states that there are over 2,700 publications, including 1,800 emperical research reports and 900 other writings, that deal specifically with the diffusion of innovations. Although diffusion research is a particular type of communication research, the vast majority of this literature is derived

from fields other than communication. A listing of these fields includes education, anthropology, geography, medical sociology, marketing, agri­

cultural economics, psychology, economics, industrial engineering, and most of all, rural sociology. Among these perspectives, the marketing or consumer behavior approach to new product adoption, is the perspective which most closely aligns with the nature and purpose of the microanalytic approach of this research. A marketing approach to new product adoption is one which seeks an understanding of why certain innovations are either adopted or rejected from the perspective of the consumer. Recognizing that the consumer is the "final arbiter of success" for a new product, market researchers strive for the identification of specific product attri­ butes that will be perceived by consumers as fulfilling certain unmet needs (see Engle, Blackwell, and Kollat, 1978, pp. 3-7). This approach most closely relates to the previously expressed objective of under­ standing the attributes of the innovation, and how these attributes affect adoption.

A marketing approach to new product adoption also attempts to specify those promotional or communication strategies that will be most effective in reaching potential adopters of a new product. A great deal of emphasis is therefore placed on the identification of personal factors which correlate with new product adoption so that marketers can

Influence the adoption process to their advantage. Thus, there is also an available body of literature which looks at the attributes of the adopting unit.

The Attributes of the Adopting Unit

Rogers and Shoemaker (1971, p. 135) summarize these two approaches by citing the apparent biases that currently exist in these two areas:

When one persues the diffusion research literature, he is impressed with how much effort has been expended in study­ ing "people" differences in innovativeness (that is, in determining the characteristics of the different adopter 28

categories) and how little effort has been devoted to analyzing "innovation" differences (that is, in investi­ gating how the properties of the innovation affect its rate of adoption).

"People differences" in diffusion research refers to those constructs or variables that are suggested to be correlates of early product adoption, where researchers attempt to measure the degree of innovativeness among members of a social system.

The identification of innovators is normally designated by the time

interval from initial product or innovation introduction, to time of

adoption. Rogers and Shoemaker (1971, p. 179) have found that the dis­

tribution of adopters by time of adoption tends to approximate a normal

distribution. As such, rural sociologists have applied the character­

istics of a normal distribution to derive adopter categories according

to the standard deviations of the population. Accordingly, the first

2.5 percent of the individuals to adopt an innovation are termed inno­

vators. The next 13.5 percent are termed early adopters, who are fol­

lowed by the early majority (the next 34 percent), the late majority

(the next 34 percent), and finally by the late adopters or laggards (the

last 16 percent). See Rogers and Shoemaker (1971, p. 181), Robertson

(1971, pp. 84-85), Zaltman and Stiff (1973, p. 481).

If marketers can distinguish and identify people who adopt at

various stages of this process, then marketing and communication pro­

grams can be directed towards unique consumer groups, who have unique

personal attributes, perceptions, personalities, life styles, and needs.

Most marketing research has therefore been a descriptive investigation

of the relationship between early adoption and those characteristics

which may be theoretically related to early adoption (Burnkrant, 1979). 29

Rogers and Shoemaker (1971) have reviewed over 3,000 research works which have related various independent variables to innovativeness (also see Rogers and Thomas (1975), Rogers and Stanfield (1968), and Robertson

(1971).) Zaltman and Stiff (1973), and Ostlund (1969a, 1969b) have conducted similar reviews for those studies which have dealt with marketing variables. This review will highlight the most relevant of

these findings, and will also review the research that has taken place since 1973.

Sociodemographics

Rogers and Stanfield (1968) specify from their review of the rural

sociology literature that innovativeness is directly related to more education, higher income and level of living, knowledgeability, higher achievement motivation and educational aspiration, more mass media ex­ posure, and more active participation in group activities. Age was not

found to be a good discriminator of innovative behavior. It should be

noted that these findings, and the majority to follow, are based on

correlation studies which do not state the directionality of causation.

Rogers (1976, p. 295) cautions that it is implied that these independent

variables lead to innovativeness, although in reality all that has been

determined is that these variables covary with innovativeness.

Robertson (1971) provides a compilation of consumer innovation

characteristics which spans twenty-one product categories (see pp. 100-

101). Among those studies which examined demographic characters, income

was found to be most closely associated with innovativeness. Nine of

the thirteen studies examined found higher income levels among high

innovators, while four show a lack of relationship. Educational level 30 and occupational status showed a similar, but not as strong relation­ ship to innovativeness, while age and number of children in the family were found to be unrelated to the degree of innovativeness. Thus innovators for consumer products tend to be better educated and have higher incomes than non-innovators.

Subsequent analysis of these findings has, however, questioned the usefulness of demographic indicators for assessing innovativeness.

Darden and Reynolds (1974) specifically address this issue in their study of innovative behavior for three product categories. By applying cluster analysis techniques to a series of demographic variables that were supplemented by personality and life style variables, a total of six innovator groups emerged. The authors point out that innovativeness is a multidimensional characteristic that goes beyond the gross cate­ gorization schemes of demographic factors, as demonstrated by their analysis. These researchers also found that overall innovative behavior

for men's products is related negatively to age, number of children, and education. It is positively related to mobility, information seek­

ing and opinion leadership for fashion, male grooming products and

hair care products, and life style venturesomeness.

Life Style

A life style approach to the study of innovativeness for a new

product is based on the premise that individuals manifest particular

and sometimes unique ways of doing things. These behavioral patterns

reflect individual attitudes and behavioral predispositions that have

developed through the course of a person's life. Lazer (1964, p. 130)

expands on this concept through his definition of life style: 31

Life style is a systems concept. It refers to the distinctive or characteristic mode of living, in its aggregative and broadest sense, of a whole society or segment thereof. It is concerned with those unique ingredients or qualities which describe the style of life of some culture or group, and distinguish it from others. It embodies the patterns that develop and emerge from the dynamics of living in a society.

The most commonly used approach to the measurement of life style has been to first determine those life style dimensions that can be hypothetically related to the behavior in question, and then generate

items that represent that dimension. Wind and Green (1974, p. 106) have

found five predominant dimensions that have been used in the past to measure and describe an individual's life style. These include:

1. The products and services the person consumes.

2. The person's activities, interests, and opinions.

3. The person's value system.

4. The person's personality traits and his concept of "self."

5. The person's attitudes toward various product classes which may

include the benefits he seeks, and the problems the product can solve.

Personality, life study, and psychographic variables are extremely

alluring to market researchers since they add considerable dimensionality

to the understanding of the innovator. The research findings in this

area are, however, somewhat inconsistent and difficult to apply across

product class. For example, Boone’ (1970) used the California Personality

Inventory in an analysis of innovators for a cable TV system. Here

innovators were found to possess greater self-confidence, acceptance

of newness, and more leadership ability. Robertson and Kennedy (1968)

applied a discriminant analysis to a series of personality and life

style measures in an effort to profile innovators for Touch Tone 32 telephones. For this study it was found that two variables, venture- someness (willingness to take new product risks) and social mobility

(movement up the social class hierarchy), accounted for most of the dif­ ference between innovators and non-innovators (64 percent of the total variance).

Arndt (1968) applied a similar procedure to assess the relative importance of twelve factors for a new brand of coffee. The following four factors, in descending order of importance, explained 84 percent of

the difference between innovators and non-innovators: lack of brand loyalty, deal proneness, high usage rate of coffee, and low perceived risk. King (1964) correctly classified 74 percent of his sample into buyer and nonbuyer groups for a line of clothing (hats). The six most

important variables were frequency of hair permanents, social visiting with other women, family income level, frequency of changes in hair

styles, attendance at spectator sports, and the psychological character­ istic of change orientation.

Specific measures of personality traits have also been proposed as

possible correlates of innovativeness. Robertson and Kennedy (1968)

have shown that venturesomeness accounts for 35 percent of the differ­

ence between innovators and non-innovators. Donnelly and Ivancevlch

(1974) used Kassarjian's 0.0. Social Preference Scale to profile inno­

vators of a new car line. This scale indicates the degree to which an

individual is either tradition-directed (oriented in the traditional

ways of their forefathers), inner directed (those who turn to their own

inner values and standards for guidance) or other directed (those who

depend upon the people around them to give directions to their actions) 33

(see Kassarjian, 1965.) Donnelly and Ivancevich found innovators of

Ford Mavericks to be more inner directed than other directed.

Robertson and Myers (1969) and Jacoby (1971) have questioned the postulated relationships of personality variables with innovative be­ havior based on the weak relationships that are evidenced in the litera­ ture, and the weak theoretical foundations upon which the research is grounded. Robertson and Myers (p. 167) state "that at best, personality measured by a standardized, normative, self-designating personality in­ ventory, have little, if any, relationship to innovative behavior."

Jacoby, however, has argued that this outright denounciation of per­ sonality variables is without justification if a theoretical framework

is provided which explains why personality should be related to an aspect of consumer behavior. As an example, Jacoby tested the relation­

ship of "dogmatism" to innovativeness within the guidelines of an appropriate theoretical framework. The results indicated that dogmatism

accounted for 10 percent of the variance between dogmatism and innovative

proneness (r = -.316; i.e., persons low in dogmatism were more likely

to be innovators). Coney (1972) supported Jacoby's findings in a repli­

cation of the study; however, Blake, Perloff, and Heslin (1970) were

unable to obtain conclusive results.

Although Jacoby's (1971) findings are not overwhelmingly powerful

in their explanatory ability, Goldberg (1976) and Churchill (1979) have

both recently supported those procedures used by Jacoby in the develop­

ment of a theoretically "relevant" and reliable/valid measure of per­

sonality. Goldberg (1976) stresses the importance of identifying the

primary functional relationships between the product and the personality 34 trait under examination. Thus a "matching" procedure between product and personality measure should take place in an a priori fashion, and not in the typical "shotgun" or ex post facto manner. The end result of such a procedure is the development of product specific measures of personality or life style that have meaning and relevance to the re­ searcher. Goldberg (p. 164) appropriately points out that while it is not expected that personality will be the exclusive variable in the explanation of innovativeness, the underlying assumption is that "appro­ priately conceptualized, personality can have a significant part to play."

Churchill (1979) has outlined a procedure for the development of reliable and valid measures of personality traits. Briefly, any measure of personality must: (1) specify the domain of the construct; (2) gen­ erate a sample of items; (3) purify the measure; (4) assess the relia­ bility with new data; (5) assess the construct validity; and (6) develop norms for comparison.

Consumer Creativity

Leavitt and Walton (1977) have recently applied the recommendations

of Goldberg and Churchill to the development of a "consumer creativity

scale." Leavitt and Walton suggest that there are a number of psycho­

logical constructs related to the outward expression of an individual's

personality that might explain why an individual would be among the

first to try a new product. These constructs are as follows: (1) de­

gree of rigidity (a low score would suggest a higher propensity to

innovate); (2) need for social approval, exhibitionism, or status

validation (a high score suggests new product use for satisfaction of 35 these desires); (3) impulsivity or self-control (high impulsivity and low self-control lead to new product purchase); and (4) cognitive style

(the tendency to try new ways of doing things in problem areas leads to a broader use of the new and unfamiliar).

Although Leavitt and Walton (1977) refer to creativity as a "psy­ chological" construct, it more closely aligns with what has previously been termed in the innovation literature a "personality" construct.

This research followed Markin’s (1974, p. 334) rather broad definition of personality as "the sum total of an individual's patterned character­ istics that make him unique." Creativity in this sense can therefore be thought of as one element of this "sum total" that uniquely defines an individual’s personality. Leavitt and Walton defined the patterned characteristic of creativity in terms of how an individual expresses this trait in relation to innovative products. Specifically, Leavitt and Walton (1977, p. 4) offered three criteria which were deemed essen­ tial for the understanding of the creative consumer:

1. The creative consumer is more open to information of all sorts and more likely to consider new and un­ familiar ones.

2. The creative consumer is more critical of both infor­ mation, sources and the idea (product), but is willing to experiment or to engage in suspension of disbelief.

3. Creative consumers solve problems by changing their frame of reference as opposed to solving problems within an existing framework.

By applying the procedures recommended by Churchill (1979),

Leavitt and Walton developed a 30 item creativity scale that incorpor— ated measures of the aforementioned personality traits. Adler (1975)

found that the creativity scale successfully separated new product 36 innovators for ten separate products to a significant degree, while

Jackson's Personality Research Form did not. Kaigler-Evens (1975) compiled fashion suggestions that had been judged to possess varying degrees of newness by fashion buyers in clothing stores. She then correlated creativity scores for a group of subjects with their degree of agreement on these fashion suggestions. Results showed an r of .71 for extremely new fashion suggestions versus .12, .09, and .08 for very, fairly, and not new (p < .001).

Further evidence for the predictive validity of the consumer crea­ tivity scale is provided by Craig and Ginter (1975) who attempted to discriminate between buyers of Mustang II's and other types of new cars.

The creativity scale successfully discriminated these buyers at a level of significance better than .05. Leavitt and Keitz (1980) used the con­ sumer creativity scale and fourteen other scales in a survey of fashion innovativeness. These researchers found the creativity scale to be correlated with new fashion ownership to a greater degree than any of the fourteen scales used (r = .31).

In summary, it appears that those personality measures that have some theoretical relationship to the trait of innovativeness, hold the most promise for describing and understanding innovative behavior (see

Kassarjian, 1971). Leavitt and Walton's consumer creativity scale has 1 demonstrated its predictive usefulness In this regard. Those studies which have used various life style and psychographic measures in an attempt to profile typical innovators have resulted in what Burnkrant

(1979) has summarized as a "very vague picture of the innovator."

This is due in part to the inconsistency of the measures that were used 37 across several different studies, and is also a result of the variations in the products that were studied.

Robertson (1971) directly addresses the issue of whether an innovator for one product can be thought of as an innovator for a dif­ ferent product. Following a review of the research evidence, Robertson concludes that "consistency of innovativeness cannot be expected across product categories, but can be expected within product categories and, sometimes, between related product categories" (p. 111). It appears to make intuitive sense that an innovator for videotex services might not be profiled in a similar manner as an innovator for a new automobile line. This does not discredit the value of life style and psychographic measures in the understanding of innovativeness, but only suggests that the findings of one study for one product class, may have limited gener- alizability for other studies within a different product class. However, if research can identify and profile the innovator for videotex services in general, this information can be extremely valuable to other research­ ers interested in this very same technology.

Interpersonal Communication

Apart from the personality, psychographic, and life style measures that have been reported thus far, there are additional independent variables that were especially relevant to this research, and its under­ standing of the adopter unit. One especially important area deals with those variables that assess how an innovator transmits and receives infor­ mation about the innovation. Basically there are two main channels of information regarding innovations that are available to a consumer. The mass media provide impersonal information from marketing controlled 38 sources such as advertisements, and impersonal neutral sources, such as news information. Personal information about innovations is the second channel of information, and is thought of as a supporting information source that can carry a considerable amount of weight in the adoption process. Katz and Lazarsfeld (1955) have summarized the role of personal influence in the spread of new ideas in their book, Personal Influence.

This book, and the seminal work of Lazarsfeld, Berelson and Gaudet

(1944) have led to the two step flow model which recognized the mass audience as a group of interacting individuals. Recent research has advocated a multistep flow model where the media provide messages to opinion leaders who in turn spread messages to other members of the audience, who in turn spread messages..., etc. (See Trodahl and Van Dam

(1965), Van Den Ban (1964), Trodahl (1966), Deutschmann and Danielson

(1960), Katz and Lazarsfeld (1955), and Robertson (1966).)

Rogers and Shoemaker (1971, p. 35) define opinion leadership as the "degree to which an individual is able to informally influence other individuals' attitudes or overt behavior, in a desired way with relative frequency." Marketers have recognized that if innovators can also be identified as opinion leaders, then the diffusion process can be greatly facilitated by word of mouth communication. Robertson (1971) reports that an association between opinion leadership and innovativeness is shown in ten out of the thirteen reviewed studies which included such a measure. Summers and King (1969) examined opinion leadership for four separate new products and found opinion leaders to be more favorably predisposed toward testing and experimenting with new products than were non-opinion leaders. The authors conclude that opinion leaders tend 39 to have fundamentally more favorable attitudes toward both new products as a concept and toward new products within the opinion leaders' specific areas of influence.

Arndt (1967) found in a new food product study that opinion leaders were not only relayers of information, but also transmitted personal evaluations of the innovation. Furthermore, these positive interpersonal evaluations tended to increase the probability of purchase for the new product. Other studies (see King and Summers (1967), Katz and Lazars­ feld (1955), Katona and Mueller (1955), Feldman and Armstrong (1975),

Blackwell and Talarzyk (1977)), have provided additional support for the importance of opinion leadership in product purchase probability. The need to identify the presence or absence of these opinion leaders within a designated group of innovators is therefore critical to the understand­

ing of the adopting unit.

Media Use

The second source of information that is available to the public

in regards to new product information comes from the mass media. Again

it might be expected that innovators would have media behaviors that

would differ from non-innovators. Robertson (1971) reports that greater

print readership is associated most often with high innovativeness,

while television viewership is generally unrelated to innovativeness.

Rogers and Shoemaker (1971, p. 189) conclude that innovators generally

have greater exposure to the mass media than do later adopters. A 1966

review by Rogers and Stanfield showed that approximately 86 percent of

the 49 published findings that included measures of media exposure,

resulted in a positive association between increased media exposure 40 and innovativeness. A recent investigation of innovativeness for eight different new consumer innovations conducted by Lambert (1972) tested the differences of media exposure for innovators and non-innovators. The four measures of media exposure (number of magazines read monthly, number of newspapers read, average number of hours-per-week listening to radio, and average number of hours-per-week watching television) were found to not significantly differ between innovators and non-innovators.

The weak relationship between media exposure and product innovation in the Lambert study might be attributed to the nature of the products themselves. Lambert looked at innovation among products such as freeze dried coffee, canned pudding, and a laundry presoaker. As Robertson

(1971, pp. 104-105) specifies, a relation between media exposure and product innovation might be expected if the new product is related in some manner to the media. For example, Robertson suggests that it might be expected that magazine readership (specifically, fashion magazines) would be highly related to fashion innovativeness. A parallel thought in this regard would be that book readership would be related to video­ tex innovation if the videotex service offered an electronic card cata­ log service to its subscribers.

The relationship between media exposure and innovativeness in many ways can be best understood from the uses and gratifications perspective

that has been proposed by Katz, Blumler, and Gurevitch (1974). Here researchers look at what functions the media performs for the individual, and how these functions might interact and help to explain media ex­ posure and use. The appropriate question to be asked for research that

investigates videotex innovation therefore becomes, what is the media 41 currently being used for, and how will the purchasing/subscribing to a videotex service either be related to, or eventually affect subsequent media use? This approach to the understanding of innovation and new product adoption is extremely relevant to an innovation such as videotex that has the capability of replacing or displacing many of the functions that are currently being performed by standard media fare. Of course, given the multiplicity of videotex services that may become potentially available to consumers (see Table 1), researchers are cautioned to explore only those videotex-media relationships that have the capability of sharing similar functional utilities for the innovating social system.

Product Use

Related to the concept of functional utility is the marketing con­ cept of product usage. Robertson (1971) reports that it has been docu­ mented in seven out of eight consumer studies he reviewed, that inno­ vators for a new product are heavy users of that product or similar products. Thus it might be expected that the innovator for a new tele­ phone product is likely to have a larger monthly telephone bill. Or innovators for a library videotex innovation might be expected to read more books and use the library more frequently. A banking videotex

innovation might also be adopted earlier by those individual who have

frequent banking use.

Taylor (1977) explored these kinds of relationships in his twelve month study of eleven innovative products. Taylor tested the hypothesis

that innovativeness is independent of product class use (amount of

product class purchased). The results of his study rejected this

hypothesis since in ten out of eleven instances, those people who purchased more of the product class, also tried the new product (the innovation) earlier. Taylor also reports that continued use of the innovation was greater among heavy users of the product class. Taylor concludes that heavy users of a product class are most likely to be innovators of a product in a similar class. Also, the reaction of the heavy users to an innovation is important because of the potential volume of business that they represent. It seems that an understanding of those behaviors related to the use of a defined videotex service is also extremely important to the understanding potential videotex innovators.

The review presented thus far has suggested that an understanding of

the innovator or adopting unit is important for five central reasons.

1. Innovators account for initial levels of penetration.

2. Innovators can influence later adopters through personal influ­

ence and opinion leadership.

3. Innovators can account for a disproportionately higher volume

of business for the innovation.

4. Communication and promotional efforts can be developed in a

more systematic and defined manner through an understanding of

the innovator.

5. The innovation itself can be altered to accommodate the inno­

vator for an increased probability of acceptance.

Although the adopter centered approach to innovation can lead to an

increased understanding of the innovator, additional, research evidence

has demonstrated that there is an alternative approach to this under­

standing that places its emphasis on the attributes of the innovation. 43

The Attributes of the Innovation

It has already been suggested in this review that more research has been completed on the attributes of the adopting unit than on the attributes of the innovation (and their subsequent effect on adoption).

Of the 6,811 diffusion studies reviewed by Rogers and Shoemaker (1971,

pp. 72-73), only 82 (1.2 percent) investigated attributes of innovations

as a factor in the adoption of innovations. This disparity is somewhat

unexpected in light of the fact that innovations exhibit a wide variety

and range of traits or attributes. Without a consideration of this

"innovation variance," most diffusion researchers have mistakenly

treated all innovations as equivalent units from the viewpoint of study

and analysis (Rogers and Shoemaker, 1971, p. 136). The fact that certain

new products are readily adopted, while others never achieve initial

penetration into society, demonstrates that innovations most definitely

vary among themselves.

In order to account for such variation, researchers must first

specify the attributes upon which such innovation variation occurs.

Furthermore, the specification of these attributes (or dimensions) must

also be based on the perceptions of the potential adopter, and not based

in the perceptions of the researcher. This adopter centered approach to

innovation perception recognizes that new product acceptance is deter- « mined by what consumers perceive the product to be, which in fact may

be quite the opposite of how researchers or product manufacturers per­

ceive the new product. Rogers and Shoemaker (1971, p. 137) underscore

this point by stating: "It is the attributes of a new product, not as

seen by experts but as perceived by the potential adopters, that really

matters." 44

Innovation Attribute Categories

Rogers and Shoemaker have concluded from their literature review that there exist five dominant innovation attributes that can be used to classify the perceived attributes of innovations in universal terms.

According to the authors, this classification scheme has been developed in an effort to achieve "maximum generality and succinctness" so that one would not have to study each new innovation in order to predict its rate of adoption (p. 137). Of course, although these dominant innova­ tion attributes may vary according to an individual’s perception of a specific innovation, Rogers and Shoemaker proposed that these attributes help explain such variation. This classification scheme does not reduce

the importance of an objective, adopter centered approach to innovation evaluation, but only serves as an empirically based set of perceptual

criteria from which such evaluation can take place.

Although the reader is encouraged to review Rogers and Shoemaker's

(1971, pp. 137-164) original description of the five innovation attri­

butes, a brief overview is provided for this literature review. The

attributes have been termed, relative advantage, compatibility, com­

plexity, trialability, and observability. Relative advantage is defined

as "the degree to which an innovation is perceived as being better than

the idea it supercedes" (p. 138). This perception can be based on either

economic or social factors depending upon the nature of the innovation

or what is specifically important to the adopter (p. 138).

Compatibility refers to "the degree to which an innovation is per­

ceived as consistent with the existing values, past experiences, and

needs of the receivers" (p. 145). A compatible innovation is therefore 45 one which is perceived as involving little change on the part of the adopter (p. 147). Complexity has been defined as "the degree to which an innovation is perceived as relatively difficult to understand and use" (p. 154). This attribute represents the simplicity-complexity continuum for innovations.

Trialability is "the degree to which an innovation may be experi­ mented with on a limited basis" (p. 155). Thus, those innovations that have the capacity to be used on a limited trial basis are thought of as possessing a trialability attribute. Finally, observability

"is the degree to which the results of an innovation are visible to others" (p. 155). Rogers and Shoemaker expand this definition so

that it also refers to how easy or difficult it is to communicate

the innovation to others (pp. 155-156). Thus the materialistic and

conceptual simplicity of the innovation would relate to its ability

to be observed by, and communicated to, other potential adopters.

There is some available research evidence that documents the

potential effect these attributes have on adoption. Rogers and Stanfield

(1968) have compiled a summary table of these relations (see Table 2)

from the agricultural literature and the following generalizations have

been put forth by Rogers and Shoemaker (1971) : the relative advantage,

compatibility, trialability, and observability of an innovation as per­

ceived by members of a social system, are related positively to its

rate of adoption. The complexity of an innovation is negatively related

to its rate of adoption. Table 2 indicates that the adoption relation­

ships for relative advantage (66 studies), and compatibility (50 46

TABLE 2

SUMMARY TABLE OF INNOVATION ATTRIBUTES

AND THEIR RELATIONSHIP TO INNOVATIVE BEHAVIOR*

Perceived Relationships (Percent) Total Number Attribute of ------of General- the Innovation Positive None Negative Conditional izations

Relative Advantage 79 15 3 3 66

Compatability 86 14 0 0 50

Complexity 19 37 44 0 16

Trialability 43 43 14 0 14

Ob servability 75 25 0 0 8

* Based on Rogers and Stanfield (1968, p. 243). 47 studies), have received the most research attention. Only eight studies have examined observability, but a positive relationship is indicated.

Both complexity and trialability do not show clear cut relationships, although they do exhibit the tendencies predicted by Rogers and Shoe­ maker (1971).

Rogers and Shoemaker (1971, pp. 140-147) indicate that the relevant studies which have examined the perceptual attributes of an innovation have been successful in explaining 49 percent to 87 percent of the vari­ ance in the rate of adoption. The number of attributes measured in

these studies varies from two to fifteen which also indicates that certain attributes were either added or droprad from the Rogers and

Shoemaker classification scheme. Further investigation also reveals

that the methodology for obtaining the perceptions of potential adopters

seems to vary from one study to the next. For example, Kivlin (1960)

used a panel of "dairy experts" to generate perceptual data for 229

potential agricultural adopters, whereas Fliegel and Kivlin (1966)

employed a split sample technique so that their respondents could react

to a total of 33 different farming innovations. The variation found in

the number and nature of the innovation attributes studied, in addition

to the multiple methodologies employed by these researchers, somewhat

limits the generalizability of these findings.

The Perceptions of an Innovation

More applicable research, however, has recently been completed in

the fields of communication and marketing. Although these studies are

relatively small in number, they do make an attempt to directly address

the issue of whether the perceptual attributes of an innovation can help 48 explain and predict innovativeness for a new product or service. Further­ more, these studies provide evidence which indicates that the initial perceptions of an innovation are the most important factor in the deter­ mination of whether an individual will either adopt or reject an innova­ tion. Ostlund (1969a, 1969b, 1972, 1973, 1974) has been primarily respon­ sible for the seminal work in this area.

In his earlier work, Ostlund (1969a, 1969b) obtained perceptual data along six attribute dimensions (the five innovation attributes defined by Rogers and Shoemaker, plus product risk) from 605 housewives. Each respondent completed a questionnaire which asked them to rate six new consumer products according to the six product perception dimensions.

In addition to the perceptual data, Ostlund obtained from each respondent measures of saven predispositional factors, plus standard demographic measures of age, education, income, and occupation. The predispositional measures included the standard personality/life style measures that have been used in the past to profile the attributes of potential adopting units. Ostlund (1969b) defines these factors as follows:

1. Venturesomeness - willingness to take risks in buying new products.

2. Cosmopolitanism - degree of orientation beyond a par­ ticular social system.

3. Social integration - extent of social participation with other members of the community.

4. Social mobility - upward movement in social class.

5. Privilegedness - perceived well-being relative to peers.

• 6. Interest polymorphism - variety and extent of one's personal interests.

7. General self-confidence (or self-esteem) with respect to (a) problem solving and (b) psychosocial consideration. 49

When a step-wise multiple regression was applied to these measures, with a measure of purchase probability as the dependent measure of new product adoption, an average explained variance (R ) of .51, with a range of .42 to .59, was found for the six products. However, of more importance to this research is the finding that the perceptual variables accounted for about 90 percent of the explained variance and were also found to be the top six rank ordered variables of entry in the step-wise regression. Ostlund (1969b) concluded that the product perception vari­ ables were found to have greater predictive power when related to pur­ chase probability than predispositional or demographic variables.

Ostlund (1969b) points out that purchase probability is by no means a perfect measure of innovative behavior. This may account for the mildly respectable coefficient of determination which reached only a mean value of .51. Naturally, a more precise measure of innovative be­ havior would be an analysis of actual purchasing behavior for this group of respondents. As a follow up to the 1969 study, Ostlund (1972) ob­

tained actual purchase behavior (for one product) from 360 of the 605 original housewives. A control sample of 200 housewives was also

employed to measure the potential contaminating effect of the first

study on the buying behavior of the 360 respondents. No significant

difference was found between the mean purchase incidence score for the

control and test samples.

By splitting the sample into two groups of 100 (sample 1) and

260 (sample 2), a multiple discriminant analysis function obtained from

sample 1 was used to classify respondents in sample 2 as buyers or non­

buyers. The discriminant function correctly identified 65 percent of 50 the innovators on an unaided product purchase basis, and 74 percent of the innovators on an aided product purchase basis. Subsequent analysis by Ostlund also indicated that when only the perceptual variables were used as predictors, only a two percent drop occurs in the proportion of buyers correctly classified. Once again, the perceptual variables were shown to have superior explanatory and predictive power in the specifi­ cation of innovativeness for a new product.

In 1973, Ostlund reported a study where he factor analyzed these six perceptual measures in an effort to potentially reduce the number of dimensions regarding innovator perceptions without a major loss in information. The perceptual data for six new products from the 1969 study of 605 housewives was first submitted to a varimax factor analysis for initial analysis. Results indicated that for each of the six prod­ ucts, a two or three factor solution was found to be most meaningful.

Although the factor loadings and perceptual dimensions varied among these factors according to the product in question, two perceptual dimen­ sions consistently loaded higher on a singular factor. These dimensions were relative advantage and compatibility. Ostlund suggests (p. 107) that these perceptual dimensions may need further specificity so that respondents could provide more precise perceptual evaluations. Ostlund

(p. 103) also points out that a considerable redundancy was found among

the six perceptual variables for the six products; however, this redund­ ancy was not parallel across all six products. This suggests that a reduction of these perceptual dimensions is most inappropriate, and that more effort should be directed towards an elaboration of the perceptual variables. 51

An expanded measure of innovation perception was therefore used in a subsequent analysis by Ostlund (1974). In this study, relative advan­ tage was expanded to include specific measures of time savings, effort savings, and the perceived monetary value of the innovation. Com­ patibility was expanded to include measures regarding self-concept, family, and existing habits. Using the same multiple discriminant pro­ cedures outlined in the 1972 study, Ostlund attempted to predict the actual purchasing behavior of 200 respondents based on their responses to the expanded six attribute perceptual measure. Ostlund also included the similar measures of demographic and predispositional variables in order to test the relative effectiveness of the perceptual measures.

Using all the predictor variables, 79 percent of the innovators were correctly classified. As was found in the 1969 and 1972 studies, personal variables were of little importance since only a small change occurred in the classification matrix when the personal variables were dropped (only one additional innovator was misclassified). Effort sav­ ings, time savings, and perceived risk were found to be the most important variables, followed by complexity and monetary value. Compatibility was not found to be as important as was true in previous studies

(although it still was among the top ten most important predictors).

Ostlund (1974, p. 28) concluded that perceptions of innovations by poten­ tial adopters can be more effective predictors of innovativeness than personal characteristic variables. The author also suggested that replication among differing types of adopters and products is needed so that perceived innovation attribute dimensions can be refined and elaborated. 52

McCain and Wall (1977) tested the relationship between innovation attribute perceptions and innovativeness for members of a social system which were defined by their level of modernity. Modernity in this case referred to the educational level and stability of the social system's members (p. 2). The authors hypothesized that the modernity of social systems would influence the saliency of an innovation’s perceived attributes, and that differences would be found between the perceptions of adopters and non-adopters.

For this study, innovation perceptions (the innovation was a school bond proposal) were elicited from an open ended question that asked respondents for the three major reasons why the school bond proposal failed. Relative advantage, compatibility, and complexity were the three major perceptual dimensions used for content analytic procedures.

Results indicated that the attributes of an innovation are perceived differently according to the perceiver's membership in a social system

(p. 14). Specifically, relative advantage was more salient to members of a modern social system and compatibility was found to be more salient

to those in a traditional social system. Although a chi square analysis also indicated that gross perceptual differences would be found between adopters and non-adopters, a post hoc analysis of these differences by

perceptual dimension indicated a significant difference for the compati­

bility and complexity dimensions only. The relative advantage attribute was therefore mentioned equally by both adopters and non-adopters. The

authors also point out that an analysis of the compatibility dimensions

according to its subdivisions of trust and need/value, resulted in the

need/value attribute accounting for the preponderance of variance 53 between the two groups. McCain and Wall conclude that this post hoc analysis of perceptual subdivisions suggests a need for further refine­ ment and development of perceptual attribute categories.

Perceived Risk

One attempt to expand upon the original perceptual attribute cate­ gories developed by Rogers and Shoemaker (1971) has been the inclusion of a perceived risk attribute in the studies completed by Ostlund (1969a,

1969b, 1974). Based on the conceptualization offered by Bauer (1960),

Ostlund defined perceived risk as "the degree to which risks are per­ ceived as associated with the innovation" (1972, p. 26) and hypothesized that this perceptual attribute would be negatively related to innovative­ ness. Results indicated that the correlation coefficients for perceived risk and purchase probability were negative and significant (p 4. .01)

(see Ostlund, 1968b). Perceived risk was also found to be ranked as the third most important variable in the 1968a and 1968b stepwise regression analysis, and ranked fourth in the 1972 analysis. These results indicate that perceived risk should be considered as an additional perceptual attribute in the study of innovativeness.

Bauer (1960) originally suggested that there are two types of risk: psychosocial risk and performance risk. Psychosocial risk refers to the reaction of significant others when an individual purchases a new prod­ uct, and performance risk refers to product performance reliability. It should be noted that Ostlund (1969a, 1969b, 1972) included these two subdimensions in his analysis. Zikmond and Scott (1977) found in a factor analysis of approximately 50 risk components associated with each of three varied products, that the risk dimensions of product reliability 54 and the reaction of significant others, consistently emerged as the pri­ mary factors for all three products. Although these two risk dimensions remained consistent across the three products, there was little consist­ ency found among the remaining risk dimensions. Zaltman and Scott con­ cluded that perceived risk was a multidimensional attribute that varies according to product characteristics. An understanding of consumers' perceptions of product characteristics is suggested as a potential means of understanding the nature of perceived risk. In terms of innovation research, such an understanding can also help explain adoption behavior.

A Brief Summary

A microanalytic approach to social change has been proposed which primarily views two elements of the overall social change model. The

first element is an analysis of the adopting unit. The vast majority of

the research in this area has attempted to characterize those people who first adopt a new product (innovators) by standard sociodemographic,

personality, and life style measures. Research results have suggested

that these findings are generalizable only to similar innovations within

a given product class. Although the innovator profiles obtained through

personality and life style variables may be generalizable only to those

products within a similar product class, a description of the videotex

innovator, or early videotex adopter would be extremely valuable for

marketing, communication, and social purposes. Profiles of potential

videotex adopters do not exist in the scholarly literature.

The second element of the microanalytic social change model deals

with the attributes of the innovation according to how potential adopters

perceive them. A small, but convincing group of research studies have 55 demonstrated that the perceptions of an innovation have greater explana­ tory and predictive power than the usual predispositional measures asso­ ciated with innovativeness. Since these perceptual dimensions have been shown to have a significant effect on the adoption decision, future dif­ fusion research should continue to examine these innovation perceptions within varying contexts and situations. Again, although a small group of studies have tested the ability of these perceptions to discriminate between adopters and non-adopters of a new product, the relative dis­ criminatory strength and importance of these perceptions in terms of their ability to identify potential adopters and non-adopters of a videotex technology has not been assessed.

Research Questions

In accordance with the research findings presented thus far, the following research questions were examined by this research:

I. The Perceptions of Channel 2000

1. What are the perceptions of Channel 2000?

2. What are the differences between the perceptions of

Channel 2000 at three points in time?

3. What are the relationships between the perceptions of

Channel 2000 at three different points in time?

4. What are the relationships between innovation perceptions,

life style, and consumer creativity?

II. The Potential Adoption of Channel 2000

5. How well do innovation perceptions, life style, consumer

creativity, and sociodemographic variables discriminate

between potential adopters and non-adopters of Channel 2000 at three points in time?

6. What are the relative importance of innovation percep­

tions, life style, consumer creativity, and sociodemo­

graphic variables in discriminating between adopters and

non-adopters of Channel 2000 at three points in time?

As a by-product of these main research questions, three dependent measures of potential Channel 2000 adoption (perceived use, purchase probability, and price sensitivity) were also evaluated in terms of their overall reliability and construct validity. CHAPTER III

THE CHANNEL 2000 SYSTEM

A great deal of discussion which surrounds the videotex phenomenon deals with the specifics of hardware design and the particular services

that could be hypothetically delivered to a group of users. Each video­

tex system design, and package of videotex services, offers a unique

set of product benefits and innovation perceptions to a group of poten­

tial adopters. Chapter III primarily outlines the mechanical operations

of the Channel 2000 system. A description of the Channel 2000 services

that were made available to the test participants is also provided in

this chapter. Those readers who are interested in obtaining more

detailed information on the hardware design of the Channel 2000 system

are referred to the report by Harnish (1981).

Channel 2000 Hardware

For the purposes of this research, Channel 2000 has been defined

as a viewdata system. Although the concept of a viewdata technology is

not unique to Channel 2000, the Channel 2000 hardware and services have

certain specifically unique features that help distinguish Channel 2000

from other viewdata systems. The reader is reminded that any videotex

system should be defined not only by its mechanical operations and its

means of data transmission, but it should also be defined according to

the services it offers, and how these services are made available to the

subscriber.

57 58

Channel 2000 was conceived and developed by OCLC, Inc. OCLC (On­ line Computer Library Center) functions as an online computer network for over 2,400 library systems in 50 states, Canada, and other countries.

Libraries use the OCLC online system to catalog books, order custom- printed catalog cards, maintain location of information about library materials, and arrange for interlibrary lending of materials. The

Channel 2000 project was centralized in OCLC's Research Department as part of the Home Delivery of Library Services Research Program. This program was developed in 1978 to explore the potentiality of delivering library and information related materials, electronically, into patrons' homes.

The essential hardware elements of Channel 2000 consisted of a tele­ phone adapter (or decoder), and a 16-key keypad. The only manditory system requirement from a user perspective was the possession of a standard telephone and television set. Channel 2000 was not a techno­ logical advancement in terms of hardware, but instead represented a new configuration of familiar components: modem, processor, keypad, and video display generator (see Bendig, 1980). A photograph of the standard

Channel 2000 hardware configuration within a home setting is provided in

Figure 4. It should be pointed out that no modifications to the tele­ phone or television receiver were required when, a Channel 2000 decoder i was brought into a user's home. For the sake of convenience, however, most users either moved a portable television set within close proximity

of the telephone, or supplimented their telephone with an extension cord.

As seen in Figures 4 and 5, the Channel 2000 decoder had a telephone

coupler mounted on its top surface. The coupler was designed to 59

Figure 4. The Channel 2000 System. 60

. Vllllfll . ■ : HONCf tANKIKf SftVtCCS . . . A v t f r l ■ I ACCCUM IKfOKMtlOH IfttWCfS f » AhK STATCtlCHt - C 0fttt«V' KMTN * IANK STATf M K T - L *tr XDJIT* *. CURREHT iHItfttSf lMTt* . ■ > t u n w iu w I ENttt A Ifu ' K N TC* A IIHI NVMIEft *H) > » 1 » •to I V • tuts . IS t K f 1E61HHIK6 8I*M.*rSt -TKC HaM .I« t> « .m v ic it.

Figure 5. The Channel 2000 Keypad. 61 accommodate standard (rounded mouthpiece and receiver ends) telephones only, which precluded the use of Princess, Trimline, or other exotic telephone handsets.

The Channel 2000 Keypad

The Channel 2000 keypad contained ten numeric and six special func­ tion keys and was designed to be comfortable and easy to use. A 25-foot cable was attached to the keypad so that users could interact with

Channel 2000 in a standard TV viewing position. A summary of available commands and keystrokes was printed on the back of the keypad for user convenience. The six special function keys were labeled with short words to enable the user to perform specific commands in a relatively simple manner. An 0CLC report (see Harnish, 1981) explained these functions:

One key - "OOPS" - was provided for wiping out mistakes during key pressing, and for requesting help (by pressing "OOPS" twice). Another key - "DO IT" - was used to con­ clude each request.

The remaining keys - digits 0-9 and the four keys "GET," "JUMP," "FWD," and "BACK." - were used in combinations to specify all the different ways of going from one screen of information to some other screen. The most important way of getting a new screen was to select an item on an index screen by keying the number of the item (followed, as always, by "DO IT"). One could also return to an index by keying "JUMP" (followed by "DO IT"). Or, to go to the first screen of a major section of the Channel 2000 information, one could refer to a printed list of the sections and their identifying numbers provided in the User Manual; then, if the number listed were 9 (as an example), the request "GET 9" would immediately display the desired screen. To undo such a request - and to return to normal activity after getting help - one would key "GET BACK."

To see items that were adjacent or nearby in an index (that is, in another branch of the tree), it was not neces­ sary to return to the preceding index. One could "browse," say, 4 items forward or back by keying "JUMP FWD 4" or "JUMP BACK 4." Simply "JUMP FWD" or "JUMP BACK" (without the number) would move one item forward or back. 62

Since many items filled more than one screen, the commands "FWD" and "BACK" were available to get to the next or the previous screen of an item. "FWD 3" or "BACK 3" would skip forward or back three screens of an item. One could skip any number of screens at a time.

Besides going from one screen to another, the Channel 2000 user also did other things by means of the keypad. To re­ quest a book from the library, one would key "GET" when viewing the "videocard" describing the item of interest. To enter a personal identification number, an account number, or a dollar amount in the banking service, one would key the digits (followed by "DO IT").

Sometimes the Channel 2000 system would ask for confirmation of a request. Then the only acceptable answers were YES and NO. These labels could have been put on any two keys; they were placed on the "FWD" and "BACK" keys.

A standard interactive session with Channel 2000 involved the fol­ lowing sequential events. The television would first be turned on and

tuned to Channel 2. The Channel 2000 decoder would then be switched

from TV to Channel 2000. This switching procedure activated a decoder­

generated "sign-on" procedure screen which displayed the computer tele­ phone number, and telephone coupling procedures for Channel 2000 inter­

action. Once this coupling had taken place, users were automatically

logged-on to the OCLC/Channel 2000 host computer. The first screen

displayed on the Channel 2000 system consisted of a personal user list

which contained the first names of each member of the household, plus a

guest identification number. After the user had identified him/herself,

the Channel 2000 main index appeared. Information in the Channel'2000

system was located by selecting items from sequential screens arranged

in a hierarchial index structure or "menu tree." To conclude an inter­

active session, users simply disconnected the telephone from the Channel

2000 decoder. 63

Channel 2000 Services

The actual Channel 2000 services that were offered to test partici­ pants, to a large extent, defined the nature and scope of Channel 2000 as a viewdata system.

A viewdata service can assume many varied forms. Channel 2000 represented a collection of information based services representative of those which might be offered over a library-based information system

(the exception of course was the banking services provided by Bank One of Columbus, Ohio). A brief listing of Channel 2000 services is pro­ vided in Table 3. A more detailed description of these services follows.

Library Catalog (Video Catalog)

The Video Catalog was essentially a computerized version of the

Public Library of Columbus and Franklin County’s card catalog. Users were able to search the catalog by author, title, or subject, for those books and materials owned by the public libraries and their branches.

Once the user had accessed any one of the 300,000 records (cards), the book item could be requested directly through the Channel 2000 system.

The library would complete the user initiated transaction by sending the requested item via the mail, complete with return postage and mailing envelope.

Video Encyclopedia

The Videopedia consisted of the complete 21 volume text of Arete’s

Academic American Encyclopedia published in paper form in 1980. The

Videopedia itself consisted of approximately 32,000 articles which

varied in length from a single screen to over three hundred screens of

text. The graphical portions of the encyclopedia were not input to the TABLE 3 WELCOME TO

CHWNEL

Channel 2000 offers eight exciting services that you can use right on your TV set in your home. WELCOME TO THE VIDEO CATALOG Video Catalog Browse through the videocard catalog of the Public Libraries of PREPARED BY THE PUBLIC LIBRARIES Columbus and Franklin County, and select books to be mailed OF COLUMBUS AND FRANKLIN COUNTY directly to your home. HOW DO YOU WANT TO SEARCH FOR YOUR BOOK? Video Encyclopedia 1 BY TITLE Locate any of 32,000 articles in the new Academic American Em 2 BY AUTHOR cyclopedia via one of three easy look-up indexes. 3 BY SUBJECT Home Banking Pay your bills; check the status of your checking and savings ac­ counts; look up the balance of your VISA credit card; look up your mortgage and installment loans; get current information on BANK ONE interest rates. Public Information Become aware of public and legislative information in Ohio.

Columbus Regional Information VIDEO CATALOG Check the monthly calendar of events for local educational and entertainment happenings. TtTLE 1 THE THIRD WAVE Math That Counts! AUTHOR 2 ALVIN TOFFLER Teach your children basic mathematics, including counting and SUBJECT simple word problems. 3 SOCIAL HISTORY 4 CIVILIZATION, MODERN Early Reader EDITION Help your children learn to read by reinforcing word relationships. MORROW, 1 9 8 0 BRANCHES Deaf Community Bulletin Board CLINTONVILLE, MAIN Check activity announcements for and prepared by the deaf community.

“\

CHANNEL 2 0 0 0 MAIN INDEX AUTHOR 1 BOOKMARK ALVIN TOFFLER 2 VIDEO CATALOG 3 VIDEO ENCYCLOPEDIA 1 THE DEATH OF MASS MEDIA 4 PUBLIC INFORMATION 2 FUTURE SHOCK 5 COLUMBUS REGIONAL INFORMATION 3 THE THIRD WAVE b MATH THAT COUNTS! 4 THE WORLD OF THE FUTURE 7 FADl V RPAT1RR 8 DEAF COMMUNITY BULLETIN BOARD 9 HOME BANKING

A project of the 0CLC Research Department

i:’: q c u c . 1125 KINNEAR ROAD ■ COLUMBUS, OHIO 43212 • (G14I 4B6-3GG1 65

Channel 2000 software. The indices for the Videopedia consisted of:

(1) an alphabetic list of article titles, (2) a topic listing by alpha­ betic word intervals, and (3) an encyclopedic index which cited occur­ rences of a given subject.

Public Information

A Columbus area volunteer group called Com-tility, provided Channel

2000 users with information in the areas of employment, environment, human services, parks, public utilities, sports, taxes, and voting eligibility.

Community Calendar (Columbus Regional Information)

Test participants could also access a listing of local events (both educational and entertainment events) and a guide to points of interest

in Columbus. This information was compiled by the Columbus Regional

Information Service (CRIS).

Math That Counts

Math That Counts was designed to provide children with a variety

of basic mathematical exercises for developmental purposes. These exer­

cises were grouped by grade level corresponding to level of complexity

(kindergarten through eighth grade). This service was developed by the

College of Education (Early and Middle Childhood Department) at Ohio

State University.

Early Reader

The second OSU service consisted of a series of 14 single screen

stories, as well as instructions to parents on how to best use a reading

service designed for children. The 'objective of each lesson was to

teach pre-school children how to read and recognize word relationships. 66

Home Banking

The Home Banking Service was available only to those test partici­ pants who had Bank One checking accounts in the Columbus area. The service itself was developed by OCLC, although Bank One obviously pro­ vided the necessary financial transactions. Users of this service could do the following: (1) display checking and savings account balances and transaction lists for the current or previous month, (2) display current interest rates, and (3) pay bills. A three level security scheme ensured privacy of personal financial information.

These services were stored and accessed through a dedicated computer system whose nucleus was a Xerox/Honeywell Sigma 7 central processing unit. According to Harnish (1981), this system included 512 kilobytes of main memory, 1.3 gigabytes of storage on eighteen moving-head magnetic

disk drives, and 56 telephone lines operating at 30 characters per

second. Also supporting the system were a fixed-head magnetic disk,

two 1600-bytes-per-inch magnetic tape drives, a high speed line printer,

and a teletypewriter. The Channel 2000 system occupied about 90 kilogytes

of main memory. CHAPTER IV

METHODOLOGY

To a large extent, the methodology which was used during the

Channel 2000 test was predetermined by the theoretical framework estab­ lished thus far, and also by the logistical restraints inherent to conducting a prototype videotex pilot study. Chapter IV describes this research design and outlines the major characteristics of this initial videotex research project.

The independent and dependent variables for this research effort are also described and documented in this chapter. For those readers who are interested in determining how the test participants were selected, a detailed description of the sampling procedures and com­ pliance percentages are included. A description of the Channel 2000 training sessions, equipment pick-up and questionnaire administration procedures are also discussed within this chapter.

Finally, the reader is provided with a description of the data analytic techniques that were used to address the previously listed research questions. This last section serves as an introduction to the actual research findings which are presented in Chapter V.

General Research Design

The research design specified by this dissertation should be viewed as only a portion of the total Channel 2000 research process.

In addition to a review of several multi-client videotex research projects, and additional consultation with industry leaders, the OCLC

67 68

Research Staff conducted a series of six focus group interviews at the conclusion of the Channel 2000 test. Although these informational sources were not documented or incorporated into this dissertation's research methodology, the knowledge gained through those techniques did in fact influence many of the preliminary decisions associated with the

Channel 2000 test. Although the proprietary nature of this information prohibits its widespread distribution, the reader is reminded that some of the implicit decisions associated with the Channel 2000 test were partially based upon the information provided through these sources.

The Channel 2000 test was designed so that the perceptions of

Channel 2000, and those user characteristics that could be theoreti­

cally linked to the potential adoption and non-adoption of Channel 2000,

could be examined at three points in time. This means that unlike most other cross-sectional innovation studies, where only a singular measure­ ment is taken, this research made use of three successive questionnaire

instruments.

The measurement instrument for this test consisted of a series of

three self-administered mail questionnaires, which were completed by all

Channel 2000 participants 16 years of age or older. The test itself

ran from October 1, 1980 to December 31, 1980. The three month time

frame allocated to the Channel 2000 test was predicated by the avail­

ability of an OCLC computer. Ideally the test could have been designed

around a six month time frame; however, the OCLC computer was made

available to the Channel 2000 research project for a three month period

only. Therefore, the first questionnaire (Time I) was completed approx­

imately two weeks prior to the October 1 startup, the second questionnaire 69

(Time II) was completed approximately three weeks after October 1, and questionnaire three (Time III) was completed one week past the conclusion of the Channel 2000 test. Table 4 lists the mailout dates for Time I-III.

Instrumentation - Independent Variables

The perceptions of Channel 2000 were organized and defined accord­

ing to the five innovation attributes found in the diffusion literature as specified by Rogers and Shoemaker (1971). These attributes are rela­

tive advantage, complexity, compatibility, trialability, and observa­ bility.

As a follow up to the work of Bauer (1960) and Ostlund (1969a,

1969b, 1974), an additional innovation attribute called perceived risk was added to the original Rogers and Shoemaker list. Table 5 lists the

six perceptual attribute categories with their corresponding definitions.

Under each attribute category, the questionnaire items that were used to

elicit perceptual response are also listed. The vast majority of these

items are a duplication of Ostlund1s (1969a, 1969b) original question­

naire. Additional perceptual items were generated and added to final

instrument in an effort to better represent the multidimensional nature

of each perceptual attribute. In addition, these extra items can be

thought of as being more "Channel 2000 specific" in comparison to the

generic perceptual items that Ostlund provided.

The perceptual attribute measurement instrument consisted of 36

questionnaire items which were evaluated on a six point Likert scale

ranging from definitely disagree to definitely agree. Respondents

indicated their agreement or disagreement with each perceptual item by

circling a number on the questionnaire form. These 36 perceptual items 70

TABLE 4

QUESTIONNAIRE MAILOUT DATES FOR TIMES I, II, AND III

Item Date

Time I September 11, 1980

Channel 2000 Test Begins October 1, 1980

Time II October 24, 1980

Channel 2000 Test Ends December 31, 1980

Time III January 3, 1981 71

TABLE 5

PERCEPTUAL ATTRIBUTES AND CORRESPONDING QUESTIONNAIRE ITEMS

Relative Advantage— the degree to which an innovation is perceived as being better than the idea it supercedes (both economic and non­ economic considerations)

*1. Channel 2000 would be worth every cent it costs when people can eventually subscribe to it. economic Channel 2000 would probably cost a lot of money.

*3. Channel 2000 offers me real advantages over the way I used to do things.

4. Channel 2000 would be very useful non- J economic ^ Channel 2000 would help me save time paying bills.

6. Channel 2000 would hlep me save time getting books from the library.

Complexity— the degree to which an innovation is perceived as being relatively difficult to understand and use.

*1. I have no difficulty understanding how Channel 2000 technically works.

2. It would be harder to pay bills using Channel 2000 than it is for me now.

*3. I would need a lot of technical know-how to use Channel 2000.

*4. I understand how to operate Channel 2000.

5. I don't think Channel 2000 would make my banking any easier to do. 72

TABLE 5 (Continued)

PERCEPTUAL ATTRIBUTES AND CORRESPONDING QUESTIONNAIRE ITEMS

Compatibility— the degree to which an innovation is perceived as being consistent with existing values, past experiences, and needs of the receivers (potential adopters).

1. I think Channel 2000 would fit right into the way I choose to live my life.

2. Channel 2000 would probably cut down the amount that we talk to each other in my family.

*3. Channel 2000 is not a product intended for a person like me. norms/ values *4. My family might object if I subscribed to Channel 2000.

5. The public libraries should not spend tax dollars on services like Channel 2000.

6. Channel 2000 would limit my personal privacy.

7. I think Channel 2000 would be boring to use.

8. I would much rather go to the library than order books through Channel 2000. past 9. In the long run, even though I could/can get library experience books through Channel 2000, I would probably go to the library as often as I used to.

10. In the long run, even though I could/can pay bills over Channel 2000, I would probably continue to pay my bills the way I used to.

*11. I would have to change some of the ways I do things if I used Channel 2000.

12. To me Channel 2000 seems to be a luxury item,

needs 13. I really need a service like Channel 2000.

14. I expect Channel 2000 will become a real necessity. 73

TABLE 5 (Continued)

PERCEPTUAL ATTRIBUTES AND CORRESPONDING QUESTIONNAIRE ITEMS

Trialability— the degree to which an innovation is perceived as available for trial on a limited basis, without a large commitment.

*1. I really wouldn't lose much if I subscribed to Channel 2000 and it didn't work out as I thought.

*2. I wouldn’t have to spend very much money to subscribe to Channel 2000 in order to try it out.

3. I would probably be permitted to subscribe to Channel 2000 on a trial basis for a month or so.

Observability— the degree to which results of an innovation will be apparent and possible to communicate to others (sometimes called communicability).

1. Channel 2000 would probably give my family something to talk about.

2. I have heard about or seen devices like Channel 2000 before.

*3. Channel 2000 is not the sort of product that I would tell my friends about after I subscribe to it.

*4. I think if I subscribe and use Channel 2000, I would have no difficulty in telling my friends what it is like.

Risk— the degree to which risks are perceived as associated with the innovation.

*1 . If I subscribe to Channel 2000 it is likely that I would not be satisfied with the way it performs. product performance *2. Evaluating how Channel 2000 performs is not something that I am very confident in doing. 74

TABLE 5 (Continued)

PERCEPTUAL ATTRIBUTES AND CORRESPONDING QUESTIONNAIRE ITEMS

*3. It is very difficult for me to evaluate how my friends . , , will react to my subscribing to Channel 2000. psychosocial *4. I think that if I subscribe to Channel 2000 most of my friends would approve.

* These items were adapted from Ostlund (1969a, 1969b, 1972, 1973). were evaluated at three different points in time (Times I, II, and III).

Thus each respondent evaluated each of the 36 perceptual items a total

of three times. The wording of these items was slightly changed from

Time I to Times II and III in order to represent the appropriate changes

in verb tense. The presentation order of the individual iltems was also

randomized to minimize ordering effects. See Appendix C for the respond­

ent instructions and questionnaire formating.

It should be noted from an examination of Table 5 that the attribute

categories of relative advantage and compatibility received the most

attention In terms of the mere number of questionnaire items that were

generated. This is in response to the suggestions by Ostlund (1973) and

McCain and Wall (1977) who recommended an expansion of these two con­

cepts. Specifically, those items representative of the relative advan­

tage dimension were broadened to include items relating to savings in

time (Channel 2000 would help me save time getting books from the li­ brary) , monetary value (Channel 2000 would probably cost a lot of money), and usefulness (Channel 2000 would be very useful). The compatibility dimension was broadened to include items relating to the individual's life style (I think Channel 2000 would fit right into the way I choose

to live my life), existing habits (I would much rather go to the library than order books through Channel 2000), and needs (To me, Channel 2000 seems to be a luxury item).

The second independent variable construct employed was Leavitt and Walton's (1977) creativity scale. The 30 item creativity instru­ ment developed by Leavitt and Walton was administered at Time II to the Channel 2000 participants to assess 76 the degree of creativity among the test population. Respondents were asked to express the extent to which each statement fit in with their feelings or views along a 5 point Likert type scale. The scale values ranged from "extremely well" to "not well at all." In addition to the items designed to tap creativity, Leavitt and Walton included in the 30 item instrument, several social desirability measures to control for response bias. Although a complete listing of all 30 creativity and social desirability questionnaire items can be found in Appendix C

(see Questionnaire II, Section B), three example items are presented below:

I like to experiment with new ways of doing things.

When I see a new brand on the shelf, I often buy it just to see what it is like.

If I got an idea, I would give a lot of weight to what others think of it (social desirability item). v Each item in the creativity scale was assigned a value by Leavitt and Walton so that the items could be summed, and a singular creativity score could be assigned to each individual.

The third independent variable construct that was used in this re­ search was a series of life style measures which were administered during

Time I of the test. It has already been mentioned that life style vari­ ables have in the past been used to describe and differentiate adopters and non-adopters for new product concepts (see Darden and Reynolds, 1974;

Boone, 1970; Robertson and Kennedy, 1968; and Arndt, 1969). However,

there currently exists no singular life style measurement instrument

that can be used across a variety of research situations. 77

Therefore, five essential life style dimensions were developed for this research. These dimensions were based on certain hypothetical rela­ tionships between life style and innovativeness which have been suggested by previous research, and in light of a generalized understanding of the potential benefits and uses of Channel 2000. As was the case with the perceptual measures, respondents indicated their agreement or disagree­ ment with all life style variables according to a six point Likert scale (see Appendix C, Questionnaire I, Section A).

The first life style dimension was opinion leadership. Rogers and

Shoemaker (1971, p. 189) state that earlier adopters have a higher degree of opinion leadership than late adopters. Five questionnaire items designed to measure opinion leadership were included in the questionnaire

(see Table 6 for a complete listing of life style items by their respec­ tive dimension).

The second life style dimension was related to degree to which an individual viewed him/herself as being conservative or traditional in values, beliefs, and behavior. Rogers and Shoemaker (1971, p. 189) sug­ gest that early adopters are more likely to have more modern than tradi­ tional norms. Six life style items designed to measure this conserva­ tive/traditional dimension were also included (see Table 6).

It was hypothesized that a major product benefit associated with

Channel 2000 would be its potential ability to save the user time. It was hypothesized that those respondents who feel that time is a commodity which they have very little of, should be more favorably predisposed to a time saving device such as Channel 2000. Five time related life style variables were included in the questionnaire (see Table 6). 78

TABLE 6

LIFE STYLE DIMENSIONS AND CORRESPONDING QUESTIONNAIRE ITEMS

Opinion Leadership

I have more self-confidence than most of my friends.

I don't like to take chances.

My opinions on things don't count very much.

I am usually among the first to try new products.

I like to be considered a leader.

Conservative/Traditional

I have somewhat old-fashioned tastes and habits.

I use many of the brands and products that my parents used to use.

Today, most young people don't have enough discipline.

The man should be the boss in the house.

I am in favor of very strict enforcement of all laws.

Everything is changing too fast.

Time Related

I find that I have less time to do the things I want to do than I used to.

I would go to the library more often if I could find the time.

I wish I didn't have to spend so much time waiting in lines.

I always find plenty of time available to get all my necessary reading done.

Sometimes I feel that there is so much news available that I can't keep up with it. 79

TABLE 6 (Continued)

LIFE STYLE DIMENSIONS AND CORRESPONDING QUESTIONNAIRE ITEMS

Information/Media Related

To keep up with the news, I primarily rely on television.

Television is my primary form of entertainment.

I enjoy science fiction.

I usually read the sports page in the daily newspaper.

Information from advertising helps me make better buying decisions.

Magazines are more interesting than television.

I don’t like television news programs.

Getting information by TV won't ever take the place of reading a book or newspaper article. 80

In that Channel 2000 is primarily an informational videotex service, eight information and media related life style variables were also in­ cluded to assess primary sources of information, media preferences, and attitudes towards the information provided by the media in general (see

Table 6). It might be hypothesized that those individuals who have a print oriented life style would be those most interested in a print oriented videotex service.

Finally a series of fifteen items designed to measure how often an

individual participated in a given event were also included as a part of

the life style instrument. These items were developed to measure those activities that could be directly related to the use of an information based, electronic, computer device (i.e., frequency of book reading,

computer operation, library attendance, etc.). A complete listing of

these activities can be found in Appendix C, Questionnaire I, Section C.

Respondents indicated how often they participated in each of the fifteen

activities during the previous 12 month time period.

The last set of independent variables that were included in the

Channel 2000 test consisted of a series of demographic measures. These

included age, marital status, sex, education, length of residency, and

employment status. All of these demographic variables were asked during

Time I of the test (see Appendix C, Questionnaire I, Section F). Thus,

the independent variables that were used in this research included

measures of innovation perceptions, creativity, life style, and demo­

graphic indicators. Table 7 lists these independent variables accord­

ing to when they were administered to the respondents. 81

TABLE 7

QUESTIONNAIRE VARIABLES BY DESIGN PHASE

Design Phase

Variable I II III

Independent Variables

Innovation perceptions X X

Creativity scale X

Life style items X

Demographics X

Dependent Variables

Perceived use X

Purchase probability X X X

Price sensitivity X X X (Guttman) 82

Instrumentation - Dependent Variables

The most apparent criterion variable for this research was the user's decision to either adopt or reject the Channel 2000 system.

Rogers and Shoemaker (1971, p. 26) define adoption as "a decision to make full use of new idea as the best course of action available." In the area of consumer behavior, adoption is usually operationalized as the purchase of a new product or service. Robertson (1971, p. 56) expanded upon this definition by stating that adoption refers to the acceptance and continued use of the innovation over time. Naturally the Channel

2000 respondents did not have the opportunity to adopt Channel 2000 in

the traditional sense of the word since the service was not made avail­ able to users after the December 31 cutoff date. All adoption measures

reported herein therefore were actually measures of potential adoption

and were treated as such throughout the research investigation.

The Channel 2000 test had four measures of potential adoption that

resulted from the research procedures. First, there was the actual use

of the Channel 2000 system over the three month trial period. Unfor­

tunately, at the time of this writing, OCLC staff are painstakingly

correcting this data set due to the programming malfunctions which

occurred during the test. Although a program was designed to record

and store each user transaction, the data itself were frequently mis-

classified and/or lost. An examination of the transactional file at

the conclusion of the test indicated that approximately 50 percent of

the data were in error. Therefore, this information was not included in

this analysis. 83

The second measure of Channel 2000 potential adoption was derived from the Time II questionnaire. Respondents were asked to indicate if they used each Channel 2000 service either every day, 6 times a week,

4-5 times a week, 2-3 times a week, once a week, or never (see Appendix

C, Questionnaire II, Section E). This measure of perceived Channel 2000 use was first summed, and then a mean score was assigned to each individ- ual

The third measure of potential adoption was derived from the liter­ ature in consumer behavior. Ostlund (1969a, 1969b, 1972) made use of a purchase probability scale developed by Juster (1966). This scaling technique asked respondents to indicate their subjective purchase prob­ ability along a 0-100 (11 point) scale. Ostlund found that purchase probability scores could be used as a dependent measure of potential adoption across a variety of product classes.

The purchased probability scale was administered to respondents at

Times I, II, and III. The exact wording of the scale instructions is given below.

Now think about what you currently know about Channel 2000 and how you would feel about subscribing to a service like Channel 2000. Select a number from the ones listed below that describes the chances that you would subscribe to Channel 2000 if it did come out on the market. If you are absolutely certain you would subscribe, the number you should circle is "100." If you are absolutely certain you would not subscribe, the number you bhould circle is "0." If you think the chances are somewhere in between, circle one of the other numbers that seems most appropriate. Again, we are only interested in your opinions since Channel 2000 will not be sold to you at any time during this test.

1. Please circle the one number you feel best describes your feelings at this time.

0 10 20 30 40 50 60 70 80 90 100 I absolutely would not I absolutely would subscribe subscribe to Channel 2000 to Channel 2000 The fourth measure of potential adoption consisted of five price sensitivity measures administered at Times I, II, and III. Essentially these measures asked respondents to indicate their agreement or disagree­ ment on a Likert scale ranging from 1 to 6, with the statement, "If

Channel 2000 cost ($5.00, $7.50, $10.00, $12.50, or $15.00) per month,

I would probably subscribe." Since these price ranges had equally in­ creasing cumulative intervals and were unidimensional, the five scores were used to create a Guttman scale. In this procedure, individuals were assigned a singular value based on their agreement or disagreement with each increasing price interval for the Channel 2000 services. In other words, if a user would not even pay $5.00 for the Channel 2000 services, he/she would receive a value of 0. If a user would pay $5.00 but not

$7.50, he/she would receive a value of 1 (agreement with the $10.00 price would therefore equal 3). Each individual received a singular

Guttman scale score at the three points in time. Table 7 lists all

three dependent variables according to when they were administered to

the respondents.

Sampling Procedures

The sampling procedures employed by this research were unfortun­

ately restricted by the number of Channel 2000 decoders that were made

available by OCLC. Although a total of 200 decoders were built, con­

tractual agreements made with another corporate participant (Bank One)

prespecified that OCLC would in effect have control on only 100 decoders.

The 100 Bank One decoders were therefore placed in non-randomly selected

households specified by Bank One management. Because the Bank One

household units may have contained a certain degree of sampling bias, 85

they were not considered to be representative of the larger population

group. Therefore, these households were omitted from the data analytic

procedures reported herein.

An attempt was made to place the remaining 100 OCLC decoders in

the homes of randomly selected households within the metropolitan area

of Columbus, Ohio. To accomplish this task, several procedures were

employed. First, a total of 1,200 randomly placed telephone calls were made to potential test participants living in Columbus. Each subject

responded to a short questionnaire designed to measure the subject’s

willingness to cooperate with a three month experiment (see Appendix D).

To determine cooperation, the following brief paragraph was read:

Our research involves a test of an electronic system that can be used in the home to give you library services such as ordering books or looking things up. It also provides banking services like enabling you to pay bills by a special connection with the bank.

We are interviewing people that might be interested in par­ ticipating in this test. Let me say quite clearly that this is not a sales approach. We are only trying to find families that might be interested in participating in a test of an electronic information system for the home developed by a local company called OCLC that works with libraries around the country.

The test involves using the service for three months and filling out three questionnaires for research purposes. OCLC would loan you a small device that attaches to your television set (IF NO TV, TERMINATE HERE). You would receive this service at no cost.

Are you interested in being considered for this test?

A total of 357 people (30 percent) agreed to this form of initial

cooperation. To gain additional information about these people, and

further test their willingness to cooperate with the Channel 2000 test,

a screener questionnaire was mailed to the individual who agreed to 86

cooperate during the first telephone contact. This questionnaire (see

Appendix E) was designed to provide information regarding ownership of

electronic equipment, use of automatic telling machines, opinion leader­

ship in regards to electronic products, demographic composition, and the

configuration of their telephone’s handset. The opinion leadership scale

used in this questionnaire was developed by Rogers and Cartano (1962).

The results of this seven item battery, in addition to the demographic

information reported, allowed for a more controlled selection procedure

which would ensure that both high, medium, and low "electronic opinion

leaders" would be included in the final test population.

Of the 357 screener questionnaires went out, 212 (59 percent) were

returned. Each respondent was first assigned an electronic opinion

leadership score by summing the values assigned to each opinion leader­

ship questionnaire item. Table 8 lists the opinion leadership question­

naire items, possible responses, and the values assigned to each response.

These scores could range from 7 (high electronic opinion leadership) to

0 (low electronic opinion leadership). Individuals who scored 0, 1, or

2 were assigned to the "low" opinion leadership group, scores 3, 4, or

5 represented the "medium" opinion leadership group, and scores 6 or 7

designated the "high" opinion leadership group. Table 9 lists the

population percentages associated with each of these three categories

for the 212 respondents (original sample).

Based on the percentages of the total 212 respondents, a sample of

100 participating households was selected according to a randomized

quota selection procedure. This process insured that the selected

sample initially contained a proportion of high, medium, and low opinion TABLE 8

OPINION LEADERSHIP ITEMS AND SCORE VALUES

Score Questionnaire Item Potential Response Value

1. In general, do you like to talk about 1. Yes +1 electronic products* with your friends? 2. No 0

2. Would you say that you give very little 1. You give very little information. 0 information, an average amount of informa­ tion, or a great deal of information about 2. You give an average amount of information. 0 electronic products* to your friends? 3. You give a great deal of information. +1

3. During the past six months, have you told 1. Yes +1 anyone about some electronic products?* 2. No 0

4. Compared to your circle of friends, are you 1. Less likely to be asked. 0 less likely to be asked for advice about electronic products?* 2. About as likely to be asked. 0

3. More likely to be asked. +1 TABLE 8 (Continued)

OPINION LEADERSHIP ITEMS AND SCORE VALUES

Seci j. Questionnaire Item Potential Response Value

5. If you and your friends were to discuss 1. You mainly listen to your friends' ideas. 0 electronic products* what part would you be most likely to play? Would you mainly 2. You try to convince them of your ideas. +1 listen to your friends' ideas, or would you try to convince them of your ideas?

6. Which of these happens more often? Do you 1. You tell them about electronic products. +1 tell your friends about some electronic products* or do they tell you about some 2. They tell you about electronic products. 0 electronic products?

7. Do you get the feeling that you are 1. Yes +1 generally regarded by your friends and neighbors as a good source of advice about 2. No. 0 electronic products?*

* Electronic products refer to such things as stereo equipment, calculators, televisions, TV games, or even home computers.

oo oo 89

TABLE 9

OPINION LEADERSHIP DISTRIBUTION FOR

ORIGINAL, SELECTED, AND COOPERATING SAMPLES

Opinion Leadership Category

Sample High Medium Low

Original Sample 28* 36 36 (N = 212) (N = 59) (N = 36) (N = 75)

Selected Sample 26 36 38 (N = 100) (N = 26) (N = 36) (N = 38)

Cooperating Sample 30 35 35 (N = 71) (N = 21) (N = 25) (N = 25)

* Values expressed as percentages. 90 leaders that was similar to the proportion found among the base of 212 respondents (see Table 9 for a comparison of the original sample to the selected sample).

These 100 households were then instructed in a letter to pick up their Channel 2000 decoder at a specified date and place so that they could participate in this test and be trained in the usage of Channel

2000. If any participant could not attend any of the pick up times that were made available, other arrangements were made to facilitate this process. Naturally, certain households dropped out of the test at this point. A total of 71 Channel 2000 decoders were successfully placed, and this sample population served as the final test group (the cooperat­ ing sample in Table 9) for this research. Although immediate attempts

to replace the missing sampling points were at first initiated, this process was stopped when OCLC decided to place the remaining decoders

in public institutions for demonstration purposes. Of the 144 individ- uals 16 years of age or older included in this 71 household group, 141

completed all three questionnaire instruments which resulted in a 98 percent completion rate.

An initial check of the attrition effect of this noncompliance on

opinion leadership proportions revealed that these proportions remained

relatively stable. Table 9 also lists the opinion leadership percentages

for those individuals who did pick up decoders according to the three

opinion leadership categories. Although high opinion leaders tended to

cooperate with the test to a greater extent than did the low opinion

leaders, this difference is quite small. 91

Channel 2000 Training Procedures

On one of two separate evenings, the Channel 2000 participants received some very cursory training instructions on the use of Channel

2000. The initial concept of Channel 2000 was profiled as a low cost, easy to use, self-installed, viewdata system, that becomes operational with a minimum of external assistance. These training sessions made an attempt to retain this original conceptualization. Thus, although the participating families had the opportunity to experiment with, and ask questions about, any one of six Channel 2000 units displayed, the pri­ mary objective of the meeting was for participants to pick up their decoders and sign a release form. The decoders themselves were packaged in a box complete with installation instructions and a user's manual.

Brief demonstrations of the Channel 2000 system were periodically given for interested parties, and attempts were made to resolve any uncertainty felt by the participating families.

Administration of the Questionnaires

It has already been mentioned that three questionnaires were admin­ istered to each Channel 2000 participant 16 years of age or older. Due to the small sample size available, every effort was made to obtain usable questionnaires from all the participants across the three data collection points. To accomplish this, respondents first received the mailed questionnaire with a cover letter (see Appendix F) which instructed the household on how to distribute the questionnaires, and when to return

them in the addressed, stamped envelope which was enclosed. Those who did not return a questionnaire were first sent a postcard reminder.

If this reminder was unsuccessful, respondents were contacted by 92 telephone. If this was unsuccessful, hired interviewers contacted the household members in their homes, and returned with the completed ques­

tionnaires. This rather deliberate attempt to increase response rates

for the three questionnaires resulted in a total of 141 (98 percent)

individuals who completed all three of the questionnaires.

Analytic Procedures

The large data set which resulted from the three mail questionnaires

required a series of sequential analytic procedures.

Phase I; Data Reduction

The initial data analytic procedures were designed with two objec­

tives in mind. First, although the sets of life style and activity

variables were carefully chosen according to their theoretical associa­

tion to the potential adoption process, the relationships or underlying

structure of these variables had not yet been determined. Second, the

vast number of these independent variables created a situation where the

mere number of variables exceeded the number of questionnaire respond­

ents. Therefore these variables had to be combined in some meaningful

way.

In an effort to determine the underlying structure of the 24 life

style variables and 15 activity variables, a series of principle com­

ponents factor analyses with oblique rotations were run on the two vari­

able sets. For the life style variables, a four, five, six, and seven

factor solution was requested, and a four, five, and six factor solution

was requested for the activity items.

Although it was also necessary to combine and reduce the perceptual

variables, the theoretical evidence in support of the six perceptual 93 dimensions established by Rogers and Shoemaker (1971) suggested that these underlying dimensions remain intact. Therefore, since these per­ ceptual dimensions had received a considerable amount of previous re­ search support, they were not subjected to additional factor analytic procedures.

Once the underlying dimensions of the life style, activity and perceptual variables had been established, it was then necessary to combine those variables representative of each dimension into a singular score. This was accomplished through a summation procedure where a singular mean score was assigned to each individual for all the inde­ pendent variable dimensions. If a composite scale contained any nega­ tively worded questionnaire items, these items were recoded so that any given composite scale contained a series of variables that were in agree­ ment in terms of their directionality.

Phase II; Independent and Dependent Scale Assessment

Since the perceptual variables had not been subjected to factor analysis, there was a possibility that some measurement error could be associated with certain individual variables which comprised a singular perceptual dimension. Therefore a series of coefficient alpha reli­ ability estimates (Cronbach’s alpha) were computed for each variable within each perceptual dimension. Those variables which tended to con­ sistently show a low item-to-total reliability coefficient across all

three points in time were eliminated from the derivation of the mean

perceptual scores.

An additional reliability estimate was also performed on all the

items which comprised the life style and activity dimensions. Since 94 these items had been previously reduced by the factor analyses, further reduction was not warranted, and the reliability estimates were used as indices of measurement stability and error.

As discussed earlier, the creativity scale was designed so that a singular score could be assigned to each individual. A split-half reliability procedure was conducted, which essentially measured the extent to which the randomly divided halves of the creativity scale tended to measure the same thing. A Spearman Brown reliability coeffi- ci .c was therefore used to estimate the creativity scale reliability.

A by-product of the previously described factor analytic techniques was an increased understanding of the construct validity associated with the life style and activity summated mean scores. As Kerlinger (1973) and Green and Tull (1978) point out, the substantive purpose of factor analysis is the search and test of constructs, or dimensions assumed to underlie manifest variables (see Green and Tull, p. 419). Those con­ structs which emerged from the life style and activities factor analyses were therefore the result of shared common-factor variances, and this

is in fact the definition assigned to validity by Kerlinger (see

pp. 685-686). Kerlinger elaborates on this concept by stating:

"...construct validity seeks the 'meaning' of a construct through the

relations between the construct and other constructs." Since factor

analysis enables the researcher to study the constitutive meaning of

constructs, it increases the researcher's understanding of construct

validity. Thus, the construct validity of the life style and activity

scales was assessed through factor analytic techniques. This validity

was also assessed through an intercorrelation matrix of the life style, 95 activity, creativity, and perceptual scales. This procedure was used to determine the relationships or homogenity of these variable constructs.

It was, of course, also necessary to evaluate the three dependent measures of potential adoption. The reader is reminded that these measures consisted of three singular scores representative of perceived

Channel 2000 use, purchase probability, and price sensitivity (a Guttman

type scale). To accomplish this, these three measures were first cor­ related to one another to determine the degree to which these variables covaried with one another. Next, respondents were classified as being either a high, medium, or low purchase probability individual according

to their scores on each of the three variables. These variables were

then crosstabulated with one another to determine the degree of simil­

arity or overlap between the measurement scales. Finally, the price

sensitivity measure was evaluated in terms of its ability to fulfill

the requirements of the Guttman scaling procedure. Thus, a coefficient

of reproductivity was calculated to determine the extent to which a

respondent's scale score was predictive of his/her actual response

pattern. The degree to which a singular Guttman scale can accurately

reflect a respondent's reaction to the five differing measures of price

sensitivity was the scale's resultant coefficient of reproducibility.

Based on these three complementary assessment techniques, a singular

dependent variable was chosen for subsequent research analysis.

Phase III; Analysis of Research Questions

Research questions #1 and #2 dealt with the perceptions of Channel

2000 at three points in time. To explore these perceptions, mean scores

for the six perceptual dimensions were first calculated at each time 96 period. These means were then tested for significant differences across time by a repeated measures analysis of variance. Those perceptual dimensions which demonstrated significant differences across time were explored further with a Scheffe post hoc multiple comparison test.

These procedures provided a straightforward analysis of the Channel 2000 perceptions over three points in time.

The third research question addressed the issue of whether or not interrelationships existed between each 'of the innovation perceptions.

An analysis of these relationships was important for two reasons. To understand these perceptual dimensions, it was important to assess the degree to which each of these dimensions could be associated with each other. If these constructs are to be useful for future innovation and diffusion research, their meaning must be clearly defined and delineated.

The correlation coefficients exhibited between these perceptual dimen­

sions contributed to this process. Secondly, the multivariate statis­

tical techniques that were used in the subsequent analysis could have been affected by the multicollinearity exhibited by the perceptual vari­ able set. Therefore it was necessary to establish the degree of multi­

collinearity among these perceptual variables at each point in time for

analytic procedures employed in subsequent phases of this research. An

intercorrelation matrix of Pearson product-moment correlations among

the six perceptual dimensions at each point in time was therefore de­

rived for theoretical and statistical understanding.

To determine the relationships between the innovation perceptions,

life style, and creativity (research question //A), a similar statistical

procedure was utilized for this variable set. Pearson product-moment 97 correlation coefficients were derived for each unique variable combina­ tion, at Time III. Here again, Time III perceptions were selected due to their theoretical importance to the adoption process.

The primary multivariate technique that was used to address re­ search questions #5 and It6 was discriminant analysis. Discriminant analysis is a statistical procedure that permitted an exploration of the differences between potential adopters and non-adopters according to

perceptual, life style, creativity, and sociodemographic variables in a simultaneous manner. Green and Tull (1978) outline the three major objectives of two group discriminant analysis as:

1. Establishing procedures for assigning new individuals, whose profiles, but not group identity are known, to one of the two groups.

2. Testing whether significant differences exist among the mean profiles of groups (group centroids).

3. Determining which variables account for intergroup dif­ ferences in the mean profiles.

Klecka (1979) divides these objectives or "discriminant analysis

activities" into those used for interpreting the group differences, and

those employed to classify cases into groups. Klecka (p. 9) expands

upon this division of research purpose by stating:

A researcher is engaged in interpretation when studying the ways in which groups differ— that is, is one able to "dis­ criminate" between the groups on the basis of some set of characteristics, how well do they (these characteristics) discriminate, and which characteristics are the most power­ ful discriminators?

These research purposes coincide with Green and Tull's second and third

discriminate analysis objectives. They also address the principal re­

search questions posed by this study. Specifically, how well do innova­

tion perceptions, life style, creativity and sociodemographic variables 98 discriminate between potential adopters and non-adopters at three points in time, and which of these variables are the most powerful discrimin­ ators? Although the degree to which the resultant discriminant function could accurately classify Channel 2000 respondents was related to the discriminatory ability of the function itself, this research objective was clearly a secondary element of this data analysis procedure.

Thus, a two group discriminant function was utilized to determine whether Channel 2000 respondents classified as potential adopters could be differentiated from those who were classified as potential non­ adopters. Since the derivation of a discriminant function is partially based upon the within-groups covariance matrix, it was mandatory for the number of subjects within the smallest group (the potential adopters), to exceed the total number of variables used in the discriminant analysis by at least a factor of 3 to 1, so that the covariance matrices could achieve an acceptable level of reliability.

To achieve this necessary accuracy, a series of preliminary dis­

criminant analyses were run on each of the variable constructs, so that only those dimensions or scales which represented the dimensions of a

singular construct were included in the discriminant analysis. For

example, a discriminant analysis was run on the innovation perception

construct which included the scale values of relative advantage, com­

patibility, complexity, trialability, observability, and risk only.

Similar discriminant runs were done on the constructs of life style

(which included the creativity scale), activities, and sociodemographics.

Those scales which achieved a significant partial F ratio at any

time period were included as independent variables in the final 99 discriminant runs.

For both the preliminary and final discriminant analyses, a stepwise procedure sequentially entered variables in terms of their differential ability (as measured by their partial F ratio) to affect discrimination between the two groups. This F-to-enter procedure was based on a partial multivariate F statistic which tested the additional discrimination intro­ duced by a variable, after taking into account the discrimination that had been achieved by the other variables that had already entered. Thus, a derived partial F which was insignificant, indicated that this vari­ able was not adding enough to the overall discrimination (Klecka, 1978,

p. 57, and Dixon, 1973, p. 241).

Once these stepwise discriminant procedures had been completed, a

canonical discriminant function was derived for each time period of the

Channel 2000 test. In order to determine both the substantive and sta­

tistical significance of these functions, several procedures were em-

ployed. First, an initial examination of the canonical correlation co­

efficients derived from the canonical discriminant functions, summarized

the degree of relatedness between the two groups and the discriminant

function. As Klecka (1978, p. 37) points out:

...the canonical correlation is useful because it reports how well the discriminant function is doing. If the groups are not very different on the variables analyzed, then the canonical correlation will be low, because we cannot create discrimination when none already exists.

A second measure of the substantive significance of the discrimin­

ant functions used in this research was an examination of the classifi­

cation matrices produced by the discriminant functions. In this pro­

cedure, the derived discriminant scores were used to classify the 100

Channel 2000 respondents as being either potential adopters or non­ adopters. As mentioned earlier, this analysis was secondary to the interpretative purposes of the discriminant procedures; however, it did provide some rough estimate of each discriminant function's predictive ability.

To test for statistical differences between the two group centroids derived from each of the discriminant functions, a Wilks' lambda test was employed at each time period. These calculated Wilks' lambda values were then converted into overall F ratios so that statistical

significance could be determined. These significance tests were used

to determine the probability that the sampling process used in the

Channel 2000 test produced cases which computed a certain degree of

potential adoption/non-adoption discrimination, when in fact, no such

group differences existed in the population. If statistical signifi­

cance is achieved through the F test, it is relatively safe to conclude

that the group differences explained by the final discriminant functions

did not occur by chance.

Finally it was important to determine the degree to which each of

the independent variables accounted for the differences between potential

adopters and non-adopters. This was accomplished by two procedures.

First, the partial F ratios and their levels of significance for each

discriminant function were examined. Those variables with the most

significant partial F ratios were interpreted as being the most effective

of the discriminating variables. Second, all the canonical structure

coefficients were examined to determine the relationships and associa­

tions between each of the independent variables and the final canonical discriminant function. These procedures directly tested the importance of innovation perceptions, life style, creativity and sociodemographic variables in the discrimination of potential-adopters and non-adopters. CHAPTER V

RESULTS

As previously mentioned, the data analysis procedures were sub­ divided into three major phases. In Phase I, single item variables were analyzed via factor analysis in an effort to combine several question­ naire items into singular measurement constructs. During Phase II, these constructs and others were assessed in terms of their overall reliability and validity. Finally, during Phase III, each of the re­ search questions were examined in light of the previously mentioned measurement devices. The chapter begins with an overview of the demo­ graphic characteristics used to describe the Channel 2000 test population.

Subjects

The sample consisted of 141 individuals who comprised 71 household units. Approximately 56 percent of the subjects were between the ages of 25 and 44. Seventy-five percent of the sample indicated they had at least some college education, while 47 percent had received a four year college degree. Forty-six percent of the Channel 2000 sample reported an annual household income of $30,000 or more. Table 10 lists the frequency distributions and relative frequencies for these socic- demographic indicators.

Variable Frequencies

As evidenced by the three questionnaires presented in Appendix C, the Channel 2000 test produced a rather large data set. In an effort to document all of the variables that were originally used in this

102 103 TABLE 10

FREQUENCY DISTRIBUTIONS OF CHANNEL 2000 SUBJECTS (N = HI)

Relative Relative Frequency Demographic Frequency (Percent)

Age 16-24 24 17 25-34 52 37 35-44 27 19 45-54 24 17 55-64 9 6 65 + 5 4

Sex Male 69 49 Female 72 51

Education Grade School 1 1 Some High School 14 10 High School Graduate 19 14 Some College 33 23 2 Year College Graduate 7 5 4 Year College Graduate 32 23 Post Graduate 34 24

Household Income Under $10,000 3 2 $10,000 - $19,999 21 15 $20,000 - $29,999 52 37 $30,000 + 65 46 104 research, the frequencies for the variables contained in the three ques­ tionnaire instruments are also presented in Appendix C.

The frequencies are potentially useful on several accounts. First, an increased understanding of how people responded to the individual items which comprised the scales used in this research can be obtained.

Second, these frequencies may generate additional research conceptual­ izations of how videotex should be researched in the future. And finally, an examination of these frequencies should develop in others an apprecia­ tion of the complexities inherent to an examination of videotex tech­ nology.

Phase I: Data Reduction

Life Style Factor Analysis

Table 11 reports the results of the factor analysis which was com­ pleted on the 26 life style variables from the Time II questionnaire

(also see Appendix G for the life style intercorrelation matrix). A

6 factor solution was selected based on the results of a scree test

(Cattell, 1966), an eigenvalue criterion of 1.0 or greater, and an exam­ ination of the factor loadings across the 4, 5, 6, and 7 oblique factor solutions.

The six factors were interpreted according to those variables which had a factor loading of approximately .50 or higher on any given factor, and those variables which did not have substantial secondary loadings on any other factors. Factor I appeared to represent a con­ servative attitudinal dimension. The highest loading items, 4 and 6, referred to the degree to which an individual felt he or she had old fashioned tastes, and accordingly, believed the man should be the boss TABLE 11

ROTATED FACTOR PATTERN MATRIX FOR LIFE STYLE

Factor I Factor II Factor III Factor IV Factor V Factor VI T Conserva- TV Evalua- Time Leader- Life Style Item ... „ Shopping Concern J tion tion Pressure ship rr °

1. To keep up with the news, I primarily rely on television. -.01 -.53 .14 -.10 .16 -.22

2. I have more self-confidence than most of my friends. -.10 -.10 .14 ^64 .09 .10

3. Television is my primary form of entertainment. .16 -.47 .00 .08 .04 -.03

4. I have somewhat old-fashioned tastes and habits. .57 .10 -.14 .06 -.21 -.14

5. I find that I have less time to do the things I want to do than I used to. .07 -.07 j_64_ -.05 -.13 -.18

6. I enjoy science fiction. -.25 .09 .36 .25 .05 .11

7. I would go to the library more often if I could find the time. -.16 .43 .55 .01 .17 .03

8. I don't like to take chances. .37 -.05 -.02 -.46 -.21 -.34

9. I wish I didn't have to spend so much 105 time waiting in lines. -.06 .14 .04 .10 .21 .06 TABLE 11 (Continued)

ROTATED FACTOR PATTERN MATRIX FOR LIFE STYLE

Factor I Factor II Factor III Factor IV Factor V Factor VI Conserva- TV Evalua- Time Leader- Life Style Item Shopping Concern tion tion Pressure ship

10. My opinions on things don’t count very much. .28 -.32 -.19 -.42 -.41 -.30

11. I usually read the sports page in the daily newspaper. .24 -.08 -.20 .49 .01 .02

12. I always find plenty of time avail­ able to get all my necessary reading done. .12 .04 -.53 .01 .07 .07

13. I use many of the brands and products that my parents used to use. .23 -.27 -.20 -.14 .02 -.14

14. I am usually among the first to try new products. -.02 -.03 -.11 ,04 .60 -.01

15. Information from advertising helps me make better buying decisions. .23 -.31 -.18 -.07 .57 -.01

16. I worry a lot more these days about the security of my family than I used to. .19 -.18 .04 -.07 18 -.59

17. Magazines are more interesting than 106 television. -.12 .61 18 .09 .09 -.03 TABLE 11 (Continued)

ROTATED FACTOR PATTERN MATRIX FOR LIFE STYLE

Factor I Factor II Factor III Factor IV Factor V Factor VI Conserva­ TV Evalua­ Time Leader­ Life Style Item Shopping Concern tion tion Pressure ship

18. I like to be considered a leader. -.11 .16 .06 .67 .10 .08

19. Today, most young people don't have enough discipline. .46 -.03 -.18 -.01 .10 -.31

20. I don't like television news programs. .05 .28 -.04 -.12 -.40 .06

21. The man should be the boss In the house. .67 -.35 -.10 -.06 -.09 -.17

22. Sometimes I feel that there is so much news available that I can't keep up with it. -.07 .07 .22 -.03 -.13 -.60

23. I am in favor of very strict enforce­ ment of all laws. .46 -.08 -.06 -.13 .07 -.08

24. Getting information by TV won't ever take the place of reading a book or newspaper article.* .07 .44 -.08 .01 -.11 .07

25. Everything is changing too fast. .36 -.03 -.18 -.22 -.10 -.67 TABLE 11 (Continued)

ROTATED FACTOR PATTERN MATRIX FOR LIFE STYLE

Factor I Factor II Factor III Factor IV Factor V Factor VI Conserva­ TV Evalua­ Time Leader­ Life Style Item Shopping Concern tion tion Pressure ship

26. Television advertising insults my intelligence -.20 .46 .21 .00 -.06 -.12

Eigenvalue 3.45 2.51 2.04 1.91 1.71 1.38

Percent of total variance . 13.3 9.7 7.8 7.4 6.6 5.3

o CQ 109

in the house. To a lesser extent, this factor was also represented by

items 19 and 23, which were also of a conservative persuasion. This

factor was therefore labeled as a "conservative" dimension of life style and items 4 and 21 were the two variables which comprised this scale.

Factor II was characterized by a series of variables which related

to how an individual evaluated and used television in their daily lives, as evidenced by the loadings of items 1, 3, and 17. Again, to a lesser extent, this factor was also represented by items 24 and 26 which also dealt with an individual's attitudes towards television.

Two of the three highest loadings (items 1 and 3) had negative

factor loadings and indicated that the individual did not depend on tele­ vision for news and entertainment. Item 17, however, had a positive

factor loading and indicated a preference for magazines over television.

If the respondents agreed with items 1 and 3, it indicated that they

tended to be more dependent on television for news and entertainment.

Agreement with item 17, however, indicated an opposite attitudinal pat­

tern. Because item 17 was contrary to the directionality implied by

items 1 and 3, it was recoded so that a respondent's agreement with any

of these items would tend to reflect a more positive orientation towards

television.

Thus, items 1, 3, and 17 served as the three variables which com­

prised the "TV evaluation" dimension of life style. If an individual

received a high mean score on this scale, it was an indication that

this person tended to be more dependent on television for news and enter­

tainment, and also felt that television was more interesting than

magazines. 110

Two life style items loaded on Factor III. These items were 5 and

12, and both related to a perceived lack of time to get things done.

Because item 12 was negatively worded, this item was recoded in a posi­ tive direction and combined with item 5 to form the composite scale of

"time pressure."

Factor IV was represented by positive factor loadings on self-con­ fidence (item 2), and leadership qualities (item 18). This factor was therefore considered to be a "leadership" dimension of life style.

Factor V related to the degree to which an individual tried new products (item 14) and the potential impact that advertising information had on purchase decisions (item 15). Because these two items related to shopping behavior, this factor was labeled as a "shopping" dimension of life style.

Factor VI had negative factor loadings on items 22, 25, and 16. It appeared that this factor represented a dimension of concern (or lack of it) about family security, and the information overload which could result from a rapidly changing society emersed in media news information.

Because this factor had a questionable degree of conceptual homo­ geneity, there was some doubt as to the usefulness of this life dimension

for this research. However, due to the exploratory nature of this study,

this factor was included, and labeled as a "concern" dimension of life style. A high score on this life style construct indicated that the

individual was concerned about family security, and a rapidly changing environment in which he or she lived. Ill

Activities Factor Analysis

The analytic procedures used in the life style factor analysis were also employed for interpretative purposes in the analysis of the 15 activity items collected at Time I. Examinations of the 4, 5, and 6 oblique rotation solutions indicated that a 5 factor solution produced the optimal factor structure in terms of its explanatory ability and uniqueness of factor loadings. These loadings are presented in Table 12

(also see Appendix I for the activities intercorrelation matrix).

The first factor was clearly a library usage activity factor as indicated by the loadings of items 12 and 13. This factor was therefore

termed "library use." Factor II was represented by the usage of a hand held calculator (item 3), credit cards (item 9), and frequent business

trips requiring air travel (item 10). This factor was labeled "business"

to represent these business related activities. Items 11 and 14, which

described banking related activities, both loaded on Factor III. This

factor was therefore labeled "finance." Factors IV and V were self-

explanatory since only a singular activity item loaded on each factor.

Factor IV was "encyclopedia use" (item 2) and Factor V was "computer

use" (item 5). Although these two factors were represented by only a

singular variable, they were retained because of their conceptual im­

portance to the Channel 2000 study. The reader is of course cautioned

in regards to the stability of a single item factor.

The underlined variables presented in Tables 11 and 12 were used

to derive the mean scale scores on each of these dimensions. Therefore,

each individual had 6 life style and 5 activity mean scores assigned to

him or her on the basis of responses given to the variables which TABLE 12

ROTATED FACTOR PATTERN MATRIX FOR ACTIVITIES

Factor I Factor II Factor III Factor IV Factor V Activity Library Use Business Finance Encyclopedia Use Computer Use

1. Read a book .48 -.16 .08 .15 .15

2. Looked up something in an encyclopedia .23 -.12 -.10 .84 .08

3. Used a hand held calculator .07 .51 .09 .08 .23

4. Used an automatic bank teller machine .02 .25 .23 -.09 .15

5. Operated a computer .04 .22 .01 -.04 .58

6. Order from a mail order catalog using an 800 number -.03 .29 -.09 .08 .08

7. Used a typewriter .23 .30 .15 .30 .07

8. Used the phone to call for weather information -.06 .07 .02 .33 -.05

9. Used a credit card .09 .56 .32 .12 -.12

10. Flew on an airplane for a business trip -.04 .54 -.13 -.01 -.01

11. Balanced a checkbook .24 .10 .51 -.07 -.20

12. Went to a library .85 .05 -.05 .04 .12 TABLE 12 (Continued)

ROTATED FACTOR PATTERN MATRIX FOR ACTIVITIES

Factor I Factor II Factor III Factor IV Factor V Activity Library Use Business Finance Encyclopedia Use Computer Use

13. Checked out a book from the library .91 .03 -.04 -.05 -.10

14. Called a bank to get information about an account -.10 -.18 .69 .06 .15

15. Used the Books By Mail service .23 -.05 .07 .10 .29

Eigenvalue 3.60 1.81 1.40 1.14 1.00

Percent of total variance 24.1 12.1 9.4 7.7 6.7 114 represented each of these factors.

Phase II: Independent and Dependent Scale Assessment

Independent Variables

The six dimensions which comprised the construct of innovation per­

ception were each tested for their internal reliability using a series

of Cronbach's alpha reliability coefficients. Table 13 lists these

reliability coefficients and also lists those variables which were elim­

inated from subsequent scale construction procedures due to their low

reliability coefficients. This scale reduction procedure reduced the

original 36 perceptual variables down to a total of 31. Again, those

variables which comprised a singular perceptual dimension, and also

exhibited sufficient reliability coefficients, were combined into a singu­

lar perceptual mean score.

Table 13 demonstrates that consistently high reliability coeffi­

cients for the perceptual dimensions of compatibility (r .84) and

relative advantage (r X; .78) were exhibited across the three points in

time. The remaining dimensions of complexity (r X .50), observability

(r x .51), trialability (r x .49) and risk (r x .58) showed a moder­

ate amount of reliability, although these coefficients indicated that

there remained a certain degree of measurement error in the scales

themselves.

As a reliability check on the life style and activity dimensions, a

series of alpha coefficients were computed for those dimensions that had

more than one variable. These reliability coefficients are presented in

Table 14. Since all these variables had been previously reduced by the

factor analytic procedures, it was not necessary to eliminate any of 115

TABLE 13

ALPHA RELIABILITY COEFFICIENTS FOR PERCEPTUAL DIMENSIONS

AT THREE POINTS IN TIME

Perceptual Dimension Time I Time II Time III

g Relative Advantage .77 .81 .78

Compatibility^ .82 .84 .85 Q Complexity .42 .50 .58

Observability^ .55 .48 .51

Trialability .57 .30 .59

Risk .64 .48 .61

Perceptual Item "Channel 2000 would probably cost a lot of money" omitted from future analysis.

k Perceptual item "I have had to change some of the ways I do things since I've been using Channel 2000" omitted from future analysis.

£ Perceptual items "It would be harder to pay bills using Channel 2000 than it is for me now" and "I don't think Channel 2000 would make my banking any easier to do" omitted from future analysis.

Perceptual item "I have heard about or seen devices like Channel 2000 before" omitted from future analysis. 116

TABLE 14

RELIABILITY COEFFICIENTS FOR LIFE STYLE

AND ACTIVITY DIMENSIONS

Dimension Reliability Coefficient

Life Style

Conservative .51

Television Evaluation .58

Time Pressure .63

Leadership .58

Shopping .57

Concern .60

Activities

Library Use .86

Business .54

Finance .46 117 these variables from the derivation of the mean scale scores. The reliability coefficients presented in Table 14, however, indicated that these measures did not achieve the levels of reliability that might have been expected. Clearly the activity dimension of library use was quite acceptable at .86, although the remaining reliability estimates tended to fall between .50 and .60. This again could be due to measurement error, although the reader is reminded that many of these dimensions were represented by only 2 or 3 variables. Naturally, a small number of variables cannot be expected to achieve the reliability levels of a multiple item scale.

This is quite apparent in light of the split half reliability co­ efficient derived for the creativity scale. Here, the Spearman Brown reliability coefficient reached .94 for the 30 item scale. Leavitt and

Walton (1977) have reported similar high reliability estimates for this scale (Kuder Richardson of .82 and a test-retest of .72).

Dependent Variables

The three dependent measures of potential adoption that were avail­ able for analytic purposes in this research included purchase probabil­ ity and price sensitivity (Guttman scale) at three points in time, and a measure of perceived Channel 2000 use. Several procedures were used to determine which dependent variable should be used for this research.

First, Table 15 shows the Pearson product-moment correlations among

these three measures. Moderately high positive correlations (r = .51

to .68) were found between the measures of price sensitivity and pur­

chase probability at their corresponding time periods. Low positive

and negative correlations (r = -.12 to .14) were exhibited between 118

TABLE 15

CORRELATION MATRIX OF PERCEIVED USE,

PRICE SENSITIVITY AND PURCHASE PROBABILITY

AT THREE POINTS IN TIME

Dependent Variable 1 2 3 4 5 6 7

1. Perceived Use -.12 .14 .09 -.04 .32 .23

Price sensitivity

2. Time I .26 .23 .51 .22 .19

3. Time II ___ .54 .26 .62 .61

4. Time III .39 .49 .68

Purchase Probability

5. Time I .33 .43

6. Time II .74

7. Time III 119 perceived use and purchase probability (r = -.04 to .32). These find­ ings indicated that there was a positive relationship between purchase probability and price sensitivity, and an extremely low correlation between these two measures and perceived use.

To determine the degree of overlap between these three measures of potential adoption, a series of crosstabulations were run on these de­ pendent measures. This procedure provided an index of similarity be­ tween those individuals who were classified into one category of poten­

tial adoption on one measure, and also classified into a similar potential adoption category on another.

For this type of analysis to be meaningful, it was first necessary

to assign individuals to three levels of potential adoption (high, medium, or low), according to their scores on each dependent measure. To do

this, the frequency distributions of these measures were first examined.

Tables 16, 17, and 18 list the frequency distributions for the three measures of potential adoption at each representative time period. The

reader is reminded that a measure of perceived use was obtained at

Time II only.

Next, the measures of price sensitivity and purchase probability

were trichotomized at Time III, along with the measure of perceived use.

This procedure was done so that each collapsed potential adopter category

contained an approximately equal number of respondents across the three

measures. Table 19 lists these measures and the corresponding frequency

counts within each potential adopter category.

When these variables were crosstabulated with each other, 61 percent

of the respondents were classified similarly by the price sensitivity 120

TABLE 16

FREQUENCY DISTRIBUTION FOR PRICE SENSITIVITY

AT THREE POINTS IN TIME

Score Time I Time II Time III

oa 17 53 58

1 20 29 29

2 32 20 29

3 27 18 13

4 13 6 4

5 31 6 7

Mean 2.66 1.34 1.26

S. D. 1.67 1.46 1.41

Note that in this table 0 « would not pay $5.00 and 5 = would pay $15.00- 121

TABLE 17

FREQUENCY DISTRIBUTION FOR PURCHASE PROBABILITY AT THREE POINTS IN TIME

Score Time I Time II Time III

oa 1 34 23

1 7 14 17

2 8 9 9

3 11 11 14

4 9 8 4

5 33 19 18

6 22 14 14

7 23 10 10

8 11 9 15

9 5 8 12

10 9 5 4

Mean 5.56 3.83 4.32

S. D. 2.31 3.16 3.18

a Note that in this table 0 = absolutely would not subscribe and 10 = absolutely would subscribe. 122

TABLE 18

FREQUENCY DISTRIBUTION FOR PERCEIVED USE

(TIME II)

Score Range Frequency

0 - 1.83a 55

1.84 - 2.33 56

2.34 - 4.33 30

Mean 1.93

S. D. 0.93

Note that in this table 0 = never used and 5 = used every day. 123

TABLE 19

FREQUENCY DISTRIBUTION FOR THREE COLLAPSED MEASURES

OF POTENTIAL ADOPTION

Adoption Category

Dependent Measure Low Med High

Price Sensitivity 58 58 24 (Time III) (o)a (1 - 2) (3 - 5)

Purchase Probability 49 60 31 (Time III) (0 - 2) (3 - 7) (8 - 10)

Perceived Use 55 56 30 (0 - 1.83) (1.84 - 2.33) (2.34 - 4.33)

The values in parentheses represent the raw score ranges within a given category. 124 measure and the purchase probability measure. A 41 percent level agree­ ment was reached between price sensitivity and perceived use measures, while a 42 percent agreement level was achieved between the measures of purchase probability and perceived use. Again it appeared that price sensitivity and purchase probability tended to approximate a similar measure of potential Channel 2000 adoption.

Since a moderately strong level of agreement was reached between these two measures, a degree of construct validity had been achieved.

Perceived use, however, appeared to be measuring something distinct from these other two measures, as evidenced by the correlation matrix (see

Table 15), and the results of the crosstabulation. This measure was

therefore not used for subsequent analysis procedures.

Since a respectable level of agreement was found between the measures of price sensitivity and purchase probability, additional criteria were

developed to gauge the effectiveness of each measure. A straightforward

approach to this problem was to simply evaluate the scaling procedure

employed in each measure. The measure of purchase probability required

the respondent to evaluate his or her probability of subscribing to

Channel 2000 at some time in the future. Unfortunately, there was no

price associated with this scale. Hypothetically, it was possible for

one subject to base a decision on a price factor quite different from

that used by another subject. The price sensitivity measure, however,

had some very specific price ranges associated with it (ranging from

$5.00 to $15.00 per month), and would therefore tend to reduce this

potential measurement error. Furthermore, a much more concrete definition

of potential adoption can be associated with a scale that actually 125 contains monetary values. This kind of information is also more "action­ able" from a marketing standpoint.

Based on this reasoning, the Guttman scale of price sensitivity was employed as the primary measure of potential Channel 2000 adoption.

As a further test of this scale's appropriateness, a Guttman scale analysis was conducted on this measure at each point in time. This pro­ cedure tested the scalability of the price sensitivity scales in terms of their ability to fulfill the scaling requirement associated with the

Guttman procedure. This measure, called the coefficient of reproduc­ ibility, resulted in coefficients of .97, .98, and .97 for the measures of price sensitivity at Times I, II, and III. Since this coefficient can vary from 0 to 1.0, these values indicated that the price sensitivity measure very accurately reflected an individual’s response pattern to the five measures of price across the three points in time.

The frequency distribution for the collapsed measure of price sen­ sitivity is presented in Table 20. This measure of potential adoption therefore served as the dependent measure in the two group discriminant analysis procedures. As noted in Table 20, potential adoption was de­ fined according to an individual’s score on the price sensitivity measure at each point in time.

Phase III: Analysis of Research Questions

Research Questions //I and #2

The means and standard deviations of the Channel 2000 perceptions according to the six summated perceptual scales are listed In Table 21.

Several general tendencies were noted among all the scales. Apart from the complexity dimension, all scales exhibited a noticeable decrease at 126

TABLE 20

FREQUENCY DISTRIBUTION FOR THREE COLLAPSED MEASURES

OF PRICE SENSITIVITY

Adoption Category

Time Low Med High Period (Potential Non-Adopter) (Potential Adopter)

I 36 59 44 (0 - i)a (2 - 3) (4 - 5)

II 53 49 30 (0) (1 “ 2) (3 - 5)

III 58 58 24 (0) (1 - 2) (3 - 5)

The values in parentheses represent the raw score ranges within a given category. 127

TABLE 21

MEANS AND STANDARD DEVIATIONS OF PERCEPTUAL SCORES

AT THREE POINTS IN TIME

Perceptual Scale Time I Time II Time III

Relative Advantage

M 4.62a 3.66 3.84

SD .80 1.13 1.08

Compatibility

M 4.27 3.83 3.90

SD .71 .88 .85

Complexity

M 4.89 5.02 4.91

SD .71 .89 .86

Observability

M 5.21 4.71 4.67

SD .72 .94 .86

Trlalability

M 4.32 3.92 4.22

SD .88 .92 .87

Risk

M 4.53 4.16 4.36

SD .85 .89 .87

Note that in this table 1 = definitely disagree and 6 = definitely agree. All negatively worded items were recoded to be unidirectional in a positive sense. 128

Time II. For the dimensions of relative advantage, compatibility, trial- ability, and risk, these mean scores then all increased at Time III.

To test for differences between each of these mean perceptual

scores at the three points in time, a series of repeated measures, one way analysis of variances were computed. These results, presented in

Table 22, showed significant differences between the mean scores of the

perceptual dimensions relative advantage (F - 64.56, df = 2/260, p <

.001), compatibility (F = 27.13, df = 2/262, p < .001), observability

(F = 23.01, df = 2/262, p < .001), trialability (F = 12.32, df = 2/262,

p < .001). The mean differences between the complexity dimension failed

to reach significance (F = 1.80, df = 2/262, p > .05).

The Scheffe test was employed to determine the means between which

significant differences existed at each time period. It was found that

among those perceptual dimensions which exhibited significant changes in

mean scores, these scores all significantly decreased from Time I to

Time II (see Table 23). This indicated that Channel 2000 was perceived

in a less favorable manner at Time II than at Time I. The dimensions

of relative advantage, compatibility and observability also exhibited a

significant decrease in mean scores between Time I and Time III.

Although significance was not reached between Time I and Time III for

trialability and risk, these dimensions did show a significant increase

( in mean perceptual scores between Time II and Time III. Thus the pre­

viously mentioned mean score Increases between Time II and Time III

were significant for these two dimensions only. TABLE 22

ANALYSIS OF VARIANCE SUMMARIES FOR THE PERCEPTUAL SCORES

AT THREE POINTS IN TIME

Perceptual Scale SS df MS F

Relative Advantage

Between Measures 67.19 2 33.59 64.56* Error 135.29 260 0.52

Compatibility

Between Measures 14.86 2 7.43 27.13* Error 71.76 262 .27

Complexity

Between Measures 1.25 2 .62 1.80 Error 90.83 262 .35

Observability

Between Measures 24.05 2 12.02 23.01* Error 136.91 262 .52

Trialability

Between Measures 11.64 2 5.82 12.32* Error 123.79 262 .47

Risk

Between Measures 9.51 2 4.76 10.88* Error 114.51 262 .44

* p < .001 130

TABLE 23 / SCHEFFE TEST OF PERCEPTUAL MEANS AT THREE POINTS IN TIME

Mean Time I Time II Time III Critical Value

Relative Advantage 4.62 3.66 3.84 ■ .217

I 4.62 .952* .774* II 3.66 .179 III 3.84

Compatibility 4.27 3.83 3.90 .157

I 4.27 .439* .373* II 3.83 .067 III 3.90

Observability 5.21 4.71 4.67 .217

I 5.21 .500* .543* II 4.71 .043 III 4.67

Trialability 4.32 3.92 4.22 .206

I 4.23 .403* .098 II 3.92 .304* III 4.22

Risk 4.53 4.16 4.36 .198

I 4.53 .379* .178 II 4.16 .201* III 4.36

* p < .05 131

Research Question #3

In order to examine the relationships between these perceptual dimensions at three different points in time, three separate Pearson product moment correlation matrices were constructed. Tables 24, 25, and 26 present these results for the six perceptual dimensions. Results indicated that the perceptions were all positively correlated to one another, although these correlations were not consistently strong

(r = .18 to .67, Time I; r = .13 to .78, Time II; r = .12 to .75,

Time III). The correlations between relative advantage and compatibil­ ity were the strongest of this group (r = .67 to .78), and the lowest correlations were found between relative advantage and complexity

(r = .]2 to .28). Other variables which exhibited moderate positive correlations were compatibility and risk (r .60, Times I and III), and observability and risk (r = .62, Time III).

Research Question //4

The relationships between the innovation perceptions and the life style, activities, and consumer creativity scales, were determined through similar correlation techniques. Table 27 lists each of the

Pearson product moment correlations between the perceptual dimensions and the summated life style, activities, and creativity scales. Only moderate correlations were found between these variables (r - -.31 to

.40); however, several observations are in order.

First, consistently higher positive correlations were found between

the life style dimension of shopping and all the perceptual dimensions, with the exception of complexity. This indicated that there was a posi­

tive correlation between an individual's tendency to purchase new Table 24 132

Correlation Matrix of Perceptual Scores at Time I

Perception 1 2 3 4 5 6

1. Relative Advantage --- .67 .28 .34 .24 .44

2. Compatibility --- .36 .48 .30 .61

3. Complexity -- .21 .26 .49

4. Observatlllty --- .28 .43

5. Trialability --- .18

6. Risk --

Table 25

Correlation Matrix of Perceptual Scores at Time XI

Perception 1 2 3 4 5 6

1. Relative Advantage --- .78 .13 .45 .30 .41

2. Compatibility -- .23 .47 .33 .46

3. Complexity -- .28 .30 .35

4. Observability -- .34 .41

5. Trialability -- .34

6. Risk --

Table 26

Correlation Matrix of Perceptual •ScoreB at Time III

.. Perception 1 2 3 4 5 6

1. Relative Advantage --- .75 .12 .49 .34 ,54

2. Compatibility -- .27 .61 .42 .60

3. Complexity -- .28 .27 .35

4. Observability -- .43 .62

S. Trialability • -- .50

6. Risk --

Note: The value of a correlation to be significant at the .OS level for a sample size of 141 Is .19. TABLE 27

CORRELATION MATRIX OF INNOVATION PERCEPTIONS BY LIFE STYLE, ACTIVITIES AND CONSUMER CREATIVITY AT TIME III

Scale Name Relative Advantage Compatibility Complexity Observability Trialability Risk

Conservative -.08 -.07 -.07 -.02 -.03 -.15

TV Evaluation .29 .30 .09 .23 .12 .30

Time Pressure .08 .08 -.10 .00 -.21 -.08

Leadership .06 .10 .19 .12 .04 .12

Shopping .30 .27 .02 .28 .28 .35

Concern .04 -.07 -.14 -.05 -.08 .00

Library Use .00 .06 .27 .10 -.06 .12

Business .03 .15 .40 .20 .18 .20

Finance .19 .18 .08 -.05 .05 .11

Encyclopedia Use .07 .03 .19 .18 .06 .15

Computer Use -.13 .30 .30 -.10 -.01 -.05

Consumer Creativity -.21 -.34 -.19 -.30 -.21 -.31

Note: The value of a correlation to be significant at the .05 level for a sample size of 141 is .19. 134 products and use advertising to make these product choices, and perceive

Channel 2000 in a positive manner along these five perceptual dimensions.

Although complexity showed no relationship with this life style dimen­

sion (r = .02), it did show a moderate positive relationship to the

activities of business (r = .40), and of computer use (r = .30). Here,

the degree to which Channel 2000 was perceived as a non-complex innova­

tion tended to positively correlate with business and computer activities.

Both the perceptual dimensions of relative advantage and risk

exhibited moderate positive correlations with the life style dimensions

of TV evaluation (r = .30), and shopping (r = .33). Thus the degree

to which Channel 2000 was perceived as offering advantages over current

behavioral modes, and being low in risk, tended to positively correlate

in a mild manner with a tendency to use television for news and enter­

tainment, as well as prefer television over magazines (the reader is

reminded that a higher mean score on the TV evaluation scale indicated

a more active use of television). Also, these perceptual dimensions

tended to positively correlate in a mild manner with the propensity to

purchase new products and listen to advertisements (shopping).

Finally, it is important to note the consistently moderate negative

correlations between consumer creativity and all the perceptual dimen­

sions (r = -.19 to -.34). This indicates that the more creative the

respondent was, the more critical he/she was of Channel 2000 along each

of the perceptual dimensions.

Preliminary Discriminant Analysis

In order to further reduce the variable set used in research ques­

tions //5 and //6, it was first necessary to run a series of four separate 135 two-group stepwise discriminant analyses on each variable construct at each point in time. Using the dichotomous criterion of potential adop­ tion or non-adoption according to the Guttman scale of price sensitivity, a stepwise discriminant analysis with an F to enter criterion of .05 was run on the perceptual, life style (which included consumer creativity), activity, and sociodemographic scale items at Times X, XX, and III.

Each discriminant run contained only those scale items which comprised a separate variable construct, so that those scale items which achieved a significant partial F ratio, could be distinguished from those that did not. This procedure was designed so that only the most important discriminating scale items would be included in the final discriminant runs.

The results of this procedure are presented in Tables 28, 29, and

30 for each respective time period. Significant discriminant functions were obtained from the perceptual and life style scales at each time r period. A significant discriminant function was obtained for the

activities scales at Times I and II only. The sociodemographic variables

did not produce a significant function at any time period which suggests

that sociodemographic variables were not very useful in terms of their

ability to discriminate between potential adopters and non-adopters of

Channel 2000.

At Time I, the innovation perception discriminant function was com­

prised of the compatibility scale only. The Time IX function also

included compatibility, and added relative advantage. At Time III, the

perceptual scales of relative advantage and compatibility again were

used to derive the discriminant function. Although significant 136

TABLE 28

RESULTS OF THE TWO GROUP DISCRIMINANT ANALYSIS

FOR EACH CONSTRUCT AT TIME I

Means Potential Potential Partial F Adopters Non-Adopters Univariate Significance Scale Name N = 44 N = 36 F Level

Relative Advantage 4.95 4.20 18.84* .16 Compatibility 4.61 3.82 31.35* .00 Complexity 4.96 4.89 .225 .14 Observability 5.33 5.00 3.95 .86 Trialability 4.75 4.15 10.11 .11 Risk 4.90 4.24 12.30* .67

Overall F = 31.35, df: = 1/79*

Conservative 3.52 3.92 2.211 .22 TV Evaluation 3.72 3.84 .507 .35 Time Pressure 3.58 3.61 .319 .83 Leadership 4.52 4.50 .127 .84 Shopping 4.17 3.46 8.662* .00 Concern 3.74 3.85 .196 .56 Consumer Creativity 63.60 68.65 4.171* .23

Overall F = 8.66, df = 1/79*

Library Use 3.67 3.26 1.109 .65 Business 4.15 4.27 .190 .08 Finance 2.84 2.18 ,5.411* .02 Encyclopedia Use 4.07 3.08 5.034* .03 Computer Use 2.86 3.44 1.118 .12

Overall F = 5.12, df = 2/77*

* p < .05 137

TABLE 29

RESULTS OF THE TWO GROUP DISCRIMINANT ANALYSIS

FOR EACH CONSTRUCT AT TIME II

Means Potential Potential Partial F Adopters Non-Adopters Univariate Significance Scale Name N = 30 N = 53 F Level

Relative Advantage 4.67 2.85 88.56* .00 Compatibility 4.59 3.15 79.78* .00 Complexity 5.06 4.97 .219 .22 Observability 5.07 4.35 10.79* .89 Trialability 4.36 3.61 12.53* .42 Risk 4.48 3.79 10.18* .96

Overall F = 53.68, df = 2/80*

Conservative 3.63 3.62 .114 .79 TV Evaluation 4.11 3.65 7.776* .00 Time Pressure 3.80 3.57 2.16 .21 Leadership 4.55 4.45 .182 .66 Shopping 4.18 3.76 2.31 .16 Concern 4.06 3.83 .998 .64 Consumer Creativity 61.67 66.06 2.685 .05

Overall F = 7.77, df = 1/81*

Library Use 3.93 3.08 4.735* .46 Business 4.49 4.11 2.031 .28 Finance 3.36 2.58 6.693* .01 Encyclopedia Use 4.21 2.85 9.354* .00 Computer Use 2.31 3.25 2.952 .04

Overall F = 6.99, df = 3/78*

* P < .05 138

TABLE 30

RESULTS OF THE TWO GROUP DISCRIMINANT ANALYSIS

FOR EACH CONSTRUCT AT TIME III

Means Potential Potential Partial F Adopters Non-Adopters Univariate Significance Scale Name N = 23 N = 58 F Level

Relative Advantage 4.83 3.04 57.99* .00 Compatibility 4.52 3.20 51.35* .01 Complexity 5.07 4.92 .483 .82 Observability 5.07 4.28 15.84* .75 Trialability 4.62 3.88 12.92* .08 Risk 4.90 3.86 24.44* .50

Overall F = 34.26, df = 2/78*

Conservative 3.27 3.58 .923 .16 TV Evaluation 3.99 3.52 5.825* .02 Time Pressure 3.73 3.51 1.413 .50 Leadership 4.44 4.52 .121 .84 Shopping 4.35 3.72 5.045* .04 Concern 3.86 3.81 .308 .82 Consumer Creativity 62.00 66.10 3.00 .22

Overall F = 5.32, df = 2/79* .

Library Use 4.02 3.34 2.885 .09 Business 4.52 4.18 1.434 .23 Finance 3.30 2.73 3.173 .08 Encyclopedia Use 4.17 3.29 3.084 .08 Computer Use 2.83 3.14 .255 .62

Overall F = n.s.

* p < .05 139

univariate differences were found between several other perceptual scales; not all these variables were useful in the multivariate discriminant

procedure.

Only the shopping scale was used at Time I to derive the discrimin­

ant function for the life style scales. At Time II, the only variable

used was the TV evaluation scale, and at Time III, both of these scales

produced a significant partial F value and were thus included in the

discriminant function. Although a significant univariate F was achieved

for the consumer creativity scales at Time I, this variable did not

produce a significant partial F ratio at any time period. It did, how­

ever, reach a significance level of .0503 at Time II which suggests some

degree of importance.

Among the activity items, finance and encyclopedia use were sig­

nificant discriminating variables at Times I and II. Only these two

variables were used in the discriminant function at Time I, and computer

use was added to the discriminant function at Time II. Again, a signifi­

cant function was not produced from the list of activity scales at

Time II.

Based on these findings, the decision was made to include any scale

item that achieved a significant partial F ratio at any time period in

the investigation of research questions #5 and //6. By using this more % liberal Inclusion criteria, all of the variables that were significant

discriminators at any time period could be evaluated in the final analysis

procedures. The exception to this rule was the inclusion of the con­

sumer creativity scale. Since this scale achieved a significance level

of approximately .05 at Time II, and was important to this research for 140 theoretical reasons, it was also included in the final analytic proced­ ures. Thus, the following scale items were included in the analysis of research questions 1(5 and #6: relative advantage, compatibility, TV evaluation, shopping, finance, encyclopedia use, computer use, and con­ sumer creativity.

Research Question 1(5

This research question dealt with the ability of the aforementioned reduced scale list to successfully discriminate between potential adopters and non-adopters at three points in time. For each time period, a two group stepwise discriminant analysis was run with price sensitivity as the dichotoraous dependent measure, and the scales of relative advantage, compatibility, TV evaluation, shopping, finance, encyclopedia use, com­ puter use, and consumer creativity as the discriminating independent variables. Again a partial F significance level of .05 was a prerequi­ site for the inclusion of any variable into the derivation of the dis­ criminant function.

A significant discriminant function was found at each time period using this scale list. Table 31 reports the results relative to research question 1(5. The magnitude of the overall F ratios in each of these discriminant functions indicated that these functions were highly signifi­ cant. The degree to which the discriminant function correlated with each of the two groups was measured by the canonical correlations.

These correlation values were moderately high (Time I, r* = .64; Time

II, r* = .79; Time III, r* = .68) for each of the discriminant functions.

When squared, these values also indicate the proportion of variation in the discriminant function explained by the groups (r^). Clearly the TABLE 31

SUMMARIES OF THE TWO GROUP DISCRIMINANT ANALYSIS AT TIMES I, II, AND III

Time Canonical Period Eigenvalue Correlation R2 Wilks' Lambda Chi Square Overall F df Probability

I .6827 .64 .41 .5943 39.81 17.296 3/76 .001 M3 CM

II 1.6085 .79 • .3834 74.79 30.965 4/77 .001

III .8687 .68 .46 .5351 48.77 33.883 2/78 .001 142

function at Time II was the strongest in that it resulted in the largest overall F (F - 30.965, df = 4/77, p K. .001), and also produced the

largest canonical correlation coefficient.

As a supplementary indication of the usefulness of each discrimin­ ant function, classification tables were produced at each time period.

This procedure classified each of the cases used in the discriminant

procedure into either the potential adopter or potential non-adopter

group according to each case's overall discriminant score. Lachenbruch

(1975, pp. 29-36) has, however, pointed out that the percent of correctly

classified cases may be overestimated when the same cases that were used

to derive the discriminant function are also used for classification

purposes. Although a split-sample technique may have solved this prob­

lem, the limited number of available cases in this research prohibited

the use of this procedure. Nevertheless, the results of the classifi­

cations, presented in Table 32, indicated that each function was extremely

successful in predicting group membership (total percent correct -

81.25, 92.77, and 85.37 for Times I, II, and III, respectively).

Research Question #6

To understand the relative importance of each independent variable

used in the preceding discriminant functions, two statistics were of

primary importance. First, those variables which produced a signifi- t cant partial F at each time period, were those variables which best

discriminated between potential adopters and non-adopters.

Second, the structure coefficients of these variables gave a

straightforward indication of each variable’s relationship to the dis­

criminant function. Since these structure coefficients are actually TABLE 32

RESULTS OF THE CLASSIFICATION ANALYSIS AT TIMES I, II, AND III

Predicted Group Membership Total Number of Potential Potential Percent Percent Time Period Actual Group Cases Adopter Non-Adopter Correct Correct

Potential Adopter 44 34 10 77.3 I 81.25 Potential Non-Adopter 36 5 31 86.1

Potential Adopter 30 28 2 93.3 II 92.77 Potential Non-Adopter 53 4 49 92.5

Potential Adopter 24 20 4 83.3 III 85.37 Potential Non-Adopter 58 8 50 86.2 144

bivariate correlations, their coefficient values were not affected by

inter-item correlations among the set of discriminating variables. Thus,

a large structure coefficient indicated that the variable was carrying

nearly the same information as the discriminant function, which in turn was a reflection of that variable's discrimination power (Klecks, 1979,

pp. 31-35). Tables 33 and 34 present the partial F significance levels

and structure coefficients for each independent variable scale item at

each point in time.

The importance of the perceptual dimensions (relative advantage and

compatibility) is apparent at all time periods. Although significance was not reached for relative advantage at Time I, it was still ranked

second in terms of its structure coefficient for that time period. Both

relative advantage and compatibility were ranked either first or second

by the structure coefficients at each point in time. The absolute

magnitude of these coefficients was another indication of their overall

importance in relation to the other variables.

Three other variables produced significant partial F ratios at

various points in time. At Time I, finance and computer use entered

into the discriminant function as significant variables. At Time II, TV

evaluation and encyclopedia use produced significant partial F's. At

Time III the two previously mentioned perceptual dimensions of relative

advantage and compatibility were the only two variables entered. The

reader is referred to Tables 28, 29, and 30 for the means and univariate

F ratios for each of these significant variables at each point in time. 145

TABLE 33

PARTIAL F SIGNIFICANCE LEVELS AT TIMES I, II, AND III

Partial F Significance Levels

Scale Item Time I Time II Time III

Relative Advantage .15 .00* .00*

Compatibility .00* .00* .01*

TV Evaluation .27 .04* .94

Shopping .16 .49 .74

Finance .01* .14 .14

Encyclopedia Use .15 .01* . 10

Computer Use .01* .39 .55

Consumer Creativity .64 .86 .75

Discriminant function based on these significant scales. 146

TABLE 34

STRUCTURE COEFFICIENTS AT TIMES I, II, AND III

Structure Coefficients

Scale Item Time I Time II Time III

Relative Advantage -.39 (2)a -.81 (1) .92 (1)

Compatibility -.78 (1) -.77 (2) .84 (2)

TV Evaluation -.11 (7) -.24 (4) .26 (3)

Shopping -.20 (4) -.21 (5) .18 (5)

Finance -.32 (3) -.01 (8) -.03 (8)

Encyclopedia Use -.05 (8) -.27 (3) -.06 (7)

Computer Use .14 (6) .03 (7) -.16 (6)

Consumer Creativity .20 (5) •17 (6) -.26 (4)

Rank order of scale importance. CHAPTER VI

DISCUSSION

This research has reported on a prototype viewdata service called

Channel 2000, In addition to the description of the mechanical and system aspects of the Channel 2000 service, this research made use of diffusion theory in an effort to guide the data collection and inter­ pretation procedures reported herein.

Specifically, diffusion theory was subdivided into two general areas of Investigation called micro and macro analytic approaches. A macroanalytic approach was characterized by those diffusion studies which examined the processes of social interaction, and the patterns of diffusion which resulted in the structural and functional changes of society. A microanalytic approach, however, was described as being concerned with the attributes of the innovation, the attributes of the adopting unit, and the degree to which these factors interacted in the adoption decision process,

Accordingly, due to the preliminary nature of videotex as an inno­ vation, this research was predisposed towards the study of Channel 2000 within a microanalytic diffusion perspective. This theoretical frame­ work directed this research towards two major areas of investigation.

First, based on the research and suggestions of Rogers (1976),

Rogers and Shoemaker (1971), Ostlund (1969a, 1969b, 1972, 1973, 1974),

and McCain and Wall (1977), the attributes of Channel 2000 were examined

according to the perceptual dimensions of relative advantage,

147 compatibility, conplexity, observability, trialability, and perceived

risk. Because these perceptual dimensions have been reported by others

to have such a dramatic impact on the adoption decision, this research

was designed to measure these perceptions at three points in time.

Furthermore, since the attributes of an innovation can only be defined

by those individuals who will in fact either adopt or reject the innova­

tion, the innovation perceptions as determined by the Channel 2000 users

were viewed as a critical ingredient in the understanding of why Channel

2000 was either accepted or rejected by the test population. Thus, one

principal concern of this research was an investigation of Channel 2000

attributes as defined by the test population’s evaluation of Channel

2000 along six perceptual dimensions. These perceptual findings were

considered to be important not only for theoretical reasons, but also

for the pragmatic aspects of developing a second generation Channel

2000 system that would have an increased adoption potential.

The second area of investigation looked at the important discrim­

inating characteristics of potential Channel 2000 adopters and non­

adopters according to a series of perceptual, life style, creativity

and sociodemographic scales. Here the emphasis was clearly upon the

attributes of the adopting unit, and the ability of discriminant analysis

to statistically differentiate two groups of Channel 2000 users. Thus, i as a form of segmentation strategy, in coordination with, and guided by

diffusion theory, an attempt was made to profile the most probable

adopters of Channel 2000. Although the generalizability of these find­

ings was somewhat restricted, the social and economic implications

inherent to the emergence of videotex technology warranted preliminary 149 research in this area. Furthermore, it was suggested that these findings could be compared to similar innovator profiles that have previously appeared in the diffusion literature. These Channel 2000 innovator pro­ files could therefore not only contribute to an increased understanding of the adoption process, but could also serve as an initial understanding of the potential videotex adopter.

In summary, the culmination of the reported findings on the per­ ceptions of Channel 2000 at three points in time, and the statistical discrimination of potential adopters and non-adopters, should be viewed as the two primary contributions of this research report. The embryonic state of videotex inquiry necessitates an exploratory approach tempered by an available body of theoretical knowledge which concerns itself with innovations, diffusion, consumer behavior, and communication. The research questions that were developed for the Channel 2000 test and the discussion which follows attempt to place the Channel 2000 findings within such an exploratory framework.

The Perceptions of Channel 2000

Prior to a discussion of the findings relative to the Channel 2000 user perceptions (research questions it 1 and if2), it is first important

to understand the relevance of these perceptions in terms of the poten­

tial adoption or non-adoption of an innovation such as Channel 2000. The perceptual dimensions of relative advantage, compatibility, complexity, observability, and trialability were originally developed by Rogers and

Shoemaker (1971) for the purpose of characterizing various innovations.

These innovation characteristics were appropriately defined according to

the perceptions of potential adopters. As Rogers and Shoemaker (1971, 150 p. 138) aptly point out: "Like beauty, innovations exist only in the eye of the beholder." Thus, it has been hypothesized in diffusion theory that the potential adopter's perceptions of innovation characteristics will in effect determine whether that individual either adopts or rejects an innovation.

The perceptual characteristics of Channel 2000 as measured by this research,■can therefore be beneficial in several regards. First, user perceptions can give a straightforward indication of those areas in which Channel 2000 was perceived in a favorable manner, both prior to the introduction of the innovation, during the Channel 2000 test, and at the conclusion of the three month trial period.

Second, these perceptions can be related to the potential adoption of Channel 2000 for the purpose of arriving at an increased understand­ ing of why people became either potential adopters or non-adopters of

this new technology. And finally, these findings can provide additional

information in regards to the practicality of using innovation percep­

tions to predict potential adoption of videotex technology.

Research questions #1 and #2 asked the question, what were the per­

ceptions of Channel 2000 and how did they change at three points in

time. The Channel 2000 research findings indicated that initially

Channel 2000 was perceived quite favorably by the test participants «

along all the perceptual dimensions. Apparently the brief descriptions

of Channel 2000 given to the test participants prior to the start of the

test, elicited strong positive response to the viewdata concept. How­

ever, after having Channel 2000 in their homes for approximately two to

three weeks, these perceptions experienced a significant decrease among 151 all the perceptual dimensions except complexity. This suggests that most of the Channel 2000 attributes as measured by user perceptions, did not measure up to the test population's expectations, and also suggests that Channel 2000 succeeded in its objective of being an easy to use viewdata system.

The most noticable decline among the perceptual scores was for the dimension of relative advantage, which is defined as the degree to which an innovation is perceived as being better than the idea it supercedes

(Rogers and Shoemaker, 1971). Some Channel 2000 users did not feel that the Channel 2000 system offered them many advantages beyond their current library, banking and information seeking behaviors. To a lesser extent, the users also experienced disappointment along the dimensions of compatibility, observability, trialability, and perceived risk.

According to diffusion theory, this means that Channel 2000 was also perceived at Time II as being less consistent with the users' exist­ ing values, past experiences, and needs (compatibility); more difficult to communicate and describe to others (observability); less likely to be made available In the future for experimentation on a limited basis

(trialability); and containing a higher amount of social and financial risk (perceived risk) than at its inception at Time I. Although the magnitude of these mean perceptual scores does not suggest that Channel

2000 was a total failure, the significant drop in these scores at

Times II and III does present some rather pessimistic evidence in regards to Channel 2000*s overall performance record.

At Time III, the dimensions of perceived risk and trialability did significantly increase from Time II, and they almost achieved their 152 original Time I mean value. As might be expected, with three months of trial experience, an innovation should be perceived as being more trial- able and also less risky. The remaining dimensions of relative advantage, compatibility, and observability, however, never returned to their original Time I mean value, which further supports the contention that

Channel 2000 was not as compatible, observable, or advantageous as originally expected prior to the introduction of the innovation. Clearly,

Channel 2000 users experienced a certain amount of product disappoint­ ment which can be an extremely dangerous phenomenon for any newly intro­ duced innovation.

Rogers and Shoemaker (1971, pp. 116-118) have coined the term "dis­ enchantment discontinuance" to describe the process whereby an adopter makes the decision to cease using a product or service as a result of dissatisfaction with its performance. This process of discontinuance can be explained by a lack of perceived relative advantage, or when the inno­ vation is less compatible with the users' beliefs and past experiences.

The levels of consumer videotex expectation as measured by the Time I

perceptions may help explain the initial disappointment with the Channel

2000 system. If consumers preconceive videotex systems as being mysti­

cal, do-all panaceas, then all videotex systems will continue to suffer

from the same "disenchantment" that the Channel 2000 users experienced.

If, however, newly introduced videotex systems at least meet, or exceed

these preconceptions, then continued consumer satisfaction should be

achieved.

Up to this point it has been suggested that innovation perceptions

directly influence the adoption decision. If this is the case, then a 153 decrease in perceptual scores should correspond to a decrease in overall potential adoption. Furthermore, significant mean score differences between the perceptions of potential adopters and non-adopters should be

found across all time periods. Although these two research hypotheses were not completely addressed by this research investigation, several

findings did emerge as a by-product of other statistical procedures.

First, if potential adoption is defined as being willing to pay $12.50

per month for Channel 2000 services, 31 percent of the users were poten­

tial adopters at Time I; 9 percent, at Time II; and 8 percent, at Time

III. This finding initially suggests that decreasing perceptual scores

corresponded to decreasing levels of potential adoption.

Second, the univariate F ratios calculated by the discriminant analysis procedures (and reported in Tables 28, 29, and 30) indicated

that potential adopters and non-adopters do significantly differ on

several of the perceptual dimensions. At Time I, differences were found

on the dimensions of relative advantage, compatibility and perceived

risk. At Times II and III, mean differences were found on these same

dimensions, in addition to the dimensions of observability and trial­

ability. The direction of these differences is consistent at the three

points in time. Potential adopters perceived Channel 2000 more favorably

than non-adopters. Of course, as previously shown, the available number

of potential adopters drops to a disappointingly low level for Times II

and III.

The evidence presented thus far, in regards to the relationship

between innovation perceptions and the process of potential adoption,

should be considered as an introduction to the discussion which follows. 154

To examine this phenomenon correctly, it was necessary to employ multi­ variate statistical techniques that could account for the relationships between many independent variables simultaneously, and that could also assess each variable's individual contribution to the discrimination of potential adopters and non-adopters. In this fashion, the relative importance of these innovation perceptions could be determined, and those perceptions which demonstrated a high degree of discrimination could be further explored and examined.

The degree to which innovation perceptions related to one another

(research question //3), and also related to life style and consumer creativity (research question it4), will be discussed after the signifi­ cance of these perceptions has been clearly delineated. This juxta­ position of research questions is best suited to the sequential organiza­

tion of the following discussion.

Discriminating Between Potential Adopters and Non-Adopters

The obvious intent of research question it5 was to explore the poten­

tiality of actually finding some identifiable differences between poten­

tial adopters and non-adopters of a videotex technology. Although

diffusion theory would suggest that these two groups should differ

according to variables such as sociodemographic indices, life style,

participatory activities, innovation perceptions, and creativity pro­

pensity, no one has provided any evidence in regards to the differences

between potential videotex adopters and non-adopters. Research ques­

tion it5 simply asks if it is possible to derive a mathematical function

that can describe two statistically distinct groups of potential Channel

2000 adopters and non-adopters based on the theoretically relevant 155 variables described in this report. This was the first step in deter­ mining which of the variables were most important in the concluding dis­ crimination procedure.

The results of the discriminant functional analysis at all three points in time suggested that potential Channel 2000 adopters and non­ adopters can be significantly differentiated. This implies that the eight variables used in this procedure quite adequately described the differences between these two groups, and can furthermore be used to predict subsequent Channel 2000 tendencies for other groups of consumers.

Although all three discriminant functions were significant at each point in time, the overall F ratios for each function noticably in­ creased at Times II and III. It could be hypothesized that this increased discriminatory significance was a reflection of the decisive­ ness obtained through experience with the innovation. In other words, the ability to discriminate between potential adopters and non-adopters of Channel 2000 was more readily achieved after the potential adopting group had the opportunity to formulate very specific and definite opinions about their own adoption probability. Although this may be the case, it is interesting to note that even at Time I, the discrimin­ ant function could accurately classify 81 percent of the Channel 2000 users prior to the introduction of the innovation. Thus, the incre­ mental gain in functional significance obtained through user experience

is somewhat limited by the strength of the discriminant function that was derived at Time I.

These findings support the basic 'assumption in diffusion theory

that Innovators or early adopters are different from laggards or the 156 late majority. This research has, however, obviously "loaded the dice" in favor of these differences, since only the opposite ends of the adopter continuum were examined. Had the discriminant functions been based on the criterion adopter categories of innovator and early adopter, these differences may not have been so pronounced. Nevertheless, the change agent who desires to facilitate the diffusion of Channel 2000 to a mass population, is in a much better position in terms of marketing strategy, when he is armed with information about the characteristics of this early adopter segment. This consumer strategy not only follows the theoretical guidelines of diffusion theory, but it also makes good mar­ keting sense.

The most critical question inherent to this strategic planning is, of course, what are the most important discriminating characteristics of this potential early adopting segment? Also, do these discriminat­ ing characteristics vary in importance over time? These two questions were the essential ingredients of research question if6.

Characterizing the Potential Channel 2000 Adopter

To understand the relative importance of each independent variable used in this research, it was first necessary to evaluate the results of the preliminary discriminant analysis procedures. In this procedure, variables which constituted a singular construct only, were tested against one another for their own discrimination power in the discrimin­ ant analysis. For any variable to be included in the final discriminant runs, it was necessary for the variable to first establish its discrim­ inant strength within its own construct category. This discriminant strength was evaluated by the presence or absence of a significant 157 partial F ratio observed in discriminant step-wise procedures at each point in time. Any variable which achieved significance at any point

in time was included in the final discriminant procedures.

Innovation Perceptions

Among the perceptual variables, the dimensions of relative advantage and compatibility were consistently the most important discriminating

dimensions. Other dimensions such as perceived risk, observability and

trialability resulted in significant univariate differences, but they were not significant in the multivariate model. This could be in part a

reflection of the powerful discriminating ability of relative advantage

and compatibility. With these two dimensions only, a very significant

discriminant function was found at each point in time (see Tables 28,

29, and 30). The inability of perceived risk, observability, and trial­

ability to enter into the multivariate model may also be a reflection of

their low reliability coefficients, or their correlation to the dimen­

sions of relative advantage and compatibility.

This initial finding in regards to the importance of relative advan­

tage and compatibility is in agreement with other research reported in

the diffusion literature. Ostlund (1974) also found perceptual dimen­

sions of relative advantage and compatibility to be by far the most

important discriminating variables in his two studies of new product

adoption. Ostlund concluded that relative advantage and compatibility

were the two most powerful predictors of innovative behavior. The

findings of this research suggest that among all six of the perceptual

dimensions, relative advantage and compatibility are the most useful of

the perceptual dimensions in terms of their ability to discriminate 158 between potential adopters and non-adopters of Channel 2000.

Life Style

Among the life style items, several variables emerged as important preliminary discriminators, and some unexpected relationships between potential adopters/non-adopters and these variables emerged. First, among the life style variables, only shopping and TV evaluation were entered into the discriminant function at any time. A positive response pattern to the shopping life style dimension indicated that the individual had a propensity to try new products and use advertising information to make these decisions. A positive response pattern to the TV evaluation dimension of life style indicated that the individual tended to use tele­ vision for news and entertainment, and voiced an interest preference for television over magazines.

The fact that new product trial and experimentation was a character­ istic of potential Channel 2000 adoption is consistent with the theoreti­ cal framework of diffusion research. Quite clearly, the fact that a person does in fact try a new product, immediately classifies him or her as an early adopter in classic diffusion theory. A more appropriate con­ ceptualization of the shopping construct, however, might be what Rogers

(1962) has termed "venturesomeness." Rogers originally used this term to characterize agricultural innovators, "The major value of the inno­ vator is venturesomeness. He must desire the hazardous, the rash, the daring, and the risking" (p. 169). Robertson and Kennedy (1968) used this construct in a discriminant analysis of consumers' reactions to a new home appliance. Their findings also indicated that the degree to which an individual would try new products and assume new product risks 159 accounted for most of the innovative behavior difference between adopters and non-adopters. This notion of new product trial and risk appears to be an important discriminating characteristic for the identification of potential Channel 2000 adopters.

The TV evaluation scale was a significant discriminating variable in the preliminary discriminant analysis at Times II and III only. The reader is reminded that a high score on this scale indicated a certain degree of television dependency for news and entertainment. It is very interesting to note that at Time I, the potential adopters scored lower on this scale than non-adopters (although a significant difference was not obtained). However, at Times II and III when this variable entered into the discriminant analysis with significant partial F ratios, these mean scores were reversed.

Not only is TV evaluation one of the most important discriminating variables among the life style list, but it also provides some very

interesting and unexpected insight to the potential adoption of Channel

2000. Those people who were classified as potential Channel 2000 adopters at Times II and III tended to rely more on television for news and entertainment. One might assume that Channel 2000 would be more

favorably received by people who chose not to be entertained or informed

by standard television, and yet the opposite is true. Channel 2000 may have been viewed as a complementary addition to the television experi­

ence. In other words, the act of organizing one's life around the tele­

vision viewing habit may have predisposed this group of users to expand

and enhance this experience with the capabilities of Channel 2000. In

light of the important perceptual dimensions of relative advantage and 160 compatibility, Channel 2000 may have been perceived as being more com­ patible with the potential adopter's current television based life style, and also perceived as offering certain extended benefits beyond the currently available commercialized television product. This finding has important implications for videotex technology and will be discussed further in a subsequent sub-heading.

Consumer Creativity

The construct of consumer creativity was introduced as a personality factor which would positively relate to the potential adoption of

Channel 2000. Several observations emerged from the preliminary dis­ criminant analyses. First, at all three points in time, the potential

Channel 2000 adopter received a lower mean score on this scale than did

the potential non-adopter. Although a significant difference was found at Time I only, the directionality of these differences was consistent across time. This suggests that the creative consumer tended to be more

critical of Channel 2000 as evidenced by their decision to become a non­

adopter. Second, the creativity scale proved to be an ineffective dis­

criminator between potential adopters and non-adopters at all points in

the preliminary discriminant procedures. Although a partial F signifi­

cance level of exactly .05 was obtained at Time II, the scale did not

enter into the final discriminant functions when all other significant

scales were entered.

In regards to the unexpected negative relationship between consumer

creativity and Channel 2000 potential adoption, several plausible explan­

ations can be derived. Leavitt and Walton (1977) describe the creative

consumer as an individual who Is more critical of both information 161 sources, and the idea (product), but is someone who is also willing to experiment, or to engage in suspension of disbelief. This critical aspect of the creative consumer may account for the inverse relationship found in this research.

Hirschman (1980, p. 286) has defined consumer creativity as "the problem-solving capability possessed by the individual that may be applied toward solving consumption-related problems." It is hypothesized by Hirschman that this problem solving capacity of the creative consumer allows this individual to evaluate the innovation in terms of the sim­ ilarities and differences among similarly adopted products within the same product class. Furthermore, the creative consumer is better equipped to mentally construct a consumption problem environment, and relate the innovation’s attributes to its ability to solve problems within this environment.

Hirschman concludes that high levels of consumer creativity do not necessarily lead to increased new product adoption, but to more competent new product evaluation. The creative Channel 2000 consumer may have compared Channel 2000, as an innovation, to other computer-like innova­ tions, and concluded that the attributes of Channel 2000 did not lend themselves to problem solving within their constructed consumption en­ vironment. In a word, it appears that Channel 2000 was critically evaluated by the creative consumer as an inferior product, and the decision was made to reject this innovation.

Life Style Activities

The activity scales used in the preliminary procedures were selected

on the basis of their hypothesized relationship to Channel 2000. For 162 example, people who frequently use the library may adopt a library based videotex earlier than those who use the library infrequently.

Of the five activity factors, the degree to which a user conducted bank­ ing business (finance), and used an encyclopedia, significantly dis­ criminated potential adopters and non-adopters of Channel 2000 at Times

I and II only. At Time II, the activity of computer use also entered as a significantly discriminating variable.

The discriminatory importance of the finance scale could be a reflection of the presence or absence of the Channel 2000 banking services in a user’s home. Because only 35 users signed up for the banking services, it might be hypothesized that those who had this serv­

ice would be more likely to become potential adopters than those who did not. A subsequent chi square test partially supports this conclusion.

Those users who did have banking services were found to be more likely classified as potential adopters, than those who did not at Time II 2 (X = 8.56, df = 2, p < .01), and significance at the .05 level was 2 almost reached at Time I (X = 5.40, df = 2, p < .10). This evidence

partially supports the conclusion that those individuals who frequently

conducted banking activities, and also signed up for the Channel 2000

banking services, could be significantly discriminated as potential

Channel 2000 adopters at Times I and II.

The frequency of encyclopedia use was an important discriminating

characteristic at Times I and II, but not at Time III. The fact that

encyclopedia users were not more likely to be considered potential

adopters at Time III might suggest that the product attribute of an

in-home electronic encyclopedia was less important at the conclusion 163 of the test. Here again, a certain amount of innovation disenchantment with the video encyclopedia may have eliminated several encyclopedia users from the potential adopter group at Time III, which would account for its diminished univariate and partial F differences.

Although it might be expected that those individuals who operate and use computers would be more likely to fall into the potential adopter category, results Indicated quite the opposite. Across all time periods, the computer use mean scores were lower for the potential adopter groups, than for the potential non-adopter groups. Although significant differences in a univariate sense were not found, this factor served as an important discriminating factor at Time II. It may be that individuals with some degree of computer sophistication perceived Channel

2000 as an elementary and restrictive tool. In light of the simplistic keypad configuration of Channel 2000, and tedious hierarchial search patterns employed, this conclusion may be well justified.

It is interesting to note that at Time III none of these activity scales reached a level of significant discrimination. This finding suggests that these activity scales diminished in discrimination power over time. The differences in the activity participation by potential adopters and non-adopters also diminished with Channel 2000 experience.

It should be pointed out, however, that the mean differences in library use, encyclopedia use, and finance for the potential adopter and non­ adopter groups were significant at the .10 level at Time III.

Demographics

Finally, the sociodemographic variables were unsuccessful in identifying differences between these two groups at all points in time. 164

Ostlund (1974) and Robertson and Kennedy (1968) have reported similar findings in this regard. Although other diffusion studies have found the early adopter to be more educated, and have higher social status or be wealthier, the findings of this research do not support these con­ tentions. This could in part be a reflection of the somewhat upscale and well-educated nature of the Channel 2000 sample. The lack of a truly heterogeneous cross section of social class may have contributed to the lack of discrimination found in this variable set.

The Final Discriminant Analysis

In the final discriminant analysis procedures those variables which were significant in the preliminary analysis were allowed to com­ pete with each other, at each time period, for the derivation of an overall discriminant function. In this manner, the relative strength of each variable across construct category could be established at each point in time.

The results of this analysis further supported the ability of relative advantage and compatibility to discriminate between the two groups of test participants. An examination of the structure coeffi­ cients across the three points in time revealed that these two perceptual dimensions have correlation coefficients that are twice as large as the third ranked variable. Compatibility was the most important discrimin- ator at Time I, whereas relative advantage was ranked first at Times

II and III.

The diminished importance of the relative advantage at Time I suggests that prior to the introduction of Channel 2000, users may not have been as aware of Channel 2000 in terms of this innovation attribute, 165 as they were at Times II and III. However, in light of this dimension's importance at Times II and III, change agents may be well advised to increase an unknowing public's awareness of videotex innovations in terms of its relative advantage early in the diffusion campaign. Had this strategy been employed for the Channel 2000 test, the dimension of relative advantage may have been the most important discriminating vari­ able at all points in time.

Of the two life style dimensions, only the TV evaluation scale resulted in a significant partial F at Time II. Here again, once the test participants had been given the opportunity to experiment with

Channel 2000, those users who were more active users of television for information and entertainment were also likely to be classified as potential Channel 2000 adopters. Even at Time III when the TV evaluation scale did not achieve the appropriate significance level to enter into the discriminant function, it was still ranked third among the structure coefficients. These results initially suggest that attitudes and behav­ iors toward television are a potentially important factor in the Channel

2000 adoption decision process.

Shopping, or venturesomeness, did not significantly contribute to a discriminant function at any point in time in the final analysis.

Its relation to the discriminant function is modest (r = .18 to .21), and it most likely was overpowered by the ability of the perceptual dimensions to differentiate the test participants. It should still, however, remain as an important consideration in future videotex research because of its theoretical value to the understanding of the adoption process. 166

The activity item of finance proved to be an important discrimin­ ator at Time I. For those people who frequently engaged in banking activities, Channel 2000 may have been initially perceived as providing them with a more efficient means of conducting their banking business.

Since the mean differences between potential adopters and non-adopters along this dimension of finance decreases at Times II and III, this pre­ conception of Channel 2000 may have diminished with time. Newly intro­ duced videotex systems that stress the relative advantage of their system along this financial dimension, may increase their available potential adoption population.

The degree to which test participants used an encyclopedia proved to be a significantly discriminating variable at Time II. Thus, the usage of services similar to what Channel 2000 offered, proved to be an important factor in identifying the potential adopter of Channel 2000.

Taylor (1977) has suggested that product usage is likely to have either one of two effects on innovative behavior. If the product/service is important to the consumer, its importance might inhibit potential adoption because of the risk involved in trying the new product. Or, heavy usage of the product/service might stimulate new product trial because of a desire to improve upon the results of current usage benefits.

When Taylor examined usage and adoption patterns for 11 innovative products, in 10 of 11 instances, those individuals who purchased more of the product class tried the newer products within the same class earlier than those who were light users. Taylor concluded that new product development should be directed towards heavy users of the prod­ uct class since they best represent the early adopter segment. 167

The results of the Channel 2000 test suggest that initially, fre­ quent banking activities and encyclopedia use are important discriminat­ ing characteristics of the potential adopter. Therefore, it might be recommended that marketing activities be directed towards these two user segments when the innovation is first introduced, since these two groups represent the early adopter segment for Channel 2000 at that point in time.

Unfortunately at Time III, significant differences between the potential adopter groups along these activity dimensions were not found.

Again it could be suggested that many of these encyclopedia users and heavy bankers were dropped from the potential adopter category at Time

III due to their overall disappointment with the Channel 2000 services.

An opposite pattern, similar to that already discussed in the pre­ liminary analysis, was found for the activity of computer use. In this case, at Time I, the degree to which test participants had more frequent exposure to computers helped identify them as a potential non-adopter.

According to Taylor's hypotheses, this group may have felt that Channel

2000 could not offer them additional benefits beyond their current com­

puter behaviors and therefore, Channel 2000 was rejected as a viable

alternative.

Since the consumer creativity scale had not been very useful as a

discriminating characteristic of potential Channel 2000 adoption in the

preliminary analysis, it was not expected to achieve significance in the

final analysis. This expectation was upheld, and this scale failed to

provide a significant amount of multivariate discrimination. Although

it was interesting to note that an opposite relationship between Channel 168

2000 potential adoption and consumer creativity was found, other re­ searchers are cautioned that this scale may not be a useful discrimin­ ator in future videotex studies.

In summary, the potential adopters of Channel 2000 could be best identified at all points in time by the degree to which they felt Channel

2000 offered them advantages over the ways in which they currently be­ have, and the degree to which Channel 2000 was perceived as being com­ patible with their existing values, needs, and past experiences. Prior to the introduction of the innovation, the degree to which a test par­ ticipant balanced his or her checkbook and called the bank about that account, proved to be an important discriminating factor in favor of potential adoption. Those individuals who had frequent encounters with computers were identified as potential non-adopters of Channel 2000 at this point in time.

Following a two week trial period, potential adopters were dis­ criminated from non-adopters on the basis of these two groups' evalua­

tion and use of television. Furthermore, the extent to which an individ­ ual used an encyclopedia was also found to be an important discriminating

factor.

At the conclusion of the three month trial period, all the afore­ mentioned variables, with the exception of the two perceptual dimensions

of relative advantage and compatibility, lost their distinctive discrim­

ination importance. The strength and consistency of these perceptual

dimensions in terms of their discrimination power at three points in

time, suggests that these constructs are very important determinants of

videotex adoption process. 169

The Relationships Among Innovation Perceptions

Research question if3 dealt with the degree to which the six innova­ tion perceptions used in this research related to one another over the three points in time. Although the results presented thus far have demonstrated the importance that these dimensions have in the potential adoption process, there is little evidence in regards to their conceptual distinctiveness.

Results indicated that all of the perceptual dimensions produced a modest positive correlation among each other at all points in time. It appears that the degree to which one dimension was viewed positively, corresponded to the positive perception of Channel 2000 along a different dimension. Most noticably, the dimensions of relative advantage and compatibility tended to exhibit the strongest correlation coefficients consistently across time.

Although this preliminary analysis of correlation coefficients can only shed light on the bivariate relationships between one pair of variables at a time, the conceptual distinctiveness of these innovation perceptions may be showing a considerable degree of overlap. The next recommended stage of analysis would be an analysis of these variables

in a multivariate sense through factor analytic techniques.

The Relationships Between Innovation Perceptions

and Life Style, Activities, and Consumer Creativity

The objective of research question if4 was to discover the relation­

ships between the innovation perceptions of Channel 2000 and the remain­

ing independent variables so that some additional meaning could be

related to each of these perceptual dimensions. If, for example, a 170 strong positive correlation was observed between relative advantage and library use, it could be hypothesized that frequent library users also perceived Channel 2000 as being able to offer them some identifiable advantages. In other words, by correlating each perceptual dimension with another user characteristic, an increased understanding of why

Channel 2000 was perceived as either possessing or lacking this attribute can be obtained. The objective of this procedure was to add dimension­ ality to the seemingly important perceptions of Channel 2000.

Unfortunately, none of the correlations achieved a level of conclu­ sive importance. Although some tendencies were observed, the relation­ ships were not strong enough to warrant definitive conclusions (see

Table 27). It is interesting to note, however, the negative correlations that were exhibited between consumer creativity and all the perceptual dimensions. This finding can serve as an additional piece of evidence in support of the critical nature of the creative consumer. The more creative the individual was, the more negatively Channel 2000 was perceived along all its attributes.

At this stage of diffusion theory, a very appropriate research strategy that could be used to explore the underlying explanations of positive or negative perceptual evaluations, would be to conduct a

series of multiple, or canonical correlations between the perceptual dimensions and a series of product specific attitudinal items. Although

this research has established that perceptual dimensions are a critical

element in the potential adoption of Channel 2000, far too little is

known about why Channel 2000 was perceived in a somewhat negative light.

This should obviously be a second stage of this analysis. 171

Implications for Diffusion Theory

First and foremost, this research has tried to re-emphasize the importance of innovation perceptions in the adoption process. Since these perceptions are unique and developed on an individual level, they can be extremely powerful tools in the analysis of innovative behavior.

A diffusion theorist interested in studying the phenomenon of innovative behavior would be well advised to identify innovators by their innovation perceptions rather than by their personal characteristics. Furthermore, careful attention should be specifically given to an individual's evalua­ tion of an innovation according to its attributes of relative advantage and compatibility. These two dimensions of innovation perception, when evaluated by a group of potential adopters, should best discriminate between those who choose to adopt the innovation, and those who do not.

The value of these perceptions in diffusion theory can also be taken one step further. The information obtained from innovation perceptions allows a change agent to modify an innovation in a manner that would elicit more favorable evaluative perceptions at a later point in time.

Through a series of sequential modifications, the innovation could be systematically altered until a desired level of potential adoption was achieved. The impact that these perceptions have on the adoption process suggests that this could be a very astute strategy indeed.

The importance of perceptual innovation evaluations in the adoption process should also create an increased appreciation among diffusion researchers, of the need for more "innovation centered" research. Dif­ fusion theory should be expanded to "include comparative studies on the characteristics of various innovations and how these characteristics 172 affect subsequent perceptions and adoption decisions. Additional innova­ tion attributes similar to those recommended by Zaltman and Lin (1971) should be included in other diffusion studies so the relationship be­ tween the dimensions of innovations and their rate of adoption can be verified. There seems to be enough available evidence in the diffusion literature to support the contention that those studies which treat all innovations as equivalent units are missing an important element in the adoption process.

The correlations found between the innovation perceptions used in

this research also implies that further conceptual distinctiveness may be needed for these perceptions in future research efforts. These inno­ vation perceptions could benefit from additional factor analyses and measurement reliability estimates.

The fact that significant differences were found between potential

adopters and non-adopters of Channel 2000 is congruent with the general­

ization in diffusion theory, that adopters possess certain character­

istics that differentiate them from non-adopters. This research has,

however, shown that the relative importance of these discriminating

characteristics may vary according to when the measurement was taken in

the adoption process. Diffusion researchers must account for the stage

of the adoption process that the respondent is in before research gen- t eralizations can be confidently established.

In coordination with the adoption stages suggested by Rogers and

Shoemaker (1971) (knowledge, persuasion, decision, and confirmation),

this research found the differential characteristics of the potential

adopter at the knowledge stage, to be somewhat different from the 173 characteristics of the potential adopter at the persuasion or decision stage. Rogers and Shoemaker (p. 160) alluded to these differences in terms of innovation perceptions by stating at the knowledge stage, com­ patibility and complexity should be most important, whereas at the

f persuasion stage, relative advantage and observability should be most important.

This implies that in order to maximize an innovation's rate of adoption, the change agent must understand the relationship between the potential adopters stage of awareness or persuasion, and those characteristics which are critical to the adoption process at that observed stage. If for example in this research, it had been known that the compatibility and financial aspects of the Channel 2000 service were the most critical dimensions at the knowledge stage, promotional efforts which stressed these dimensions may have substantially increased the hypothetical share of the market. This kind of analysis seems to offer a great deal to the understanding of adoption over a specified period of time.

Consumer creativity as a construct in diffusion theory needs further elaboration and definition. Although the efforts of Hirschman (1980) represent a promising trend in this direction, the concept in its cur­ rent form needs more specificity. The findings of the Channel 2000 test represent the first piece of evidence in support of Hirschman1s conclu­ sion that the creative consumer will not always be among the first to purchase an innovation. The creative individuals in the Channel 2000 test population tended to be the most critical users, as evidenced by

their perceptions of Channel 2000. It could be hypothesized, however, 174 that these critical evaluations of Channel 2000 were the result of crea­ tive consumers' tendencies to evaluate the innovation in much greater detail than the less creative individual. These relationships most definitely warrant further consideration.

It is also important for diffusion researchers to very carefully consider alternative measures of adoption in their analysis of new product/service acceptance and rejection. This research considered and examined three measures of potential adoption before arriving at a final criteria. In the case of videotex research, it is especially important for researchers to compare and contrast research results in light of the criterion variable which was selected for the individual study.

Had this research made use of perceived use as the measure of potential adoption, it is quite possible that a very different set of results may have emerged. Other measures such as actual videotex use examined over a period of time may again have produced a contrasting set of conclusions. Because adoption has been defined as the acceptance and continued use of the innovation over time (Robertson, 1971, p. 56), it is recommended that future videotex researchers incorporate such videotex usage rates in their research strategies. Once a baseline data base of usage is established, videotex research may have the potential for making a very significant contribution to diffusion theory.

This research is indebted to diffusion theory as a conceptual

device which helped conceptualize, organize, and interpret what might

have been an uninterpretable research problem. At the very least, this

research has hopefully lent additional support to diffusion theory as an

important tool in the understanding of a very complex and important 175 innovation such as Channel 2000. The beauty of diffusion theory is that it coincides so nicely with other perceptual and learning theories, theories of media effects, marketing segmentation and life style strate­ gies, and an overall understanding of consumer and communication behavior.

Hopefully, future videotex investigations will build upon this important theoretical foundation so that others can better understand why video­ tex became such an enormously successful innovation, or why it was such a devastating failure.

Implications for Videotex Technology

An important implication of this research relates to the fundamental purpose of the Channel 2000 test. This experiment was designed so that an increased understanding of the videotex phenomenon could be under­ stood from the eyes of the Channel 2000 test participants. This research has hopefully initiated a trend in videotex research which emphasizes user perceptions of a new communication technology. This approach has been suggested as the most productive means of obtaining an in­ creased understanding of how people react to, and eventually use video­ tex technology. This research has also provided a very preliminary example of how such an inquiry might be conducted within an actual field setting. The value of this research perspective can be best evaluated in light of the implications and conclusions that can be drawn from the research findings themselves.

The most apparent reason why the innovation perceptual dimensions of relative advantage and compatibility best predict potential and subse­ quent adoption is that these variables appear to penetrate the essential reasoning which underlies the adoption decision process. When these two 176 concepts are expanded into videotex-relevant applications, their immedi­ ate value becomes even more apparent.

Rogers and Shoemaker (1971, p. 138) state quite simply, "relative advantage is the degree to which an innovation is perceived as being better than the idea it supercedes." This concept is normally expressed

in terms of economic profitability, but as will be seen, there are some very important non-economic advantages that can be associated with an

innovation such as videotex.

In terms of monetary advantages associated with Channel 2000, the

potential adopter might have considered such factors as gasoline savings,

book expenditures, reduced postage costs, or educational service fees.

Examples of the non-economic advantages associated with Channel 2000

may have included time savings (which of course could be considered an

economic advantage), actual physical energy saved (as a result of

electronic in-home service), or the perceived educational values that

Channel 2000 offered to adults and children.

These are just a few of the many advantages that could be hypo­

thetically associated with using Channel 2000. Regardless of what

these advantages are, one singular point is very clear. For videotex to

succeed, it must be perceived as replacing a current mode of behavior

with one that is quicker, cheaper, easier, and produces a better end « result. The potential videotex adopter will not be impressed by

industry-based technological break-throughs in videotex hardware and

services if the system itself cannot meet up to the standards Implied by

the innovation dimension of relative advantage. Newly developed videotex

services should therefore very carefully evaluate a potential adopter’s 177 perception of their system along this dimension. To a very large extent, this should determine whether or not a market exists for their product.

One finding that resulted from the Channel 2000 test was the in­ creased importance of relative advantage after the users had been given the opportunity to experiment with the system on a trial basis.

Because videotex technology is a somewhat complex communication, informa­ tion, and entertainment medium, many test participants may have been unable to evaluate Channel 2000 along this dimension of relative advan­ tage until they had obtained enough information about the specifics of its operation.

The obvious implication of this finding is that if a videotex system can be developed that excels along this dimension of relative advantage, potential videotex users should be first given an opportunity to experi­ ment with the system on a free trial basis. Once these potential sub­ scribers have been given the opportunity to see for themselves that the videotex system does in fact offer them both economic and non-economic advantages, theoretically speaking, the probability that these trial participants do in fact become videotex users should be dramatically increased.

The inherent complexity of a videotex system may also suggest that the system itself be presented to potential adopters as a very simplis­ tic concept. This of course relates to the perceptual dimension of com­ plexity, and it appeared that Channel 2000 was perceived as a relatively easy-to-operate and understand innovation. However, the innovation attribute of complexity goes beyond the mere mechanics of the system. 178

Because the Channel 2000 test participants were unable to evaluate the advantages inherent to this innovation at Time I, there was a strong possibility that they were also unable to conceptually understand exactly how the system would deliver all of the services it promised.

If only one simplistic aspect of Channel 2000 had been presented to this group of test participants prior to the start of the test, they may have been in a much better position to evaluate the relative advantages of Channel 2000 at this initially important awareness stage.

Therefore, if the most important aspect of a videotex system in terms of the relative advantage it can afford the potential adopter is presented in a totally comprehensible manner at a very early intro­ ductory stage, the potential adopting group should be able to more easily evaluate the innovation in terms of its relative advantage, which should in turn facilitate the adoption decision process. This strategy is of course directed towards the objective of increasing the speed at which the videotex adoption process takes place. An outstanding videotex service that cannot be readily evaluated in terms of its rela­ tive advantage will be likely to fail in the marketplace.

The perceptual dimension of compatibility consistently demonstrated its relative importance as a discriminating variable in the potential adoption of Channel 2000 at all three points in time. Apart from its t most apparent statistical significance, this innovation attribute should be looked upon as the singularly most important concept in the understand­

ing of videotex adoption and rejection.

The compatibility of an innovation relates to the degree to which

it is perceived as being consistent with the existing values, past 179 experiences, and needs of the potential adopters. The issue at hand is, given this concept’s importance, what values, past experiences, and identifiable needs are most important to the videotex adoption decision?

Several speculative relationships can be hypothesized.

Videotex could offer potential adopters a wealth of information and educational services. These services could be packaged for either adults or children. Individuals who place a high value on more standard forms of education, and those processes associated with standard educa­ tion (for example, finding information from books in libraries), may frown on passive computer-programmed educational services. Some parents may find disdain in their child's ability to locate information for homework assignments by sitting in front of the TV set, keypad in hand.

Others may feel that without videotex services, their child would be missing an important process of socialization within a computerized society.

Some people may find the mechanical, computer-like aspects of video- r tex technology to be too cold, and impersonal, while others may long for

this "distant" approach. The thought of transferring funds electroni­ cally may impinge upon the values associated with a standard, and secure, savings passbook.

The recreation values that an individual has may determine whether

that person stays home on Saturday night playing a videotex version of

"Space Invaders," or goes to the opera. Graham (1956) found that only

24 percent of his upper middle class sample adopted television, whereas

72 percent of the lower middle class population readily accepted this

innovation. Quite clearly, this was a case of contrasting recreational

values. 180

Computer experience was shown to have an interesting effect on potential adoption in this research. In this case, the degree to which

Channel 2000 was perceived to be compatible was based upon a user's previous experiences with other computer-like innovations. It was quite obvious that Channel 2000 did not exceed the standards or criteria established by this experience; thus, the innovation was rejected.

Rogers and Shoemaker (1971, p. 147) offer a plausible explanation by

stating: "Old ideas are the main tools with which new ideas are assessed.

One cannot deal with an innovation except on the basis of the familiar and the old-fashioned.11

This implies that diffusion strategies for experienced computer

users must be different from those developed for non-computer users. It

is important for change agents to remember that each group's evaluation

of the videotex innovation in terms of its compatibility will be based

on a unique set of past experiences.

Another factor which may have a significant impact on how compatible

videotex is with previously established habits and experiences, has to

do with the manner in which videotex information is located and displayed.

Will users who are accustomed to a serendipity search through the Sunday

paper, or library book stacks, find the purposive linear tree-structured

search patterns of videotex cognitively compatible with their past exper­

iences? Will potential adopters find reading from a videotex screen

compatible with their established reading habits? Previously established

habits of media information processing could have a dramatic effect on

the compatibility of newly introduced videotex systems. 181

This research has also provided some preliminary evidence to suggest that those individuals who rely upon television for news and entertain­ ment may perceive videotex as an extension of their current viewing experience. In other words, since television viewing has been incorpor­ ated into the lives of these individuals, a device which has the poten­ tial for further enhancing this video habit should be perceived as a very compatible innovation. The intensity of the television viewing habit dramatizes the implications this finding has for videotex tech­ nology. If this preliminary finding can be supported by other research evidence, entrepreneurs will have additional support for the contention that a potential market for their services exists. An "enhanced" tele­ vision experience promised by videotex services should facilitate the diffusion of videotex technology.

The "bottom line" issue of videotex compatibility has to do with the degree to which videotex meets a need felt by a potential adopter.

A consumer need is normally prefaced by a state of problem recognition which in turn motivates the individual to seek out and compare viable products or services that can fulfill this need (see Engle, Blackwell and Kollat, 1978). As Professor Blackwell of Ohio State University often points out, a consumer buys a drill bit not because he needs the

tool, but because he needs a hole.

At its current stage of development, most consumers are not at the

level of recognizing felt needs that could be solved by videotex tech­

nology. Although most would agree that need assessment is an important

preliminary stage of videotex development, the degree to which these

videotex needs can be assessed seems to be highly questionable. Videotex 182 development may be caught in its own "Catch 22" where consumer needs will regulate the acceptance of videotex technology, yet these needs will be impossible to ascertain. This will continue to be a nagging issue in videotex research and it will surely continue to be referred to as the one factor which will determine whether this technology either succeeds or fails.

This research has also demonstrated that the perceptions of Channel

2000 differed according to adopter group. This implies a diffusion strategy based on a segmentation approach to the marketplace. Videotex

r change agents are encouraged to very carefully evaluate their potential adopters according to the aforementioned perceptual dimensions. As has been suggested, each of these dimensions implies a unique promotional

strategy in light of the differential relative advantages and compati­

bility issues that may affect each potential adopter segment. If change

agents take the time to develop discriminant functions similar to those

reported in this research, their projections for videotex penetration

levels should be dramatically Increased.

At this point it seems appropriate to speculate on what is needed

for a successfully diffused videotex service based on the research find­

ings presented in this report. Obviously there are the mechanical and

software pragmatics of putting together an interactive community.

Ball (1981) captured the essential Ingredients that are needed for a

successful national videotex operation by stating:

Success will depend on the development of a cheap widely accessible digital communications network, the develop­ ment of suitable applications databases by existing businesses, and a large number of user terminals to gen­ erate the necessary business, (p. 15) 183

The emphasis of this research has not been on this important topic, and

this concluding discussion will not begin to address these issues.

An additional caveat is also in order. The recommendations which

follow are based on the assumption that any organization interested in

providing a mass audience with a videotex service, regardless of its con­

tent or form, is interested in a successful adoption rate. Adoption is

successful when the individual continues to use and incorporate the inno­ vation into their daily lives for an extended period of time. For videotex to be considered a successful innovation, the following recom­

mendations are offered.

The videotex innovation that is made available to potential adopters

must be simplistic in concept. Simplicity does not imply uselessness.

The innovation must be readily perceived as offering one or two major

advantages over current consumer practices, and it must be relatively

compatible with the potential adopter’s value system, and previous

experiences.

By offering potential adopters an innovation that can be easily

evaluated in terms of its relative advantages, the adoption process

should be heightened. If research indicates that potential adopters

cannot evaluate the videotex innovation with any degree of conviction

due to the inherent difficulty of conceptualizing the benefits associated

with videotex, then a reasonable low-risk trial period should be made

available.

The videotex service that is made available should be positioned

as an extension of currently established behavioral patterns. The

availability of widespread, and almost addictive television viewing 184 habits in the United States, implies that an enhanced television exper­ ience via videotex technology could set the stage for initial videotex acceptance. The videotex innovation that is perceived as being comple­ mentary to this television experience should be adopted sooner than that which is not. Since this "complementary" innovation represents little change to the potential adopter, it can be viewed by the change agent as a vehicle which paves the way for later, more complex videotex innovations.

Finally, once this initial videotex service has been adopted, it must meet or exceed the expectations of the potential adopters. If potential adopters buy the videotex concept, and then find dissatis­ faction with its service, the innovation will rapidly fail. Videotex

services must be reliable, and more importantly, expandable. As the

television industry readily discovered, the public devours vast amounts

of programming material, and quickly tires of the old. Thus, the

initial and simplistic concept of videotex must be expandable into the more complex. This expansion must also be based on the rate at which a

potential adopter desires the new.

The above recommendations do not consider the unforeseen economic,

regulatory, and social factors that may interact with this adoption

process. These recommendations also do not account for the social and

philosophical issues associated with videotex. It can be stated quite

confidently, however, that these recommendations can be put to use for

the betterment or detriment of mankind.

These procedures should be used to encourage the equality of

electronic information access among the public at large. Those 185 interested in the humanistic concerns of videotex technology are advised to face the realities of a capitalistic society and employ those market­ ing and communication techniques that will insure the equitable distribu­ tion and adoption of videotex. The technology will be made available; the question is, who, and what will it be made available for.

Limitations

Within the introductory sections of this dissertation, a very early attempt was made to frame this research within the context of diffu­ sion theory. As such, every effort was made to provide other researchers who are interested in diffusion theory and the adoption of newly intro­ duced technologies, with additional information and supportive data on

this important topic. At the same time, this dissertation has attempted

to provide others who are interested in videotex as a communication and marketing device with some insight and an increased understanding of what makes up the videotex phenomenon.

Most definitely, this dissertation could not completely cover both

of these areas in a totally comprehensive manner. Certain areas were

stressed and dealt with in greater depth, while others were not. The

Channel 2000 data could be viewed from a number of alternative per­

spectives, several of which are strongly recommended in a subsequent

section of this dissertation. It is quite apparent that the time and

space limitations inherent to a singular research project limit the

scope and breadth of the research itself. Hopefully in the future,

others will pursue and expand upon many of the important issues which

have been initially raised by this research project. Videotex offers

to many, a totally new and unexplored area of communication inquiry. 186

This dissertation has only scratched the surface of this exciting new area of scholarship.

A point made repeatedly throughout the course of this dissertation has been that videotex is best defined according to the services that it offers. Because viewdata and teletext are an emerging class of media whose boundaries remain largely undefined, it is important for other re­ searchers to be keenly aware of exactly what were the "package" of services

that were made available to a group of subjects in any research investiga­ tion. An evaluation of the Channel 2000 results must also be placed within the context of those videotex services that were made available to the test participants. Hopefully, some underlying generalizations which resulted

from this research can be verified by other studies, but researchers are

forewarned that these results were to a large extent a reflection of the videotex services that were offered to the Channel 2000 test participants.

The small test population that was used for this research has

already been mentioned as an initial research limitation. This factor,

in association with the moderate initial test cooperation rates that were achieved, limits the generalizability of these results to those

kinds of people who would be likely to cooperate with a videotex experi­

ment such as Channel 2000. Undoubtably a certain degree of self­

selection resulted from the sample selection procedures employed by

this research. The reader is forewarned that those people who cooperate

with this kind of research may not be representative of many down-scale,

socioeconomically deprived, non-technology oriented members of society.

Of course it could be argued that the types of people who cooperated

with the Channel 2000 test best represent the potential and available 187 market for videotex services anyhow.

A more troublesome limitation dealt with the fact that not all the Channel 2000 services were made available at the October 1 startup date. Although the Video Card Catalog, Columbus Calendar, Public

Information, Math That Counts, and Early Reader were initially available, the banking services and electronic encyclopedia were not up and running until the first week in November. It might be expected that the respond­ ents experienced a certain amount of disappointment with Channel 2000 since not all the services that were promised were delivered on time.

This factor should be considered in light of the resulting perceptions and attitudes towards Channel 2000.

Unfortunately, there also was a discrepancy in the total number of services that were made available to the test participants. Only

35 Channel 2000 users had access to the Channel 2000 banking services.

This was due to the fact that access to this subsystem was predicated on two conditions. First, the respondent had to have a Bank One checking account, and second, the respondent had to sign a Bank One release form which enabled him or her to use the banking services. The total effect that this inequitable Channel 2000 service had on the test participants is not entirely known.

As reported in Chapter V, the reliability coefficients associated with some of the innovation perceptions and life style scales were not very high. The reader is cautioned in light of these coefficients, and also in light of the fact that some of these scale items were represented by only one or two variables. It seems safe to conclude that more

elaborate scale construction techniques would produce more reliable and 188 valid indices of certain life style dimensions in particular. This is

especially true for the television use and dependency dimension that was

used in this research. Until other studies confirm the relationships

that were found between television evaluation and Channel 2000 potential

adoption, this finding must be viewed only as a preliminary indication

that such a relationship exists. Most definitely, this area of videotex

research cries out for further research evidence.

Ideally, a diffusion study similar to the Channel 2000 test would

have examined the actual adoption of the innovation over time. In this

test, Channel 2000 was not made available to the test participants for

adoption in the classic sense of the word. Instead, measures of poten­

tial adoption had to suffice. Although these measures of potential

adoption were very carefully examined in terms of their measurement pre­

cision, they may not accurately reflect the actual adoption of Channel

2000. As is the case with most research, a paper and pencil test had

to be a substitute for actual behavior.

The element of time is also not adequately dealt with in this

research. Although the data offer some unique opportunities in terms of

time series data analysis, this research chose an alternative research

strategy which first stressed data reduction techniques, reliability and

validity assessment, and an examination of construct relationships and 4 importance. The second step in this analysis should most definitely

include a detailed analysis of these variables over a three month time

period.

Finally, the reader is encouraged to appreciate the exploratory

nature of the Channel 2000 field test. The total lack of previously 189 published works on this topic necessitated a somewhat open minded approach to the collection of the data. Certainly there remains a number of unanswered questions, although the objective of this research was to first explore the most obvious, and the most theoretically relevant issues that face this totally unexplored area of communication technology.

Recommendations for Future Research

Because videotex technology represents a multitude of interrelated and multifaceted areas of inquiry, the task of defining what should be studied is not a simplistic matter. Videotex crosses all boundaries of scholarly inquiry, and given its potential impact on society, it should emerge as a very prominent field of academic endeavor within the next

10 years. Because of the breadth of this topic, the recommendations put forth in this discussion may seem to be relatively specific and oriented mostly towards diffusion theory. This restriction was based on the decision to provide others with an established database and group of related findings from which subsequent research can build upon and

expand.

The relationships between innovation perceptions and the adoption

of an innovation warrants addition research attention. Two general areas

are in most need of study. First, the perceptions of an innovation as

they correspond to the specifically defined stages of the adoption

process should be examined through sequential time series procedures.

If, as this research has partially determined, various innovation attri­

butes are in fact more important at different stages in the adoption

process, change agents can intelligently coordinate the introduction of

an innovation as it proceeds through its life cycle. This important 190 area of diffusion research has been grossly neglected.

Second, the perceptual differences experienced by adopter groups could be an invaluable aid towards an increased understanding of why certain innovations either fail or succeed. Here again, if these differ­ ences can be identified, change agents can modify their diffusion strat­ egies accordingly. Coupled with additional information on the sequential perceptual changes that occur throughout the adoption process, the effectiveness of diffusion campaigns could be significantly increased.

Although the six perceptual dimensions used in this research were shown to be extremely useful conceptual tools, additional dimensions need to be generated and tested for their validity across a number of innovations. Zaltman and Lin (1971) have proposed innovation attributes such as terminality, reversibility, commitment, public vs. private, and ego-involvement as examples of expanded attribute categories. The defin­ ition of these various concepts, and the research conducted on them, is a much needed step towards a theory of adoption and subsequent diffusion.

One step in this direction would be a series of factor analyses on those adjectives that best describe an innovation. If similar perceptual dimensions continually emerge across a number of innovations, a clearer understanding of just how many of these perceptual dimensions are actu­ ally needed in diffusion research could be determined.

The relationships between innovation percpetions and various life style and personality characteristics should be explored. If, for example, an Innovation is found to be compatible, researchers need to

understand why it was perceived in this manner. The perceptual dimen­

sions used in this research, although proven to have considerable 191 predictive, and discriminatory power, need dimensionality. Although the research evidence on this topic may be somewhat innovation specific, those relationships which underlie the perception of videotex tech­ nology could be a very valuable body of knowledge.

For example, some evidence was provided in this research to suggest that Channel 2000 was perceived as being an innovation that was compat­ ible to the television viewing experience. Without the conceptual tie between the life style pattern of television viewing and compatibility, the reasoning behind this perception may have been left unexplained.

Diffusion researchers need to combine both the perceptions of an innova­ tion with the humanistic explanations which can account for these per­ ceptions. This approach to videotex research should provide the most lucid insights into the personal aspects of this very personal technology.

The relationships between the adoption of videotex and established media habits should be further explored. Will the gratifications cur­ rently derived from the media be replaced by videotex services? What effects will videotex have on current media behaviors? What are the relationships between heavy television viewing and videotex adoption? A very profitable area of analysis seems to be an investigation of how people currently use the media to be informed or entertained, and relate these patterns to potential videotex use.

A most obvious extension of this logic is the analysis of how people interact with, and learn from, currently available computer technologies.

The need for expanded research in the area of human-computer interaction seems to be a critical ingredient in the understanding of videotex tech­ nology. 192

Most of the aforementioned recommendations assume a research meth­ odology which would require a measure of videotex adoption or potential adoption. Unfortunately there is no available evidence regarding the relationship between measured adoption (as in a questionnaire) and actual adoption.

What is needed is a comparison of projected potential adoption measurement results, and actual adoption patterns. If the potential adoption measures reported in this research accurately reflect actual adoption tendencies, diffusion researchers will have a very powerful analysis device.

Ultimately, researchers who study the diffusion of videotex tech­ nologies, should be able to compare and contrast the profiles of videotex innovators derived from varying videotex systems. In doing so, a very important contribution can be made to diffusion theory, as well as to the understanding of videotex technology. Clearly what is needed is a growing body of knowledge based on established theoretical foundations that can organize constrasting investigations into a unified whole.

A number of available perspectives in communication theory seems to be especially relevant to this topic. Those fields of study which look at media information processing and search strategies, the uses and gratifications of the media, media displacement effects, and of course t the diffusion of innovations, all offer a strong empirical foundation from which others can build. Those interested in studying media organ­ izations have a totally new area of inquiry and investigation. Others may investigate the legal and economic conditions which will determine the parameters within which videotex technology must coexist. Historical 193 relationships between the emergence of radio and television technologies could provide some very interesting insights to resultant trends in media technology adoption and diffusion. The social issues associated with videotex technology alone could be the topic of several dissertations and books.

Many predict that videotex technology will transcend our current communication behaviors and revolutionize the way we spend both our dis­ cretionary and non-discretionary time. It has been suggested that video­ tex technology will affect our democracy, our society, our legal system, our homes, our businesses, and our lives. For this technology to develop unguided by the perceptions, needs, and wants of*those people for whom it will have such a disruptive effect, would indeed be a saddening state of affairs. Had others been aware of television’s impact upon society prior to its commercialized introduction, its development may have pro­ ceeded along a more cautious path. If history does repeat itself, let us alter its direction through careful analysis and coordinated decision making. If videotex does change our lives, let that change be to the benefit of society.

Conclusions

The results of this study can be summarized by the following con­ clusions:

1. A certain degree of disenchantment with the Channel 2000 system as measured by innovation perceptions was experienced over the three month test period. Channel 2000 was perceived as being less compatible, observable, trialable, offered fewer relative advantages, and contained more risk at Time II than at Time I. Channel 2000 was not perceived as 194

being complex.

2. Innovation perceptions differ between adopters and non-adopters

of Channel 2000 along the dimensions of relative advantage, compatibility,

and perceived risk at Time I; relative advantage, compatibility, observ­

ability, trialability, and perceived risk at Times II and III. Thus,

some perceptual differences exist between potential adopters and non­

adopters of an innovation.

3. Innovation perceptions exhibit a moderate amount of correlation with one another, and with other life style constructs.

4. Potential adopters and non-adopters of Channel 2000 can be

significantly discriminated on the basis of their perceptions of Channel

2000 along the dimensions of relative advantage and compatibility, and

in terms of their evaluation and use of television, and their frequency

of conducting banking related activities, and using an encyclopedia.

5. Those constructs which differentiate potential adopters and

non-adopters of Channel 2000 change in terms of their importance over

time although relative advantage and compatibility consistently are the

most important discriminating items.

6. Innovation perceptions are more useful in discriminating between

potential adopters and non-adopters of Channel 2000 than are life style,

consumer creativity, and sociodemographic variables.

7. Compatibility is more important at the awareness stage of adop­

tion, and relative advantage is more important at the persuasion and

decision stages.

8. Although consumer creativity is not a useful discriminating

construct, it is negatively correlated to potential Channel 2000 adoption. 195

9. Sociodemographic variables appear to be of limited value in discriminating between potential adopters and non-adopters of Channel

2000.

Less conclusive findings which resulted from this research are:

1. The degree to which an individual receives information and entertainment from television relates positively to the adoption of

Channel 2000.

2. The degree to which individuals engage in an activity that can be replaced by Channel 2000 positively correlates to potential Channel

2000 adoption.

3. Consumer venturesomeness positively relates to Channel 2000 adoption.

4. Purchase probability and price sensitivity measures of potential

Channel 2000 adoption appear to be measuring the same thing. Perceived use does not exhibit a strong relationship to these two variables.

This research has not assumed that these findings will in fact with­ stand the test of time. The pace at which this technology is currently progressing will more than likely antiquate the concepts and procedures that were used to design the Channel 2000 system. However, the general principles established by diffusion theory, and those findings reported herein that relate to this theory, should remain as viable explanations of the videotex phenomenon for many years to come. FOOTNOTES

For additional introductory information on teletext see: Educa­ tional and Instructional Television, 1979, June, 31-39 (special section on teletest); see also "TV Takes on the Printed Word," Marketing and Media Decisions, 1980, May, 74-75, 130-136; see also IEEE Transactions on Consumer Electronics, 1979, 25:3, 23-28.

2 For additional introductory information on viewdata see: Tele­ communication Policy, 1980, March, entire issue; see also "Why TV Sets Do More in Columbus, Ohio," Fortune Magazine, 1980, October 6, 67-73; see also Itheal de Sola Pool, Talking Back: Citizen Feedback and Cable Technology (Cambridge, Mass.: MIT Press, 1973).

3 Microcomputers and viewdata systems. Information Retrieval and Library Automation, 1980, 15:10, 1.

4 Try out the new media now. Broadcasting, March 24, 1980, 56-57.

5 Top Time exec sees teletext as threat to newspapers. Editor and Publisher, May 24, 1980, 37.

^Advertisers eye cable TV impact. Editor and Publisher, May 17, 1980, 39.

^TV takes on the printed word. Marketing and Media Decisions, May 1980, 136.

196 REFERENCES

Adler, R. Scales of open processing and adoption. Unpublished report. Columbus, Ohio, 1975.

Arndt, J. Profiling consumer innovators. In Arndt, J. (ed.), Insights into Consumer Behavior. Boston: Allyn and Bacon, Inc., 1968.

Arndt, J. Role of product related conversations in the diffusion of a new product. Journal of Marketing Research, 1967, 4, 291-295.

Ball, A. J. S. Videotex: chimera or dream machine. Canadian Mbrary Journal, 1981, 38:1, 11-15.

Barnett, H. G. Innovation: The Basis of Cultural Change. New York: McGraw-Hill^ 1953.

Bauer, R. A. Consumer behavior as risk taking. In Proceedings of the American Marketing Association, 1960, American Marketing Associa­ tion, 389-398.

Blackwell, R., & Talarzyk, W. Consumer Attitudes toward Health Care and Malpractice. Columbus: Grid Publishing, Inc., 1977.

Blake, B., Perloff, R., & Heslin, R. Dogmatism and acceptance of new products. Journal of Marketing Research, 1970, 7, 483-486.

Blumler, J. G., & Katz, E. The Uses of Mass Communication. Beverly Hills: Sage Publications, 1974.

Boone, L. E. The search for the consumer innovator. The Journal of Business, 1970, 43:2, 135-140.

Burnkrank, R. E. Interpersonal influence in consumer behavior. In Calder, B. J. (ed.), Current Perspectives in Consumer Behavior. New York: Wiley, 1979.

Carey, J. W., & Quirk, J. J. The history of the future. In Gerbner, G., Gross, P., & Melody, W. H. (eds.), Communications Technology and Social Policy: Understanding the New "Cultural Revolution.11 New York: John Wiley and Sons, 1973.

Cattell, R. (ed.). Handbook of Multivariate Experimental Psychology. Stokie, 111.: Rand-McNally, 1966.

Churchill, G. A. A paradigm, for developing better measures of marketing constructs. Journal of Marketing Research, 1979, 16, 64-73.

197 198

Coney, J. Dogmatism and innovation: A replication. Journal of Market­ ing Research, 1972, 9, 453-455.

Craig, S. C., & Ginter, J. L. An empirical test of a scale for inno­ vativeness. In Schinger, M. J. (ed.), Advances in Consumer Re­ search, 2, 1975.

Criner, K. Teletext and videotex in North America. U.S. videotex activities and policy concerns. Telecommunications Policy, 1980 March, 3-8.

Darden, W. R., & Reynolds, F. D. Backward profiling of male innovators. Journal of Marketing Research, 1974, 11, 79-85.

Deutschman, P. J., St Danielson, W. Diffusion of knowledge of the major news study. Journalism Quarterly, 1960, 345-355.

Dixon, W. J. (ed.). BMP: Biomedical Programs. Berkley: University of California Press, 1973.

Donnelly, J. H., & Ivancevich, J. M. A methodology for identifying innovator characteristics of new brand purchasers. Journal of Marketing Research, 1974, 9, 331-334.

Eastman, A. T. Viewdata - a product in search of a market? Telecommuni­ cations Policy, 1980 March, 25-30.

Engel, J. F., Blackwell, R. D., & Kollat, D. T. Consumer Behavior, 3rd Ed. Hindale, Illinois, 1978.

Feldman, L. P., St Armstrong, G. M. Identifying buyers of a major auto­ motive innovation. Journal of Marketing, 1975, 39, 47-53.

Fliegel, F., & Kivlin, J. Attributes of innovations as factors in dif­ fusion. The American Journal of Sociology, 1966, 72:3, 235-248.

Garvey, M. A. The Silent Revolution: or the Future Effects of Mankind. London: William and Frederick G. Cash, 1852.

Goldberg, M. E. Identifying relevant psychographic segments; how specifying product functions can help. Journal of Consumer Re­ search, 1976, 3, 163-169.

Goldman, R. J. Teletext and videotex in North America: Demand for telecommunications services in the home. Telecommunications Policy, 1980 March, 25-30.

Graham, L. S. Class and conservativism in the adoption of innovations. Human Relations. 1956,‘9, 91-100.

Green, P. E., & Tull, D. S. Research for Marketing Decisions. Engle­ wood Cliffs, N.J., 1978. 199

Harnish, T. Channel 2000: Project Report. An executive summary of the Channel 2000 project distributed by OCLC, Inc., 1981.

Hirschman, E. C. Innovativeness, novelty seeking, and consumer creativ­ ity. Journal of Consumer Research, 1980, 7, 283-295.

Jacoby, J. Personality and innovative pronenass. Journal of Marketing Research, 1971, 8, 244-247.

Johansen, R., Nyhan, M. J., & Plummer. Teletext and videotex in North America: Issues and insights for the USA. Telecommunications Policy, 1980 March, 31-41.

Juster, T. F. Consumer buying intentions and purchase probability: an experiment in survey design. Journal of the American Statistical Association, 1966, 61:35, 658-696.

Kaigler-Evans, K. L. Fashion information source effects in interpersonal communication. Unpublished doctoral dissertation, Columbus, Ohio. The Ohio State University, 1975.

Kaplan, A. The Conduct of Inquiry. San Francisco: Chandler, 1964.

Kassarjian, H. H. Personality and consumer behavior: A review. Journal of Marketing Research, 1971, 8, 409-418.

Kassarjian, H. H. Social character and differential preference for mass communication. Journal of Marketing Research, 1965 May, 146-153.

Katona, G., & Mueller, E. A study of purchasing decisions. In Clark, L. (ed.), Consumer Behavior: The Dynamics of Consumer Reaction. New York: New York University Press, 1955.

Katz, E. The two step flow of communication: an up-to-date report on a hypothesis. Public Opinion Quarterly, 1957, 21, 61-78.

Katz, E., Blumler, J. G., & Gurevitch, M. Utilization of mass communi­ cation by the individual. In Blumler and Katz, op. cit.

Katz, E., & Lazarsfeld, P. E. Personal Influence. New York: The Free Press, 1955.

Kerlinger, F. N. Foundations of Behavioral Research. New York: Holt, Rinehart and Winston, Inc., 1973.

King, C. W. The innovator in the fashion adoption process. In Pro­ ceedings of the Winter Conference of the American Marketing Associa­ tion, 1964.

King, C. W., & Summers, J. 0. Overlap of opinion leadership across con­ sumer product categories. Journal of Marketing Research, 1970, 7, 43-50. 200

Kivlin, J. Characteristics of farm practices associated with rate of adoption. Unpublished Ph.D. dissertation, University Park: The Pennsylvania State University, 1960.

Klecka, W. R. Discriminant Analysis. Beverly Hills, Calif.: Sage Publications, 1978.

Kotler, P., & Saltman, G. Targeting prospects for a new product. Journal of Advertising Research, 1976, 16:1, 7-18.

Kuhn, T. S. The Structure of Scientific Revolutions. Chicago: The University of Chicago Press, 1970.

Lambert, Z. V. Perceptual patterns, information handling, and inno­ vativeness. Journal of Marketing Research, 1972, 9, 427-431.

Lazarsfeld, P., Berelson, B., & Gaudet, H. The People's Choice. New York: Columbia University Press, 1944.

Lazer, W. Life style concepts and marketing. Toward Scientific Market­ ing, Proceedings of the American Marketing Association, 1963, 130- 139.

Leavitt, C., & Keitz, J. Innovativeness and women's fashion. Working paper if80-6, Columbus, Ohio. The Ohio State University, 1980.

Leavitt, C., & Walton, J. Personality and new product use. An un­ published research report. Columbus, Ohio, 1977.

Leduc, N. F. Teletext and videotex in North America: The Canadian Perspective. Telecommunications Policy, 1980 March, 9-16.

McCain, T., & Wall, V. River Ridge II: Social system effects on the perceived attributes of a school board proposal. Western Speech Communication Journal, 1977, 42:3.

Midgley, D. F., & Dowling, G. R. Innovativeness: The concept and its measurement. Journal of Consumer Research. 1978, 4, 229-242.

Noll, A. M. Teletext and videotex in North America: Service and system implications. Telecommunications Policy, 1980 March, 17-24.

Ostlund, L. E. Product perceptions and predispositional factors as determinants innovative behavior. Unpublished doctoral disserta­ tion, Boston: Harvard University, 1969a.

Ostlund, L. E. The role of product perceptions in innovative behavior. In McDonald, P. R. (ed.), Marketing Involvement in Society and the Economy. Proceedings of the 1969 Fall Conference, American Market­ ing Association, Chicago, 1969b. 201

Ostlund, L. E. Identifying early buyers. Journal of Advertising Re­ search, 1972, 12:2, 25-30.

Ostlund, L. E. Factor analysis applied to predictors of innovative behavior. Decision Sciences, 1973, 4, 92-108.

Ostlund, L. E. Perceived innovation attributes as predictors of inno­ vativeness. Journal of Consumer Research, 1974, 1, 23-29.

Peterson, R. A. A note on optimal adopter category determination. Journal of Marketing Research, 1973, 10, 325-329.

Plummer, R. P., Johansen, R., Nyhan, M. J., & Holmlor, P. G. 4004 Futures for teletext and videotex in the U.S. - IEEE Transactions on Consumer Electronics, 1979, 25:3, 318-325.

Robertson, J. P. Interpersonal influence in election campaigns: Two step flow hypotheses. Public Opinion Quarterly, 1976, 40, 304-319.

Robertson, T. S. Innovative Behavior and Communication. New York: Holt, Rinehart and Winston, 1971.

Robertson, T.. S., & Kennedy, J. N. Prediction of consumer innovators: Application of multiple discriminant analysis. Journal of Market­ ing Research, 1968, 5, 64-69.

Robertson, T. S., & Myers, J. H. Personality correlates of opinion leadership and innovative buying behavior. Journal of Marketing Research, 1969, 6:2, 164-168.

Rogers, E. M. Diffusion of Innovations. New York: The Free Press, 1962.

Rogers, E. M. Modernization Among Peasants. New York: Holt, Rinehart and Winston, Inc., 1969.

Rogers, E. M. New product adoption and diffusion. Journal of Consumer Research, 1976, 2, 290-301.

Rogers, E. M. & Cartano, D. Methods of measuring opinion leadership. Public Opinion Quarterly, 1962, 26:3, 435-441.

Rogers, E. M. & Shoemaker. The Communication of Innovations. New York: The Free Press, 1962.

Rogers, E. M., & Shoemaker. The Communication of Innovations. New York: The Free Press, 1971.

Rogers, E. M. , & Stanfield, D. Adoption and diffusion of new products: Emerging generalizations and hypotheses. Paper presented at the Conference on the Application of Sciences to Marketing Management, Purdue University, 1966. 202

Rogers, E. M . , & Thomas, P. C. Bibliography on the Diffusion of Inno­ vations. Ann Arbor: Department of Population Planning, University of Michigan, 1975.

Simons, D. M. A primer on teletext and viewdata and using them without getting nicked. Paper distributed by Antiope Videotex Systems, Inc., 1725 K Street, N.W., Washington, D.C. 20006, 1980.

Summers, J. 0., & King, C. W. Interpersonal communication and new product attitudes. In McDonald, P. R. (ed.), Marketing Involvement in Society and the Economy. Chicago: American Marketing Associa­ tion, 1969.

Taylor, J. W. A striking characteristic of innovators. Journal of Marketing Research, 1977, 14, 104-107.

Thio, A. 0. A reconsideration of the concept of adopter-innovation compatibility in diffusion research. Sociological Quarterly, 1971, 12, 56-68.

Toffler, A. The Third Wave. New York: William Morrow and Company, 1980.

Toles, T. M. Videotex: Beyond the technology questions. Paper presented at the Pacific Telecommunications Conference, January 1981.

Trodahl, V. C. A field test of a modified "two step flow of communication mode." Public Opinion Quarterly, 1966.

Trodahl, V. C., & Van Dam, R. Face to face communication about major topics in the news. Public Opinion Quarterly, 1965, 29:4, 626-634.

TV takes on the printed word. Marketing and Media Decisions, 1980 May, 74-75, 130-136.

Tydeman, J., & Lipinski, H. Technological assessment on teletext/video­ tex. Institute for the Future, Menlo Park, California, 1980.

Van Den Ban, A. W. A revision of the Two Step Flow of Communication hypothesis. Gazette, 1964, 10:3, 237-249.

Vermilea, D. Where will teletext be in 2000 A.D.? Educational and Instructional Television, 1980 July, 23-24.

Wallace, W. C. The Logic of Science in Sociology. Chicago: Aldine Publishing Company, 1971.

Weizenbaum, J. Once more: The computer revolution. In Dertouzos, M. L., & Moses, J., The Computer Age: A Twenty Year View. Cambridge, MA: MIT Press, 1979.

Westin, A. F. Information Technology As a Democracy. Cambridge, MA: Harvard University Press, 1971. 203

Wind, Y., & Green, P. E. Some conceptual, measurement, and analytical problems in life style research. In Wells, W. (ed.), Life Style and Psychographics. Chicago: American Marketing Association, 1974.

Zaltman, G., & Lin, N. On the nature of innovations. American Behavioral Scientist. May-June, 1971.

Zaltman, G., & Stiff, R. Theories of diffusion. In Ward, S., & Robert­ son, T. S., Consumer Behavior: Theoretical Sources. Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1973.

Zikmund, W. G., & Scott, J. E. An investigation of the role of product characteristics in risk perception. Review of Business and Economic Research, 1977, 13:1, 19-34.

4 APPENDIX A

RECENT VIDEOTEX PROJECTS RECENT VIDEOTEX PROJECTS EXPERIMENTS BY COUNTRY

Country and Project Sponsors Description Date Number of Participants Address and Telephone Number

UNITED STATES:

KSL-TV KSL-TV Oracle and 1978 Transmission test only KSL-TV Salt Lake City, UT Ceefax Tele­ 145 Social Hall Avenue text trial; Salt Lake City, UT 84111 Touch tone (801) 237-2500 teletext

Infotext Micro TV Teletext 1978 Transmission test only Micro TV Philadelphia, PA 3600 Conshohocken Avenue Philadelphia, PA 19131 (215) 879-9000

KMOX-TV CBS Teletext- 1979 Transmission test only KMOX-TV St. Louis, MO Antiope and One South Memorial Drive Ceefax St. Louis, MO 63102 (314) 621-2345, or CBS, New York, NY

KCET-TV KCET-TV Teletext 1980 Transmission test only KCET Los Angeles, CA 4401 Sunset Boulevard Los Angeles, CA 90027 (213) 666-6500

Inteltex Microband Used MDS Trans-1978- Teletext test; Microband Microband Corp. of America Washington, DC Corp. mitter in 1979 will offer a viewdata 655 Third Avenue Washington service to businesses, New York, NY 10017 for test of government and educa- (212) 867-9590 Antiope tional institutions 205 \

Country and Project Sponsors Description Date Number of Participants Address and Telephone Number

Cabletext Southern Teletext- 1980 Transmission test; will Southern Satellite Systems Douglasville, GA Satellite Ceefax via be offered to cable 3530 Bomar Road satellite companies Douglasville, GA 30135 (404) 949-6600

Electronic Informa- AT&T Viewdata 1979 84 AT&T tion Service Albany, NY Albany, NY (518) 463-5022 Parsippany, NY (201) 428-3000

Green Thumb U.S. Dept. Viewdata 1980 100 University of Kentucky Lexington, KY of Agricul­ College of Agriculture ture & Na­ Cooperative Extension Service tional Lexington, KY 40546 Weather (606) 255-5582 Service

Viewtron Knight- Viewdata 1980 160 Viewdata Corp. of America Coral Gables, FL Ridder News­ 1444 Biscayne Boulevard papers , Inc. Miami, FL 33132 & AT&T (305) 350-6082

Channel 2000 OCLC & Banc Viewdata 1980 200 OCLC, Inc. Columbus, OH One Corp. 1125 Kinnear Road Columbus, OH 43212 (614) 486-3661

Citibank Citibank Viewdata- 1980 100 Citibank New York, NY banking at 399 Park Avenue home New York, NY 10043 (212) 559-2588 NJ o O n Country and Project Sponsors Description Date Humber of Participants Address and Telephone Number

Qube Warner-Amex Existing two- 1977 25,000 Qube Columbus, OH Corp. way inter­ 930 Kinnear Road active cable Columbus, OH 43212 operation (614) 459-3300

Home Book Club OCLC & Qube Book discus- 1979 Qube audience OCLC, Inc. Columbus, OH sion via in­ 1125 Kinnear Road teractive Columbus, OH 43212 cable (614) 486-3661

Woodlands, TX To com & Operational 1975 2,000 Tocom, Inc. Mitchell two-way cable Box 47066 Energy De­ provide Dallas, TX 75247 velopment security pro­ (214) 438-7691, or tection CATV 2407, Suite A Timberloch Woodlands, TX 77373 (713) 367-2288

Las Colinas Dow Jones & Two-way cable 1979 8 Dow Jones & Company Irving, TX Company to Apple Com­ P.O. Box 300 puter owners Princeton, NJ 08540 (609) 452-2000

Park Cities Project Sammons Com- Cable informa- 1980 Unrestricted, but at Dow Jones & Company Dallas, TX munications, tion retrieval least 2,000 P.O. Box 300 Inc.; Dow and finance Princeton, NJ 08540 Jones; Mer- analysis (609) 452-2000 rill Lynch;

A. H. Belo 207 Country and Project Sponsors Description Date Number of Participants Address and Telephone Number

Annapolis, MD Communica- Cable 1981 3,000 Communications Technology tions Tech­ Mgt. nology Man­ P.O. Box 915 agement Inc. 6861 Elm Street and Prime McLean, VA 22101 Cable (703) 734-2724

San Diego, CA Cox Cable Cable banking 1980 100-300 Cox Cable Communications, Communica- and catalog Inc. tions, Inc.; shopping 219 Perimeter Pkwy, West, HomServ; tf500 . Fingerhut Atlanta, GA 30346

Lexington, KY Public Li­ One-way cable 1980 Unrestricted Lexington Public Library brary of initially; 251 West Second Street Lexington, library infor­ Lexington, KY 40507 KY & Tele­ mation (606) 252-8871 cable of Lexington, Inc.

WETA Corp. for Broadcast tele- 1980 60 WETA-TV Washington DC Public text using Corp. of Public Broadcasting Broadcast- Telidon Washington, DC 20036 ing, NSF, (202) 293-6160 National Telecommuni­ cations & In­ formation Admin., Dept, of Health & Human Services oto oo Country and Project Sponsors Description Date Number of Participants Address and Telephone Number

Chemical Bank Chemical Viewdata-Bank- 1980 200 - nine county area Chemical Bank New York Bank ing, In-home 20 Pine Street shopping, en­ New York, NY 10017 tertainment, (212) 770-1246 educational games, News- wires

United California United Cal­ Telephone-view-1981 200-400 United California Bank Bank ifornia data banking 700 Wilshire Boulevard Los Angeles, CA Bank Los Angeles, CA 90017 (213) 614-4111

CANADA:

Vista Bell Canada Viewdata, in- 1979- 1979 - 34 Bell Canada Canada itial experi- 1980 1980 - 1,000 25 Eddy Street Dept, of ment used Fifth Floor Communica­ Prestel; Teli- Hull, Quebec tions don will be (819) 776-7647 used in newest experiment

Vidon Alberta Viewdata — 1979 115 Alberta Government - 30F Gov't Tele- Prestell and 10020-100 Street phone and Telidon Edmonton, Alberta Harris Corp. T5J 0N4 (403) 425-2460

Grand River Canadian Cable 1980 75 Grand River Cable TV Kitchener, Ontario Cable Sys­ 209 tems (519) 893-2101 Country and Project Sponsors Description Date Number of Participants Address and Telephone Number

TV Ontario Canadian Teletext and 1980 55 Ontario Educational Communi­ Dept, of Viewdata cation Authority Communica­ using Teli- P.O. Box 200 - Station 0 tions and don Toronto, Ontario Ontario M4T 2T1 Educational (416) 484-2621 Communica­ tions

Mercury New Bruns- Viewdata 1980 75 New Brunswick Telephone Co., wick Tele- (Telidon) and Ltd. phone Co. of alarm connected One Brunswick Square Canada to Central St. John, New Brunswick computer by E2L 4K2 telephone (506) 648-2340

Ida Manitoba Cable augment- 1979 100 Manitoba Telephone Telephone ed by tele­ BE 301A Systems phone lines P.O. Box 6666 Winnipeg, Manitoba R3C 3V6 (204) 947-4111

London, Ontario Canadian Two-way cable 1977 300 Canadian Cablesysterns Cable- London, Ontario systems and N5W 2T1 London (579) 672-7704 Cable TV Country and Project Sponsors Description Date Number of Participants Address and Telephone Number

Elie Canadian Fiberoptic 1981 150 CTCA Telecom­ cable offer­ 1 Nicholas Street munica­ ing tele­ Suite 700 tions Car­ phone view­ Ottawa, Ontario riers' data and (613) 238-3038 Ass'n, and regular tele­ Canadian vision. Department Telidon of Communi­ cations

FINLAND;

Telset Sanoma Pub- Viewdata 1976 200 Sanoma Osakeyhtio lishing P.O. Box 240 Viewdata 00101 Helsinki 10 Co., Hel­ Finland sinki Tele­ phone Co., & Nokia Electronics

NETHERLANDS:

Viditel Netherlands Viewdata 1976 200 Nederlandse Omrdep Postal & Stichting (NOS) Telecommuni­ P.O. Box 10 cations Serv­ 1200 JB Hilversum ices, Nether­ The Netherlands

lands Broad­ 211 casting Corp. & Modern Media Foundation Country and Project Sponsors Description Date Number of Participants Address and Telephone Number

HCNG KONG:

Hong Kong Viewdata 1980 N/A Hong Kong Telephone Co. Ltd. Telephone P.O. Box 479 Company General Post Office Hong Kong

FRANCE:

Antiope TDF, the Teletext, 1977 500 Centre Commun d'Etudes de nat'l. TV Viewdata Television et Telecommuni­ network cations 35 Rennes, Frances or Teletel CCETT- Viewdata 1981 2,000-2,500 Centre Intelematique Commun Tour Maine Mantparnasse d'Etudes de 28th Floor Television PS 118 et Telecom­ Paris, FRANCE munications

Electronic PTT-Postes Viewdata 1982 270,000 Directory Telecom­ munications et Tele- graphes Country and Project Sponsors Description Date Number of Participants Address and Telephone Number

WEST GERMANY:

Broadcast German Fed­ Teletext test- 1975- N/A Deutsche Bundespost Teletext eral Post ing, Ceefax Pres­ P.O. Box 5000 Office and Antiope ently D 61 Darmstadt West Germany Bildschirmtext German Fed- Viewdata 1980 4,000 eral Post Office

JAPAN:

CIbS (Character NHK-TV Teletext- 1978 Transmission test only Nihan Hoso Kyokai Information Broad­ 2-2-1 Jinnan casting Station) Tokyo 150 JAPAN

Captains Japanese Viewdata 1979 1,000 CAPTAIN System Ministry Research and Development of Posts Institute and Tele­ RM 105 MPT Iikura Annex communica­ 6-19 Azakudai 1-Chome tions and Minato-Kee Nippon Tokyo, JAPAN Telephone d Telegraph Public Cor­ poration Country and Project Sponsors Description Date Number of Participants Address and Telephone Number

CCIS Japanese Cable 1976 N/A Ministry of Posts and Tele­ Ministry of communications Posts and 3-2 Kasuniigaseki Telecommuni­ 1-Chome cations Chiyada-ku Tokyo, JAPAN

Hi-Ovis Visual In- Optical Fiber 1973- 158 Ministry of Posts and Tele­ formation cable project communications System De­ was 3-2 Kasuniigaseki velopment first 1-Chome Association dis­ Chiyada-ku and Japanese cussed Tokyo, JAPAN Gov1t.

GREAT BRITAIN;

Ceefax BBC-1 Teletext 1973- 30,000-equipped sets BBC Television Centre BBC-2 tested (Ceefax) Wood Lane 1974- London W12 7RJ live ENGLAND

Oracle Independ- Teletext 1975 N/A IBA (Independent Broad­ ent Broad­ casting Authority) casting 70 Brompton Road Authority London SW2 ENGLAND Country and Project Sponsors Description Number of Participants Address and Telephone Number

Prestel British Viewdata 2,822-terminals Prestel Postal installed British Post Office Service 2,411-business Telephone House 411-consumer Temple Avenue London EC4 Y OHL ENGLAND APPENDIX B

VIDEOTEX APPLICATIONS VIDEOTEX APPLICATIONS

Application Areas: Community Services

Application Description Users

Community bulletin An up-to-the-minute listing of local events within a Profit and non-profit board specific geographic area, accessible by subject type, organizations. place, etc. Also proposed as a public information Individuals. utility. Electronic referral services.

Mass transit/travel A listing of all route schedules plus intercity con­ Commuters, shoppers, information nections, including airline, bus and train. In more tourists. sophisticated version, trip planning capabilities are available.

Emergency and urgent Latest reports on accidents, road conditions, weather Police, fire, disaster Information flashes, air pollution, etc. officials. Travelers pilots, commuters, bus/truck drivers.

Government information Meeting agenda for government agencies. Listing of Local, regional, state meetings. Notice of reports, regulation changes, etc. and federal agencies. Contracting organiza­ tions. Lobby groups.

Referenda or quasi­ Citizen input to government officials on controversial Public agencies. referenda issues.

Housing availability Multiple listing service. House, apartment, condominium Buyers, sellers, sales and rentals. Hotel and motel space availability. renters, visitors. Application Description Users

Electronic gossip Links up residents with similar interests to share Local residents. information, opinions, and ideas. (Would need full messaging capability.)

Electronic hotlines Instead of telephone hotlines, these would match up Depending on subject requests with information (e.g., poison control area. information — type of poison entered and recommended action immediately displayed).

Foreign language Translations of community announcements, emergency, Non-English speaking service and other information. residents and visitors

Captioning Subtitling of TV programs, news, etc. Hearing-impaired.

Electronic directories Open or closed systems for providing listings of Hotels, shopping centers employees, buildings, stores, hours of service, etc. government agencies, campuses.

Application Areas: Advertising and Marketing

Application Description Users

Electronic yellow pages Directory-type information updated regularly (rather Telephone companies, than annually) accessed by a number of different cate­ consumers, businesses. gories. In more sophisticated systems, could be com­ bined with order-taking capability (e.g., restaurant listing with menu preview and reservation capabilities).

Supplement to TV National ads with local dealer and price information. Advertisers, consumers. advertising More detailed description on product (e.g., nutritional ingredients, suggested recipes). Could have order-taking capability. Application Description Users

Classified advertising Similar to newspaper classified advertising, but with Small advertisers, more capabilities for indexing and retrieval. employment agencies, general public.

Display advertising Retail and chains and supermarkets. Show weekly specials Advertisers, con­ and product information. Could be used in conjunction sumers . with other ads. May include ordering capability.

Market research Use of new product is tracked through online polling of Market research groups, consumers.

Comparison shopping Umbrella group collects information on current prices New umbrella service for particular products and provides that information groups, consumers. to system subscribers.

Consumer action Complaints and reactions to products of services are Consumers. channeled through consumer ombudsman service (similar to action-line services in newspapers and TV) with feedback to participants.

Electronic catalogues Online access to catalogues (e.g., Sears) with order- Mail order houses, taking capability. catalogue users.

Entertainment options Ticketron-type services available at home or office. Theater owners, guide- Leisure time planning. Vacation options. Could include book publishers, reviews, ratings, guidebook information. Application Areas: Games and Entertainment

Application Description Users

Video games Downloaded games with potential of adding multiple Game enthusiasts, players. Could be combined with TV programs such that TV program producers. participants at home become Involved and influence course of program.

In-home travel Accessing travelogues (video, sound, and smell).

Gambling Quotations, pool size, risk information. Potential State lotteries extension of off-track betting offices.

On-demand TV Dial-up program selection. Pay-TV subscribers.

Sporting events Audience participation, providing comment and feedback on Males, 25-45 years. play choices, player performance, etc.

Pornography On-demand adult-oriented material. "X" and "R" - rated movie (and magazine) customers.

Application Areas: Home-based Services

Application Description Users

Telemonitoring Remote sensors: Meter reading. Emergency fire, police Home owners. notification.

Energy management Control and regulation of household energy use. Home owners, utility companies. 220 Home computing Provide additional computing power to home terminals. Students, self-employed. Application Description Users

Personal Information Private electronic files for home-related information. storage

Financial management Personal finance management. Small investors.

Application Area: Education

Application Description Users

Course listings Extension courses, night school classes, private school Extension service offerings available by subject, location, fee, etc. students.

Computer-assisted Course material programmed to move with individual instruction/ learner's speed and capability. Capable of giving Computer-managed homework assignments and testing. instruction

Special services for Interactive "correspondence" courses. Handicapped students. homebound students

Supplemental materials Online "Sesame Street" type materials to practice Target student group, for educational TV ’lessons/ideas from broadcast program. programs

Do-it-yourself training Step-by-step instruction for home repair, car repair, etc. Could be offered by school system or product manufacturer,

Literacy training Basic language and mathematical skills. New citizens, anyone needing help to upgrade literacy capability.

Retraining Tutorial programs including linking new job interests Unemployed, laid-off and positions available. workers. Application Area: Person-to-person communication, person-to-group communication

Application Description Users

Internal business Intra-office memos; closed user groups. communication

Serendipity machine Unplanned meeting with people of like interests via system.

1 'Videot exgram" Efficient way of sending same message to multiple locations.

Conferencing Shared textual space by group (other media could be added).

Electronic welcome Information on community and neighbors provided by Newcomers to community, wagon individual who serves as online welcomer.

Application Area: Library and Reference Service

Application Description Users

Catalogue review Listing of all library materials. Capability of Library users, reserving/requesting items from local or distant libraries.

Data base access Opportunity to access specialty data bases from home, Researchers, planners. allowing some work (e.g., legal research) to be done at home. 222 Application Area: Business, Office of the Future

Application Description Users

Telework Text editing, file maintenance, data entry and analysis. Office workers

Extensions of corporate Available to workers at home and at other convenient Office workers. management systems locations (e.g., hotel rooms).

Inventory/stock Extension of present computerized systems to smaller Franchise outlets, control and more decentralized companies and franchise organiza­ small businesses. tions.

Application Area: Banking

Application Description Users

Electronic checkbook Purchase made over system; purchase amount automatically Bank and savings and deducted from purchaser's checking account. loan customers, auto­ mated teller users.

Electronic funds * Bank customer has access to all accounts via home termin­ Bank and savings and transfer al. Can manipulate accounts and make transactions. loan customers, auto­ mated teller users.

Electronic credit cards Account number is entered in system for purchases, bill Credit card users and paying, etc. companies. 223 Application Area: Health

Application Description Users

Store-front medicine Autodiagnostic systems. General public, personnel offices.

First-aid information What to do in case of ..., could include follow-up numbers to call.

Health and safety Assist in at-home care of chronically ill, elderly, etc. monitoring systems

Application Area: Electronic Publishing

Application Description Users

Specialized newsletters Professional, high-cost newsletters with timely information.

Electronic newspapers Text or headline service from local papers and wire services. Ability to retrieve background stories.

Supplement to TV Critiques, references, explanation of ideas not TV viewers programs explained fully in program due to broadcast restraints. APPENDIX C

CHANNEL 2000 QUESTIONNAIRES This questionnaire is for

General Instructions 1. This questionnaire is to be completed only by the person named in the upper right hand corner of this page.

2. Alt of your opinions and answers are strictly confidential.

3. Do not discuss the questions with other family members.

4. Please answer all the questions. Do not skip any.

5. Move quickly through the questionnaire. Do not spend too much time on any one question.

REMEMBER TO RETURN THE QUESTIONNAIRE, ALONG WITH YOUR FAMILY’S QUESTIONNAIRES, IN THE POSTAGE PAID ENVELOPE BY SEPTEMBER 21. Ill SECTION A: ABOUT INTERESTS AND OPINIONS

In this section, a number of statements are listed about interests and opinions. For each state* ment listed, we’d like to know whether you personally agree or disagree with the statement After each statement, there are six numbers from 1-6. The higher the number, the more you tend to agree with the statement. The lower the number, the more you tend to disagree with the state­ ment. The numbers from 1-6 may be described as follows: 1. i definitely disagree with the statement 2. I generally disagree with the statement 3. I moderately disagree with the statement 4. I moderately agree with the statement 5. I generally agree with the statement 6. I definitely agree with the statement For each statement, please circle the number that best describes your feelings about that state­ ment. You may think many items are similar. Actually, no two items are exactly alike, so be sure to circle one number for each statement.

Jr .J£ tr &£ V a r . $ tfJ <£/V $■ V V B1 To keep up with the news, I primarily rely on television ...... 5" 11 11 33 26 14

B2 I have more self confidence than most of my frie n d s ...... 3 5 15 28 36 13 \ B3 Television is my primary form of entertainment ...... 11 18 12 23 25 11 ** B4 I have somewhat old-fashioned tastes and habits...... 5 14 16 26 25 14 1

B5 1 find that 1 have less time to do the things 1 want to do than 1 used to 4 6 10 18 27 36

B6 1 enjoy science fiction ...... 18 ■14 9 16 18 26

B7 1 would go to the library more often if 1 could find the tim e...... 7 9 9 23 23 26 1

B8 f don't like to take chances ...... 10 15 23 21 20 11 1

B9 1 wish 1 didn't have to spend so much time waiting in lines ...... 1 1 4 15 19 60 1

BIO My opinions on things don't count very m u c h ...... 31 31 18 13 4 3

B l l 1 usually read the sports page in the daily newspaper...... 34 18 10 11 10 18

B 12 1 always find plenty of time available to get all my necessary reading d o n e ...... 28 31 18 10 8 4. 1

B 1 3 1 use many of the brands and products that my parents used to use . 11 22 19 32 9 6 1

B 14 1 am usually among the first to try new products ...... 4 8 16 26 28 18

B 15 Information from advertising helps me make better buying deci­ sions ...... 6 7 18 39 20 11

B 16 1 worry a lot more these days about the security of my family than 1 used t o ...... 4 8 16 26 28 18

B 17 Magazines are more interesting than television...... 8 11 30 27 16 8 B 18 1 like to be considered a leader ...... 1 4 5 13 38 21 * Numbers represent percentages (N=l4l) ** No response/missing data B19Today, most young people don't have enough discipline ...... * B20| don’t like television news programs ......

B21The man should be the boss in the house......

B 2 2 Sometimes I fee! that there is so much news available that I can’t keep up with i t ......

B 2 3 1 am in favor of very strict enforcement of all law s ......

B24 Getting information by TV won’t ever take the place of reading a book or newspaper article ...... - ......

B25 Everything is changing too f a s t ......

B2 6 Television advertising insults my Intelligence ......

* Numbers represent percentages (N=141) ** No response/missing data 229 SECTION B: YOUR OPINIONS ABOUT CHANNEL 2000

YOU MUST READ THESE NEXT PARAGRAPHS IN ORDER TO COMPLETE THE REST OF THIS QUESTIONNAIRE.

What does Channel 2000 do?

Channel 2000 is the name for a service that allows you to use your television set in a variety of new ways. By pressing the appropriate buttons on the Channel 2000 keypad (see diagram below), you can have printed information appear on your own TV screen. For example, by pressing certain buttons you can look things up in an encyclopedia that will appear as printed words on your TV screen. By pressing other buttons you can look up what books are in your local library and have them mailed to your home. Or, by pressing others, you can find out what events are happening in your community. If you have a BANK ONE checking account you can look up your balance; or you can pay your bills.

How does Channel 2000 work?

Channel 2000 is an electronic device that connects your television set to an information center located in Columbus. To hook up to this information center, you just dial a local telephone number and then place your telephone receiver onto the Channel 2000 unit.

A diagram of what Channel 2000 looks like is directly below.

Channel 2000 U n it *

© © ■«Your TV Set

Telephone*

Channel 2000 -« Keypad

Now that you have a general idea of what Channel 2000 is, we would like you to answer some ques­ tions about yourself and how you feel about Channel 2000.

PLEASE ANSWER THESE QUESTIONS YOURSELF AND DO NOT DISCUSS THEM WITH OTHER MEMBERS OF YOUR FAMILY. THANK YOU. 230 V a r . #

Please circle the number that gives your answers. The numbers for your answers are to the right of each question. & «.* <1 * . * f # tiP 'A v* , * f j / / / / it B27 I really don't yet understand how 1 will operate Channel 2000 ...... 16 33 17 19 7 6

B 28 Channel 2000 would be worth every cent it costs when people can eventually subscribe to i t ...... 2 4 13 52 17 9

B29 Channel 2000 would offer real advantages over the way 1 now do th in g s...... 1 2 6 34 36 19

B 30 | expect Channel 2000 to be very useful ...... 1 2 4 31 32 30

B 31 Channel 2000 would help me save time paying b ills ...... 6 3 9 28 26 28

5 3 2 Channel 2000 would help me save time getting books from the library...... 4 0 7 18 29 40

B33 1 would much rather go to the library than order books through Channel 2000 ...... 21 22 30 16 6 4

B 34 in the long run, even though 1 could get library books through Chan­ nel 2000,1 would probably go to the library as often as 1 do n o w___ 21 21 25 16 10 6

B35 In the long run, even though 1 could pay bills over Channel 2000, 1 would probably continue to pay my bills the way I do n o w ...... 21 21 25 16 . 8 9

B36 To me, Channel 2000 seems to be a luxury item ...... 6 7 15 41 16 14

B37 Channel 2000 would probably cost a lot of money ...... 1 6 18 37 20 IS

B38 My family might object if 1 subscribed to Channel 2000 ...... 37 23 21 11 3 5

B39 Channel 2000 is not a product intended for a person like m e ...... 35 27 24 6 4 3 B40 Evaluating how Channel 2000 will perform is not something that 1 am very confident in doing ...... 41 25 17 8 3 5

B41 If Channel 2000 cost $10.00 per month, I would probably subscribe . 8 11 27 26 17 8

B42 I really need a service like Channel 2000 ...... 5 12 32 34 11 5

B43 I would have to change some of the ways I do things if I used Chan­ nel 2000...... 4 6 16 39 23 11

B44 I think Channel 2000 would be boring to use...... 41 34 18 3 1 1

B45 1 have heard about or seen devices like Channel 2000 belore ...... 34 12 7 19 11 16 B46 1 would have to have a lot of technical know-how to use Channel 2000...... 48 26 17 6 1 0

B47 If Channel 2000 cost $5.00 per month, 1 would probably subscribe .. 2 4 8 26 23 36

* Numbers represent percentages (N=141) ** No response/ missing data 231

/ 4 7 * & f. tr•A* tr •«? a* ct

V a r . # / fe«- 6* <5> /

B48 1 really wouldn't lose much if 1 subscribe to Channel 2000 and it . u JLJ, didn't work out as 1 thought ...... 6" 4 16 36 25 13 1 B49 | wouldn't have to spend very much money to subscribe to Channel 2000 in order to try it o u t ...... 2 6 16 33 23 21

B50 |f Channel 2000 cost $15.00 per month, 1 would probably subscribe . 32 23 19 16 4 5 1

B51 1 expectChannel 2000 will become a real necessity ...... 8 11 31 31 11 9

B52 1 think if 1 subscribe and use Channel 2000,1 would have no difficul­ ty telling my friends what it was lik e...... 1 1 2 9 36 52

B53 Channel 2000 is not the sort of product that 1 would tell my friends about after 1 subscribe to i t ...... 57 29 7 2 2 2

B5 4 | don’t think Channel 2000 would make my banking any easier to do. 29 19 23 16 7 6

B55 | have no difficulty understanding how Channel 2000 technically is going to w o rk...... 4 4 20 19 26 27

B56 The public libraries should not spend tax dollars on services like Channel 2000 ...... 25 28 26 9 4 7 1 B57 It would be harder to pay bills using Channel 2000 than it is for me 31 22 33 9 1 4

B58 channel 2000 would limit my personal privacy ...... 38 21 16 15 4 6 1

B59 If Channel 2000 cost $12.50 per month, I would probably subscribe . 18 23 26 17 11 5

B60 I think Channel 2000 would fit right into the way I choose to live my 6 4 9 40 28 12

B61 1 would probably be permitted to subscribe to Channel 2000 on a trial basis for a month or so...... : ...... 2 0 5 32 37 23 1 B62 If 1 subscribe to Channel 2000 It is likely that 1 would not be satisfied with the way it perform s...... 16 39 36 6 2 1

B63 If Channel 2000 cost $7.50 per month, 1 would probably subscribe .. 7 7 14 36 19 17 B64 1 think that If 1 subscribe to Channel 2000 most of my friends would approve ...... 1 1 4 26 40 28 1 B65 Channel 2000 would probably give my family something to talk about...... 2 1 2 26 34 36

B66 It’s very difficult for me to evaluate how my friends will react to my subscribing to Channel 2000 ...... 16 23 23 20 13 5

B67 Channel 2000 would probably cut down the amount that we talk to each other in my fam ily ...... 46 30 18 2 2 1 * Numbers represent percentages (N=141) ** No response/missing data 232 SECTION C: ACTIVITIES IN PAST YEAR

This section lists some things that you may or may not have done. For each activity listed, please circle the appropriate number to indicate how often during the past 12 months you, yourself, have engaged in this activity. (If you have not taken part in one of the listed activities during the past 12 months, circle the “None in the Past Year" for that activity.) Please make sure that you circle a number for each activity and that you circle only one number for each activity.

No. of Times Participated in Past 12 Months v # CIRCLE ONLY ONE NUMBER None * FOR EACH STATEMENT In Past 1-4 5-8 9-11 12-24 25-51 52 or Year Tim es Tim es Tim es Tim es Tim es M ore * irk B68 Read a book...... 10 6 19 18 13 20 14 l

B69 Looked up something in an encyclopedia ...... 21 21 14 8 17 9 9 2

B70 Used a hand-held calculator ...... 6 3 2 8 11 16 53 1

B71 Used an automatic bank teller machine ...... 45 10 2 4 11 11 14 1

B72 Operated a com puter ...... 50 13 4 5 5 6 16 1

B73 Ordered some merchandise from a mail order catalog using an 800 (toll free) telephone num ber ...... 54 31 8 3 4 0 1 1

B74 Used a typewriter ...... 14 14 12 11 11 7 31 1

B75 Used the telephone to call for weather information ___ 16 13 12 10 17 13 19 1

B76 Used a credit card ...... 9 4 4 6 15 18 43 1

B77 Flew on an airplane for a business trip ...... 69 21 6 0 2 0 1 1

B78 Balanced my checkbook ...... 25 7 9 9 36 4 8 3 q 79 Went to a library ...... 11 20 11 18 18 14 6 2

B80 Checked out a book from a lib ra ry ...... 23 18 17 14 18 8 1 1

B81 Called the bank to get information about my checking account ...... 39 33 1.4 8 4 1 1 1

B82 Ordered a book from the library and received it by mail ("Books By Mail" Service) ...... 92 6 1 0 1 0 0 1

* Numbers represent percentages (N=14l) ** No response/missing data 233 SECTION D: LIBRARY USE AND BANKING Var. # Now J few questions about libraries. Please circle the one number that gives your answer.

B83 1. Some libraries now have a machine that looks like a TV that you can use to look 1. yes 57 up a book or author, instead of the card 2, no (please skip to #3 below If you have catalog. Have you used one of these not used one) 42 new machines? No response 1 B 8 4 2. If you have used the machine to look up a book, how did you feel about it? Was it 1. easier 38 easier or more difficult than the card 2. more difficult 6 catalog? 3. no difference 13 No response 43 B85 3. What is the principal type of activity that you do at the library? (Place a check school related work 21 next to your most frequent activity.) career/job related work 19 personal reading 35 Information unrelated to school or job 14 check here if you don’t use the library 8

B8S 4. Does your household own an No response 3 encyclopedia? 1. yes 65 2. no 34 9. don’t know 1

Now a few questions about banking.

B 8 7 1. When the bank sends a statement at the end of the month some people handle it differently. 1. balance it exactly 57 Which of these fits best with the way you 2. quick looking over 32 handle your checking account statement? 3. let the bank tell me if there’s a problem 6 No response 6 B88 2. Do you usually handle your banking during regular bank hours (9:00 AM -5:00 PM) or ■ 1. during regular banking hours 70 do you take care of it at another time 2. at another time 28 (say, after work,1 or on Saturday)? No response 2 B89 3. In an average month, approximately how many checks do you personally write? 1. none 14 2. 5 checks 12 3. 6-15 checks 31 4. 16-25 checks 24 5. 26-35 check3 11 6. 36 or more 6

B90 4. Do you use a telephone bill payment No response 1 service offered by one of the local banks? 1. yes 13 2. no 87 No response 1

* Numbers represent percentages (N=141) 234

People find certain things to be difficult for some and easy for others. j

Please tell us how easy or how difficult each of the following things is for you to get done now. Circle the number that gives your answer. j

Var. « & f J ? .t? & I A B 9 1 1. Finding out about a check you have written that has not yet been returned by the bank * ** with your statement?...... 13 16 19 14 8 7 21 3

B 92 2. Paying your bills (not making the money to pay them, but actually going through the process of paying your bills)? ...... 20 28 24 13 4 2 ! 9 1

B 93 3. Finding the answer to a particular question In an encyclopedia ...... 22 23 23 16 4 3 9 1

B 94 4. Looking up a book in the card catalog in the library...... 23 26 21 14 6 3 ! 6 2

* Numbers represent percentages (N=141) ** No response/missing data SECTION E: RADIO / TV / NEWSPAPERS / BOOKS 235

In this section we would like to know how much you currently use radio, television, newspapers, and books. First we would like to know how much time you spend on an average weekday (Iram when jlou wake up until when you go to bed) listening to the radio, watching TV, or reading a newspaper. Please answer these questions according to how much time you currently spend doing iach activity for one average weekday (Monday-Friday). V a r . # B 95 1. fedio: On one average weekday, how much time do you usually spend listening to the radio?...... 1. none 2. less than 15 minutes 10 3. 16-30 minutes 21 4. 31-60 minutes 24 5. more than 1 hour but less than 3 hours 30 6. more than 3 hours but less than 6 hours 9 7. more than 6 hours 6

B 9 6 ^Television: On one average weekday, how much time do you usually spend watching TV?...... 1. none 2. less than 30 minutes 1 3. 31-60 minutes 16 4. more than 1 hour but less than 2 hours 21 5. more than 2 hours but less than 4 hours 40 6. more than 4 hours but less than 6 hours 16 7. more than 6 hours but less than 8 hours 4 8. more than 8 hours 1

B 9 7 3. Newspapers: On one average weekday, how much time do you usually spend reading the newspaper? ...... 1. none 1 62. less than 15 minutes 162. 3. 16-30 minutes 45 4- 31-60 minutes 23 5. more than 1 hour but less than 2 hours 9 6. more than 2 hours 3

B 9 8 4. Newspapers: How many newspapers do you read during an average 7 day week? (If you read one newspaper per day, you would have read 7 newspapers during an average 7 1. none 6 2- 1-3 17 3. 4-6 19 4. 7-9 34 5. 10-13 9 6. 14 or more 15

* Numbers represent percentages (N=141) 236

5. Newspapers: Here Is a list of some of the parts of a daily newspaper. Read through this list and circle the number that best describeswwwuw*... your jww. normal...... reading • - — ■■a habits for -N that newspaper section. <4^

//// ✓ V a r . # £ £ #* ^ ^

•ft ‘Arflr B 99 Main news s e c tio n 31 40 26 1 1 2

B100 Frontpage 53 33 12 1 1 1

B i d Editorial pages 9 16 50 18 5 3

B 102 Movie and amusement pag es... 11 22 58 6 0 4

B 103 TV schedules 12 27 48 11 1 2

B104 Classified pages (want ads) 4 11 53 27 4 3

B105 Sports section 12 17 37 30 2 2

BIO 6 Business & financial news 9 17 43 25 2 4

B107 Stock market tables 4 11 27 53 2 3

6. Television: Which of the following cable TV * systems do you subscribe to? ...... 1. QUBE 16 2. All American Cable 3 B 108 3. KBLE 1 4. Although cable TV is available in my neighborhood, I don’t subscribe 41 5. Cable TV is not available in my neighborhood 32 9. I don't know if cable TV is available in my neighborhood 4

Television: Do you subscribe to any No resP°nse 4 premium movie service such as Home 1. yes 9 Box Office, The Star Channel, Show 0 __ on Time, etc.? a n0 89 3. don't know BX09

* Numbers represent percentages (N=141) ** No response/missing data 237 SECTION F: ABOUT YOU

Now think about what you currently know about Channel 2000 and how you would feel about subscribing to a service like Channel 2000. Select a number from the ones listed beiow that describes the chances that you would subscribe to Channel 2000 if it did come out on the market. If you are absolutely certain you would subscribe, the number you should circle is "tOO." If you are absolutely certain you would not subscribe, the number you should circle is "0." If you think the chances are somewhere in between, circle one of the other numbers that seems most ap­ propriate. Again, we are only interested in yaur opinions since Channel 2000 will not be sold to you at any time during this test. Var. # 1. Please circle the one number you feel best describes your feelings at this time.

B110 0* 10 20 30 40 50 60 70 80 90 100 1568 6 23 16 16 846 1 I absolutely would not I absolutely would subscribe to Channel 2000 subscribe to Channel 2000

We now have a short series of questions that wilt be used for classification purposes only. We assure you that your responses are strictly confidential. * 2. Which category includes your age? 1. 12-15 years 0 (Circle one number) 2. 16-17 years 6 Bill 3. 18-21 years 6 4. 22-24 years 5 5. 25-34 years 37 6. 35-44 years 19 7. 45-54 years 17 8. 55-64 years 6 9. 65 years or older 4

3. Are you currently ...... 1. Married 78 B 112 2. Widow/Widower 1 3. Divorced/Separated 1 4. Single 20

4. Are y o u ...... 1. Male 49 2. Female 51 B 113 5. What is the highest level of school 1. Grade school or less 1 you completed? 2. Some high school 10 B 114 3. High school graduate 14 4. Some college 23 5. Two year college graduate 5 6. Four year college graduate 23 7. Postgraduate 24

* Numbers represent percentages (N=14l) ** No response/missing data 238

6. How long have you lived at your 1. Less than 1 year 6 present address? 2. 1*5 years 38 B 1 1 5 3. 6*10 years 23 4. 11-20 years 23 5. More than 20 years 9

7. Are you employed outside your home? 1. Employed full time 59

B 1 16 2. Employed part time 19 3. Not employed outside home 21

8. If you are employed outside your home please No response l • describe what you do at work. Please do not say something like "self employed," or "I work for the government." For example, you might say:

I work in an office as a secretary, or I drive a train as an engineer or I am a high school student or I drive a COTA bus and so f o r t h ...... (print in your response) B 1 1 7

* Numbers represent percentages (N=14l)

Thank you for your cooperation and help! Remember to place your questionnaire along with your family’s questionnaires in the stamped envelope and drop it in the mail. If you live alone, just return your own questionnaire.

YOUR OPINIONS COUNT!! Thanks again. This questionnaire is for

QUESTIONNAIRE U “ Now What Do You Think?”

General Instructions 1. This questionnaire is to be completed only by the person named in the upper right hand corner of this page.

2. Alt of your opinions and answers are strictly confidential.

3. Do not discuss the questions with other family members.

4. Please answer all the questions. Do not skip any.

5. Move quickly through the questionnaire. Do not spend too much time on any one question.

REMEMBER TO RETURN THE QUESTIONNAIRE, ALONG WITH YOUR FAMILY'S QUESTIONNAIRES, IN THE POSTAGE-PAID ENVELOPE BY DECEMBER 1. 240 SECTION A: YOUR OPINIONS ABOUT CHANNEL 2000

Now that you have had a chance to see how Channel 2000 works and what services it offers, we would like you to respond to several statements about Channel 2000.

Naturally we are interested in whether your feelings have changed or have remained the same, so many of these statements are similar to those in the first questionnaire.

For each statement, please circle the number that best describes your feelings about the statement. You may think many items are similar. Actually, no two items are exactly alike, so be sure to circle one number for each statement.

# 'f ■ * * *

/ / / > ’ < < V a r . # * f / / / j t>® &

C l To me, Channel 2000 seems to be a luxury ite m ...... 5 8 13 32 23 u 6

C2 Channel 2000 would probably give my family something to talk 5 6 9 23 33 18 6

C3 1 have no difficulty understanding how Channel 2000 technically w o rk s...... -...... 6 6 8 9 30 33 8

C4 1 don't think Channel 2000 would make my banking any easier to 14 21 18 11 15 13 9

C5 1 really wouldn't lose much if 1 subscribed to Channel 2000 and it didn't work out as 1 thought...... 13 17 19 20 16 7 8

C6 If Channel 2000 cost $10.00 per month, 1 would probably subscribe ...... 29 23 21 10 9 3 6

C7 Channel 2000 offers me real advantages over the way 1 used to 18 16 6 6 do things...... 16 16 21

C8 1 think Channel 2000 is boring to u s e...... 17 29 23 14 6 4 6

C9 t think Channel 2000 would fit right into the way 1 choose to live my life ...... 15 15 16 26 15 8 6

CIO Channel 2000 would probably cut down the amount that we talk to each other in my fa m ily ...... 36 35 16 2 3 1 7

C l l If Channel 2000 cost $7.50 per month, 1 would probably subscribe ...... : ...... 27 14 18 15 14 6 7

C12 Channel 2000 is not a product intended for a person tike m e ___ 27 18 20 11 9 7 7

C 13 I would much rather go to the library than order books through Channel 2000 ...... 19 18 17 12 14 13 7

C l i 1 really need a service like Channel 2000 ...... 23 10 25 23 8 4 7

CIS Channel 2000 would be worth every cent it costs when people can eventually subscribe to i t ...... 14 14 22 28 10 5 7

* C 16 In the long run, even though 1 could pay bills over Channel 2000,1 would probably continue to pay my bills the way 1 used t o ...... 15 16 14 19 14 14 9

* Numbers represent percentages (N=141) ** No response/missing data 241 S/ /V ✓ var. f / / / / ✓ / / /V / & V C l7 It’s very difficult for me to evaluate how my friends wilt react to tflr ** my subscribing to Channel 2000 ...... 16 26 22 16 9 5 6

28 26 18 6 8 4 9

C19 If Channel 2000 cost $15.00 per month, I would probably subscribe ...... 55 20 13 4 1 ■ 1 7

27 14 9 16 14 13 8

6 11 10 29 23 12 9

C22 Channel 2000 is not the sort of product that I would tell my 38 25 15 6 5 5 6

11 10 11 23 18 17 10

C24 Evaluating how Channel 2000 performs is not something that I am very confident in doing ...... 23 24 20 14 7 6 6

C25 Channel 2000 would probably cost a lot of money 77 16 16 19 19 23 23 18 9 9

C26 I wouldn't have to spend very much money to subscribe to 6 11 20 24 18 14 7

18 16 27 15 9 8 8

C28 Channel 2000 would help me save time getting books from the library...... 10 9 6 28 17 23 7

C29 |f Channel 2000 cost $5.00 per month, I would probably subscribe ...... 22 6 10 23 14 18 8

C30 If I subscribe to Channel 2000 it is likely that I would not be satisfied with the way it perform s...... 9 26 21 14 10 13 8

C31 It would be harder to pay bills using Channel 2000 than it is for me n o w ...... 23 22 26 9 4 6 9

C32 My family might object if I subscribed to Channel 2000 ...... 40 23 18 6 4 4

C33 I need a lot of technical know-how to use Channel 2000 51 22 14 4 2 l

C34 I think if I subscribe and use Channel 2000, I would have no difficulty telling my friends what it is lik e ...... 4 1 3 13 34 39

C35 If Channel 2000 cost $12.50 per month, I would probably subscribe ...... 38 28 16 7 3 0

C36 t think that if I subscribe to Channel 2000 most of my friends would approve ...... 4 4 5 27 34 20

C37 The public libraries should not spend tax dollars on services like 22 22 20 6 8 15

2 2 1 9 31 47 * Numbers represent percentages (N=141 ) ** No response/missing data 242 The statement fits with my opinion: var- * / > ./ /■ /. f JS c ? V C55 in hunting for the best way to do something, it Is usually a good 6 * Idea to try the obvious way first ...... 23 30 35 5 1 6 * *

C56 1 like to wait until something has been proven belore 1 try i t ...... 4 14 38 32 7 6

C57 My table manners at home are as good as when 1 eat out in a restaurant...... 18 36 24 11 4 6

C58 When it comes to taking chances, I would rather be safe than so rry...... 15 18 43 14 4 6

C59 1 like people who are a little shocking...... 8 16 32 30 7 7

C60 When 1 see a new brand on the shelf, 1 often buy It just to see what it is lik e...... 12 12 31 31 7 6

C61 1 feel that too i.'uch money is wasted on new styles ...... 9 14 27 31 11 7

C62 1 am sometimes Irritated by people who ask favors of m e ...... - 2 9 20 45 18 6

C63 1 often try new brands before my friends and neighbors d o...... 9 19 43 16 7 6

C64 1 enjoy being with people who think like 1 d o...... 14 40 33 4 2 8

C65 At work, 1 think everyone should work on only one thing thereby becoming more of an e x p e rt ...... 2 4 11 48 28 7

C66 1 like to experiment with new ways of doing th in g s ...... 25 34 28 5 1 7

C67 1 have never intensely disliked anyone ...... 11 10 24 23 25 6

C63 In the long run the usual way of doing things is the best...... 4 7 32 43 8 7

C69 Some modern art is stimulating ...... 15 21 31 19 8 6

C70 1 like to tool around with new ideas even if they turn out later to be a total waste of time ...... 16 28 36 10 5 6

C71 Today is a good day to start a new project...... 25 31 30 5 4 6

* Numbers represent percentages (N=14l) ** No response/missing data t>* cf 6*, C39 In the long run, even though I can get library books through * Channel 2000,1 probably go to the library as often as I used to... 14 14 11 16 19 19

C40 | have had to change some of the ways I do things since I’ve been using Channel 2000 40 18 16 12 3 4

C41 I would probably be permitted to subscribe to Channel 2000 on a trial basis for a month or s o...... 4 2 6 31 26 23

SECTION B: MORE OPINIONS

On the following pages are statements that reflect people’s opinions. They are not right or wrong, but just opinions. Please show how well each of these statements of opinion fits in with your views or feelings by circling a number to the right of each statement.

The statement fits wtth my opinion:

/ > v

C42 1 always try to practice what 1 preach ...... 21 35 35 4 0

C43 1 like to take a chance ...... 4 26 36 18 9

C 44 1 don’t like to talk to strangers...... 5 11 28 29 20

C45 The unusual gift is often a waste of m oney...... 3 10 18 41 21

C46 1 enjoy looking at new styles as soon as they come o u t ...... 21 21 26 18 6

C47 1 never resent being asked to return a favor ...... 19 38 28 6 2

C48 Buying a new product that has not yet been proven is usually a waste of time and m oney...... 6 7 26 42 14

C49 Often the most interesting and stimulating people are those who don't mind being original and different ...... 36 36 16 3 3

C50 1 would like a job that required frequent changes from one kind of task to another...... 36 31 18 6 4

C 51 If people would quit wasting their time experimenting, we would get more accomplished ...... 2 1 4 28 58

C52 1 have never said something to deliberately hurt someone’s feelings...... 13 22 31 22 6

C53 If 1 got an idea, 1 would give a lot of weight to what others think of I t ...... 11 28 36 14 6

C54 1 like to try new and different things ...... 31 31 26 4 1

* Numbers represent percentages (N=14i) ** No response/missing data 244 SECTION C: SUBSCRIBING TO CHANNEL 2000

Now think about what you currently know about Channel 2000 and how you would feel about subscribing to a service like Channel 2000. Select a number from the ones listed below that describes the chances that you would subscribe to Channel 2000 if it did come out on the market. If you are absolutely certain you would subscribe, the number you should circle is "100." If you are absolutely certain you would not subscribe, the number you should circle is "0". If you think the chances are somewhere in between, circle one of the other numbers that seems most appropriate. Again, we are only interested in your opinions since Channel 2000 will not be sold to you at any time during this test.

V a r . # C72 • Please circle the one number you feel best describes your feelings at this time.

10 20 30 40 50 60 70 80 90 100 24” 10 6 8 6 14 10 7 6 6 4 I absolutely would not I absolutely would subscribe to Channel 2000 subscribe to Channel 2000

SECTION D: YOUR BEHAVIOR

In this section we are interested in whether Channel 2000 has changed the way you behave. Remember to answer the questions according to how you currently feel. Please circle the number that best describes your feelings about that statement.

^ <& <£> / 4 / SINCE 1 HAVE HAD CHANNEL 2000 IN MY HOUSE IT SEEMS THAT . . . «* vO" V * > * C 7 3 1. I listen to the ra d io...... 0 2 87 2 2 6**

C74 2. I watch television ...... 0 3 82 8 1 6

C 75 3. I read magazines...... i 5 83 2 1 3

C 76 4. I read the new spaper...... l 4 84 3 1 6

C 77 5. I read books...... i 9 78 4 1 7

C78 6. I talk to my fam ily...... l 7 84 1 1 7

C79 7. I do banking-related activities ...... 2 7 79 3 1 8

C80 8. 1 am aware of community events ...... 4 43 44 0 1 8

C 81 9. 1 talk to my children (if applicable) ...... 0 9 50 1 1 40

C82 10. My knowledge of library services I s ...... 5 37 50 1 1 7

* Numbers represent percentages (N=14l)

** No response/missing data 245 SECTION E: RATE THE CHANNEL 2000 SERVICES

Listed below are questions and statements about several of the services available on Channel 2000. Please tell us how often you use each service, how easy or difficult it is for you to use the service, and how useful or useless the service is by circling the number to the right of each item. Video Catalog

6 4-5 2*3 Once V a r ., # limes times times a Everyday a week a week a week week Never ^ C83 1. How often do you use the * 9? l l " video catalog? ...... 1 4 4 1 3 45 22 (if 6, skip to #4)

Very Very Easy Difficult C84 2. I find using the video catalog i s ...... 16 21 14 11 5 4 29

Very Very Useful Useless C85 3. For me, the video catalog I s ...... 9 16 IB 10 11 7 29

Video Encyclopedia

6 4 5 2*3 Once limes limes tirnsa a Everyday a week a week a week week Never C86 4. How often do you use the 1 3 video encyclopedia? ...... 0 2 4 1 5 44 22 (if 6, skip to #7)

Very Very Easy Difficult C87 5. I find using the video encyclopedia i s ...... 14 24 12 14 3 3 31

Very Very Useful Useless C88 6. For me, the video encyclopedia i s ...... 1 0 16 18 1 6 4 6 29

Home Banking

With the Channel 2000 home banking service, it is possible for you to check the balance you have In any of your accounts, pay your bills, or examine your monthly banking transactions.

Yes No C89 7. Do you use any of these Channel 2000 home banking 19 services? ...... 17 64 (if 2, skip to #17)

6 4-5 2*3 Once times limes times a Everyday a week a week a week week Never C90 8. How often do you check the balance In any of your accounts with the Channel 75 2000 home banking service? 1 1 1 6 7 9 (if 6, skip to #11)

* Numbers represent percentages (N=14l) ** No response/missing data 246 V a r . i Very V ery Ea9y D iffic u lt C91 9. I find that checking the balance in any of my accounts with the Channel . * •

Very Very U seful U seless C92 10. For me, being able to check the balance in any of my accounts with the Channel 2000 home banking service i s ...... 9 3 1 3 2 3 80

6 4-5 2-3 O nce tim es tim e s tim es a Everyday a week a week a w eek w eek N ever How often do you pay your bills with the Channel 2000 75 home banking service? ...... 0 0 1 3 9 12 (if 6 , Skip to #14)

Very Very Easy D iffic u lt I find usii.g the Channel 2000 bill payment service i s ...... 11 3 0 0 0 2 84

Very Very U seful Useless For me, the Channel 2000 bill payment service i s ...... 6 4 2 0 1 2 84

6 4-5 2-3 O n ce tim es tim es tim es a Everyday a w eek a w eek a w eek w eek N ever C96 14. How often do you examine your monthly banking trans­ actions with the Channel 2000 home banking service? 0 0 1 2 13 9 (If6, skip to^17)

Very V ery Easy D iffic u lt C97 15. I find that examining my monthly banking transac­ tions with the Channel 2000 home banking service Is ----- 11 3 1 1 0 3 82

Very Very U seful U seless C98 16. For me, being able to examine my monthly bank transactions with the Channel 2000 home banking service is ...... 5 6 2 0 1 4 82

* Numbers represent percentages (N=141) ** No response/missing data 247

Columbus Calendar

6 4-5 2-3 O nce Var. # tim es tim es tim es a Everyday a w eek a w eek a w eek w eek N ever C 99 17. How often do you use the Columbus Calendar 11 Service? ...... A 21 52 11 (If 6, skip to Section F)

Very Very Easy Difficult C l0018. I find using the Channel 2000 Columbus Calendar service i s ...... 53 16 10 18

Very Very U seful U seless C 1 0 1 1 9 - For me, the Channel 2000 Columbus Calendar service i s ...... 16 16 23 11 11 17

SECTION F: YOUR CHANCE TO TELL US

Below we have several questions that allow you to tell us about your experiences with Channel 2000. This kind of information is extremely valuable to us and usually quite difficult to capture in standard questionnaire format. So, please read each question carefully and print or write clearly your responses.

1. What do you like about Channel 2000?

2. What do you dislike about Channel 2000?

* Numbers represent percentages (N=14l) ** No response/missing data 248

3. What are some of the problems that you have had with Channel 2000?

4. What additional or other kinds of services or features would you like to see on Channel 2000?

Var. # C 102 5. And finally, since you have had Channel 2000, how many different people (outside your im­ mediate family) have you shown Channel 2000? (circle your answer)

0 1 1-3 4-6 7-10 11-15 16 or more I haven't person people people people people people shown it to anyone * 13 3 20 23 20 11 11

* Numbers represent percentages (N=14l)

Again we thank you for your cooperation. Please place your questionnaire along with your family’s questionnaires in the stamped envelope, and drop it in the mail. If you live along, just return your own questionnaire.

PLEASE RETURN YOUR QUESTIONNAIRE PACKAGE BY DECEMBER 1.

Having any problems) Call 457-9305 (8:30 a.m.-10:00 p.m.) about Channel 2000. Call 224*0623 (9:00 a.m.-5:00 p.m.) about this questionnaire. This questionnaire is for

QUESTIONNAIRE Iff “ The Final Evaluation”

General Instructions 1. This is the last and most critical questionnaire and is to be completed only by the person named in the upper right hand corner of this page. 2. All of your opinions and answers are strictly confidential. 3. Do not discuss the questions with other family members. 4. Please answer all the questions. Do not skip any. 5. Move quickly through the questionnaire. Do not spend too much time on any one question.

REMEMBER TO RETURN THE QUESTIONNAIRE, ALONG WITH YOUR FAMILY’S QUESTIONNAIRES, IN THE POSTAGE-PAID ENVELOPE BY JANUARY 14, 1981. 250 SECTION A: YOUR OPINIONS ABOUT CHANNEL 2000

We begin with several statements that you have seen before. The purpose of these statements Is to see if your feelings have changed or remained the same during the course of the Channel 2000 test.

For each statement, please circle the number that best describes your feelings about the statement. Move quickly, and be sure to circle one number for each statement.

.tP J5 s/6 / ^ / ^ > ^ \ 4^ Var. # 6*

D2 I understand how to operate Channel 2000 ...... 2 3 4 4 33 52

D3 Channel 2000 is not a product intended for a person like m e 23 .28 17 13 10 6

D4 If Channel 2000 cost $7.50 per month, I would probably subscribe ...... 25 23 14 20 9 9

D5 Channel 2000 would probably give my family something to talk about...... 7 5 9 39 25 15

D6 I have had to change some of the ways I do things since I've been using Channel 2000 ...... 38 22 16 13 6 4

D7 Channel 2000 would probably cost a lot of money ...... 9 24 22 22 16 5

D3 My family might o‘bject if I subscribed to Channel 2000 ...... 38 30 16 7 5 4

D9 Channel 2000 would limit my personal privacy ...... 36 26 15 9 9 5

01 0 I really wouldn’t lose much if I subscribed to Channel 2000 and it didn’t work out as I thought...... 4 11 23 31 22 8

D l l In the long run, even though I could pay bills over Channel 2000,1 would probably continue to pay my bills the way I used t o ...... 12 17 16 18 19 16

D 12 Channel 2000 would help me save time getting books from the library...... 9 12 11 22 23 21

D 13 I h3ve heard about or seen devices like Channel 2000 before .... 32 15 6 18 13 16

D14 1 think that if I subscribe to Channel 2000 most of my friends would approve ...... 3 1 4 26 36 28

D 15 I would probably be permitted to subscribe to Channel 2000 on a trial basis for a month or s o...... 0 4 4 33 34 24

D16 Channel 2000 would probably cut down the amount that we talk to each other in my fa m ily ...... 46 32 14 3 2 1

D 17 If Channel 2000 cost $10.00 per month, I would probably subscribe ...... 37 23 21 10 8 1

D18 If I subscribe to Channel 2000 it is likely that I would not be satisfied with the way it performs...... 9 20 30 16 15 8 * Numbers represent percentages (N=141) ** No response/missing data *ar. # f / y J ? / ^v j T > F j ? ^v t

The public libraries should not spend tax dollars on services tike Channel 2000 ...... 21 26 19 14 8 10 1

D21 1 don’t think Channel 2000 would make my banking any easier to 23 19 16 11 11 2

D22 To me, Channel 2000 seems to be a luxury ite m ...... 4 9 17 31 28 10 1

D23 1 would much rather go to the library than order books through Channel 2000 ...... 16 13 16 23 14 15 3

D24 1 think Channel 2000 would fit right into the way 1 choose to live my life ...... 11 9 14 35 16 12 4

D25 If Channel 2000 cost $12.50 per month, I would probably subscribe ...... 45 28 17 5 1 4 1

D26 Channel 2000 would help me save time paying b ills ...... 13 9 12 23 25 16 2

D27 Channel 2000 is very u s e fu l ...... 7 6 14 34 18 18 2

D28 1 really need a service like Channel 2000 ...... 22 16 24 27 6 3 3

D29 If Channel 2000 cost $15.00 per month, I would probably subscribe ...... 61 19 12 4 2 1 1

D30 1 expect Channel 2000 will become a real necessity ...... 14 18 26 24 10 6

D31 1 need a lot of technical know-how to use Channel 2000 ...... 43 31 18 4 3 1 1

D32 1 think if 1 subscribe and use Channel 2000, 1 would have no difficulty telling my friends what it is l ik e ...... 0 4 4 12 38 40 1

D33 In the long run, even though 1 can get library books through Channel 2000,1 probably go to the library as often as 1 used to... 7 16 18 19 26 14 1

D34 Channel 2000 would be worth every cent it costs when people can eventually subscribe to i t ...... 15 12 13 28 17 5 4

D35 1 have no difficulty understanding how Channel 2000 technically works ...... 4 6 16 18 27 28 1

D36 If Channel 2000 cost $5.00 per month, 1 would probably subscribe ...... 20 12 13 16 20 18 1

D37 Evaluating how Channel 2000 performs is not something that 1 am very confident in doing ...... 23 31 21 11 6 8 1

D38 Channel 2000 is not the sort of product that 1 would tell my friends about after 1 subscribe to i t ...... 31 31 26 3 6 3 1

D39 Channel 2000 offers me real advantages over the way 1 used to do things...... 10 15 18 28 18 9 1

D40 It would be harder to pay bills using Channel 2000 that it is for me now ...... 22 28 31 6 6 4 2

D41 1 wouldn’t have to spend very much money to subscribe to Channel 2000 in order to try it o u t...... 2 8 18 33 23 14 3 * Numbers represent percentages (N=14l) ** No response/missing data 252 SECTION B: SUBSCRIBING TO CHANNEL 2000

Now think about what you currently know about Channel 2000 and how you would feel about subscribing to a service like Channel 2000. Select a number from the ones listed below that describes the chances that you would subscribe to Channel 2000 if it did come out on the market. If you are absolutely certain you would subscribe, the number you should circle is “100." If you are absolutely certain you would not subscribe, the number you should circle is “0". If you think the chances are somewhere in between, circle one of the other numbers that seems most appropriate. Again, we are only interested In your opinions since Channel 2000 will not be sold to you at any time during this test.

1. Please circle the one number you feel best describes your feelings at this time.

0 * 10 20 30 40 50 60 70 80 90 100 16~ 12 6 10 3 13 10 7 11 -9 3 l ’ t absolutely would not I absolutely would subscribe to Channel 2000 subscribe to Channel 2000

2. Next we would like you to order each of the Channel 2000 services. Place a number 1 next to the service that you would most likely pay money to have in your home. Place a number 2 next to your second choice and so on, up to number 7.

Channel 2000 Service Your Numbers Here 2 3 4 5 6 7 D 43 Video C a talo g...... 23 Y 7 ~ vr 4

D 4 4 Video Encyclopedia ...... 16 32 18 i i 12 2 4

D45 Home Banking...... 43 5 10 6 14 5 14

D 4 6 Public Information ...... 7 18 24 19 14 9 5

D 4 7 Columbus Regional Information . . 3 12 13 23 32 8 5

D 4 8 Math That Counts...... 4 4 9 10 10 49 11

D 4 9 Early Reader...... 2 4 3 9 6 18 53

* Numbers rep resen t percentages (N=141) ** No response/missing data 3, A system like Channel 2000 has the capability to offer many other services, some of which are not currently available. Listed below are some of these services. We would like you to tell us how willing you would be to pay $3.00 per month for each service listed.

For each service, please circle one number to the right of each service that best describes your feelings about paying for that service. Don't worry about a total cost for all of the ser­ vices combined. We are only interested in your reaction to one service at a time.

Potential Service From Would you pay S3.00 per Channel 2000 System month for just this service?

? y y . y y y * y y y Access to articles from the local newspaper via your televi­ * sion screen ...... 23 26 15 27 8

.Self instructional educational assistance for children ...... 21 18 16 30 13

.Paying bills directly from your home ...... 11 16 21 32 19

.Look at local and national restaurant m enus...... 21 29 23 19 6

Catalog-type shopping which would allow you to shop at h o m e ...... 4 16 17 43 17

.Getting investment information ...... 14 27 18 30 9

Access to articles from 5 or 6 major national newspapers via your television screen ...... 11 21 23 29 14

.Access to information on public transportation (e.g., schedules, bus routes, e tc .) ...... 21 28 23 18 8

Access to a full set of library services—including the ability to order books from the library's card catalog ...... 4 13 20 39 23

A home security/protection system ...... 3 6 19 39 31

Keep a schedule of your social and other engagements ___ 29 32 19 13 5

_Access to a complete encyclopedia ...... 5 18 23 36 15 • Have an in-home computer ...... 5 6 21 42 24

Access to the classified advertising that appears in news­ papers ...... 18 44 22 10 4

A listing of all local entertainment (e.g., musicals, shows, nightclubs, movies, etc.) ...... 15 23 14 36 11

Review your overall financial situation, including savings, Investments, and insurance programs ...... 13 23 18 30 15

. Being able to play a number of sophisticated electronic vid eo g am es ...... 4 13 12 43 26

Self Instructional educational assistance for adults ...... 8 14 21 40 16

* Numbers represent percentages (N=141) ** No response/missing data 254

Potential Service From Would you pay $3.00 per Channel 2000 System month tor just this service?

A Var. # / V y y A £

D6 9 .Apply fora loan or credit by using this d ev ice ...... 23 31 22 18 3 2

D70 .Having a system that immediately contacts your doctor, and if necessary, a hospital and ambulance in the event of a medical em ergency 6 14 24 36 19 1

D71 .Access to a complete dictionary to find correct spelling and pronunciation of w o rd s 17 27 17 28 9 3

D l l _A system for keeping household financial records ...... 21 23 23 27 6 1

D73 .Provide up to date analysis and recommendations on in­ vestments ...... ^7 25 29 23 5 1

D74 _A control system which manages all household energy.... 6 13 28 41 11 1

D75 .Look at lists of merchants and stores that carry certain ite m s 9 16 27 35 11 1

D76 4. You have just read several potential Channel 2000 services listed above and on the previous page. Now we would like you to put in order the 6 services you would most likely pay to have in your home on a monthly basis.

Dll- Place a number 1 next to the service that you would most likely pay for to have in your home. Place a number 2 next to your next choice and so on, up to number 6.

IMPORTANT: Place your six numbers in the space provided to the left of each service. Please review both pages before selecting your six prefered services.

D102 5. Now think about the six potential Channel 2000 services that you have just selected in ques­ tion #4.

Would you be willing to pay $15.00 per month for all of these services combined?

Again we are only interested in your opinions.

Please circle the one number you feel best describes your feelings at this time.

(L 10 20 30 40 50 60 70 80 90 100 14 6 5 12 6 13 9 17 7 6 6 I absolutely would not I absolutely would pay $15.00 per month pay $15.00 per month for these six services. for these six services.

* Numbers represent percentages (N=141) ** No response/missing data

« SECTION C: USING CHANNEL 2000 255 On the following pages are statements about Channel 2000 which allow you to tell us how you feel about using Channel 2000.

For each statement, please circle the number that best describes your feelings about the statement. Again, please be sure to circle one number for each statement.

$ J * * ./ /V * 4 V a r . # / V / ■ / ✓ ✓ D 103 Channel 2000 was easy for me to install...... 1 4 4 14 36 37

D 104 Channel 2000 is more for children than for adults ...... 15 28 33 13 6 2

D 105 There should have been more instruction on how to use Channel 2000...... 13 26 26 19 8 5

D 106 Channel 2000 tied up my television too o fte n ...... 26 33 18 9 7 4

D 1 07 1 frequently used the Channel 2000 Users Manual ...... 14 24 26 22 6 4

D 1 08 The Channel 2000 keypad was very easy to u s e ...... 2 4 6 21 43 23

D 109 The keys on the Channel 2000 keypad should have been larger .. 14 26 33 14 10 1

DUO 1 had no difficulty reading the instructions on the back of the Channel 2000 keypad...... 2 3 5 27 35 26

D i l l 1 wish Channel 2000 didn't have to be connected to my telephone 4 7 9 21 20 38

D 112 The Channel 2000 unit cluttered up my house ...... 21 21 20 21 10 5

D 113 There were too many technical problems with Channel 2000 ----- 8 10 18 28 15 19

D 114 There is a definite future for Channel 2000 type devices ...... 0 2 4 18 30 45

D 115 It was easier to get help from the Users Manual than from the help screens ...... 4 6 25 31 23 9

D 116 It was very annoying not to have all the Channel 2000 services working at the beginning...... 5 4 10 14 26 39

D 117 1 found reading from the TV screen to be very tiring ...... 11 27 22 24 11 3

D11S 1 never did figure out how to use all the Channel 2000 commands ...... 23 27 16 14 9 9 * D 119 It seemed as though 1 was involuntarily disconnected from Channel 2000 about 25 percent of the time that 1 used i t ...... 9 12 18 17 19 22

D 120 It took too much time to find an encyclopedia article that 1 wanted ...... 4 6 16 36 18 14

D 1 21 It took too much time to find a book .that 1 w anted...... 6 6 18 31 15 16

D 122 1 understood the Channel 2000 system well enough to find what 1 wanted to fin d ...... 4 4 9 26 36 20

D 123 The words were printed on the Channel 2000 system too slowly . 11 17 22 21 19 8

D 124 It took too much time to do my banking on Channel 2000 (11 1 applicable) ...... 9 9 7 4 3 4 * Numbers represent percentages (N=14l) ** No response/missing data 256 SECTION D: RATE THE CHANNEL 2000 SERVICES

Listed below are statements about several of the services available on Channel 2000. Please tell us how easy or difficult it is for you to use the service, and how useful or useless the ser­ vice is by circling two numbers to the right of each item.

Very Very Very Channel 2000 Service Very Easy D ifficult Useful Useless to Use to Use to M e Var. « 4 to M e V a r . # •A*.'»» D 125 Video Catalog ...... , 2 2 30 18 18 4 2 16 31 22 16 6 3 D 126 6

D127 Video Encyclopedia ...... 43 36 13 4 1 0 52 31 8 3 1 0 D128 5

D 129 Public Information ...... 37 26 13 9 2 1 47 26 9 2 1 0 D 130 14

D 131 Columbus C alen d er...... 17 15 24 18 12 6 12 18 25 22 11 5 D 132 8

D 133 Math That C ounts...... 8 19 29 15 14 8 9 19 26 14 16 9 D134 8

D 135 Early Reader...... 6 6 9 12 17 38 5 5 6 8 14 52 D 136 10

This next section is to be completed only by those people who signed up for the Channel 2000 banking services.

If you did not sign up for these services, please go to Section E.

Channel 2000 Banking Service Very Very Very Very Easy D ifficult Useful Useless DX37 Checking the balance in any of ,0j[Use to Use to M e to M e D 1 3 8 your accounts ...... 18 4 1 1 0 15 5 1 1 0 1 76

D 139 Bill payment service ...... 16 4 3 0 0 14 5 2 0 0 1 D 140 76

D 141 Examining your monthly banking transactions ...... 11 6 4 3 1 1 13 4 1 2 1 4 D142 76

D 143 Checking current interest rates. . . 9 6 3 3 2 1 3 2 6 4 4 5 D 144 75

* Numbers represent percentages (N=141) ** No response/missing data

i 257 SECTION E: YOUR BEHAVIOR

In this section we are interested in whether Channel 2000 has changed the way you behave. Remember to answer the questions according to how you currently (eel. Please circle the number that best describes your feelings about that statement. / *V SINCE I HAVE HAD CHANNEL 2000 IN MY HOUSE S

D 1 4 6 2. I watch television ...... 1 0 83 15 0 1

D 1 47 3. I read magazines...... 0 5 91 2 0 2

D 1 4 8 4. I read the newspaper...... 0 5 89 4 1 1

D 149 5. I read books...... 3 13 79 3 0 3

D 1 50 6. I talk to my fam ily...... 1 7 89 1 0 2

D 1 5 1 7. I do banking-related activities ...... 2 8 81 3 0 6

D 1 52 8. I am aware of community events ...... 7 43 48 1 0 1

D 1 53 9. I talk to my children (if applicable) ...... 0 7 55 0 0 38

D 154 10. My knowledge of library services i s ...... 5 35 57 1 0 2

SECTION F: YOUR FINAL THOUGHTS

Please carefully print or write in your response.

1. What recommendations do you have for us that would make Channel 2000 a better service?

2. In your own words, what did you like or dislike about Channel 2000?

* Numbers represent percentages (N=14l) No response/missing data 3. As a final question, we are extremely interested in your overall evaluation of the Channel 2000 system.

Please circle the number next to the statement that best represents your feelings.

1. A good idea and a good system.

2. A good idea but not such a good system.

3. Not such a good idea but no complaint about the system.

4. Not such a good idea and a poor system besides.

No response

* Numbers represent percentages (N=141)

Thank you again for your valuable assistance. Place your questionnaire along with your family’s questionnaires in the stamped envelope, and drop it in the mail.

PLEASE RETURN THE QUESTIONNAIRE PACKAGE BY JANUARY 14.

Upon receipt of this package, OCLC will be sending to your family a "thank you” gift. APPENDIX D

TELEPHONE QUESTIONNAIRE 260

TELEPHONE QUESTIONNAIRE

Phone number called:

Hello, I'm calling for Decision Research Corporation, a local public opinion research firm. We are conducting a very brief survey today in connection with library and banking services. I would like to speak with the (MALE/FEMALE) head of household. Are you that person? (IF NOT, ASK TO SPEAK WITH THE HEAD OF HOUSEHOLD AND REPEAT INTRO­ DUCTION). I would like to ask you just five or six questions.

1. first, have you used any library in the past year? 1. yes (2 ) 2 . no (3) 3. not sure (term)

2. How often would you say you have used it; from 4. 1-11(4) one to eleven times in the past year, or more 5 . 12+ (4) often than that? 6. not sure (as S i g n to 1-11 quota)(4 refsd.(term)

3. Has anyone else in your household used a library 1. 1-11 (4) in the past year? (IF YES, HOW OFTEN?) 2. 12+ (4) 3. no (4)(quota 40) 4. not sure (term)

4. Our research involves a test of an electronic system that can be used in the home to give you library services such as ordering books or looking things up. It also provides banking services like enabling you to pay bills by a special connection with the bank.

We are interviewing people that might be interested in participating in this test. Let me say quite clearly that this is not a sales approach. We are only trying to find families that might be in­ terested in participating in a test of an electronic information system for the home deleloped by a local company called OCLC that works with libraries around the country.

The test involves using the service for three months and filling out three questionnaires for research purposes. OCLC would loan you a small device that attaches to your television set (IF NO TV, TERMINATE HERE). You would receive this service at no cost.

Are you interested in being considered for this test? 1. yes (5) 2, no (term) 3. not sure (probe as on instr sheet)

5. Would you say that you are very interested in partici-1. very (^ pating, or just somewhat interested. * 2. somewhat(i) 3. not sure^O 261 (HERE ASK THE SUPPLEMENTARY QUESTIONS IF ASSIGNED TO YOU)

6. Then I need to ask a few additional questions. 1. yes (7,11 only) First, do you have children living at home with you? 2. no (8,9,10,11) 3. refsd (term)

7a.How old are they?

child #1: #2____ #3 #4 #5____

b.Are any of them in technical school or college? 1. yes 2. no

8.Approximately how old are you? . 1. under 25 2. 25-34 3. 35-44 4. 45-54 5. 55-64 6. 65+ 7. refsd

9.How many people including you live in your household? 1 2 3 4 5 6+

10. And the final question. Are you employed outside 1. yes,female of the home? (ALSO CODE SEX HERE). 2. no, female 3. yes',male 4. no,male

11.Thank you very much for your cooperation. We will be examining responses people give us in this survey and choosing several familes for the next stage of this test, in which we mail out a questionnaire.

For that purpose I would like to verify your name and address:

you are-.______

and your address is______

and I called you at __

Thank you. You will be hearing from us again soon, by mail. APPENDIX E

SCREENING QUESTIONNAIRE SCREENING QUESTIONNAIRE

August 4, 1980

Dear Respondent:

Thank you for your cooperation in our recent telephone interview. We are continuing the process of selecting the homes in which Channel 2000 will be installed.

You can help us by completing the enclosed questionnaire and returning it to us in the enclosed, stamped envelope.

Please complete and return the questionnaire by Monday, August 11.

We would like the person named on the envelope in which this questionnaire was sent to you to be the person who fills it out.

Please answer all questions as frankly and honestly as you can. We will be selecting people of all types — people with large and those with small families; people with and those without children; people who like to be the first to try new products, and people who do not; people who like to discuss electronic products with their friends, and people who do not.

We have enclosed a picture of Channel 2000 in operation. Basically it is a small electronic system that links your standard telephone with your standard television set. The connection is temporary. It does not interfere with your use of your telephone or television while Channel 2000 is shut off.

There is no charge for the Channel 2000 services. We will be placing the systems in homes for research purposes only.' We ask only that if selected to participate as a participating home or as a back-up home, you and other members of your house­ hold will cooperate by completing four questionnaires during the course of three months. All information in these question­ naires will be completely confidential.

Channel 2000 is a project sponsored jointly by OCLC, Bank One, and the Public Libraries of Columbus and Franklin County.

Again, thank you for your cooperation.

Decision Research Corporation

Decision Research Corporation Offices m CctLmbus and Cleveland 21 East State S treet * Columbus • Ohio • 4 3 2 1 5 • Phene- 6 1 4 -2 2 4 -0 6 2 3 264

!|oWNNaWUOINIJEX*| j©

A. Your telephone

B. Channel 2000 device which clips to the antenna of your tv C. Your own TV D. Channel 2000 push button control.

4 265

Please answer each question by circling the appropriate response. You may use either pen or pencil. Give the answer which best expresses your point of view.

1. Does any member of your household presently own 1. yes an electronic pocket calculator? 2. no

2. If so, in about what year did anyone in your 19 household first obtain an electronic pocket — — calcualtor? 3. Does anyone in your household regularly use a 1. yes bank machine in which you put a plastic card to get money, pay bills, or make deposits? 2. no (That is; once a month or more often?)

4. Would you say that you are generally one of the 1. one of the. first people in your area to purchase a new first electronic product or that you are not? 2. not one of the first

The following questions all relate to how much you yourself talk about electronic equipment with your friends. When you are answering a question that uses the words "electronic equipment , think of this as meaning such things as new types of stereo systems, televisions, calculators, TV games, or even home computers. 266 5. In general, do you like to talk 1. yes about electronic products*with your friends? 2. no

6. Would you say that you give very 1. You give very little little information, an average information. amount of information, or a great 2. You give an average deal of information about electronic amount of information products*to your friends? 3. You give a great deal of information.

7. During the'past six months, have you 1. yes told anyone about some electronic products?* 2. no

Compared to your circle of friends, 1. less likely to be asked are you less likely, about as likely, or more likely to be asked 2. about as.likely to be asked for advice about electronic products?* 3. more likely to be asked

9. If you and your friends were to 1. You mainly listen to discuss electronic products* your friends1 ideas. what part would you be most likely to play? Would you 2, You try to convince them mainly listen to your friends' of your ideas. ideas, or would you try to convince them of your ideas:

10.Which of these happens more often? 1. You tell them about Do you tell your friends about electronic products some electronic products* or do they tell you about some 2. They tell you about electronic products. electronic products

11.Do you get the feeling that you are generally regarded by your friends and neighbors as a good sources of advice about electronic products?*

*Electronic products refer to such things as stereo equipment, calculators, televisions, TV games, or even home computers. 267

12. The Channel 2000 project will be conducted 1. yes from about September 15 through December 15. Will you and other members of your household 2. no be available at home during that whole period to use Channel 2000 services and to 3. someone in complete questionnaires? household gone for part of tha time.

13. If one or more members of your household will be 1 2 3 4 5 unavailable during that time, please circle how many people in you household will NOT be available.

14. For how long will that person (those persons) not 1 week be available during this time period? (approximate) 2 weeks

3 weeks

4 or more week

15. How many people live in your household? 12 3 4, 5

6 7 8or more

\ 16. Do you have cable tv in your home? 1. yes 2. no

17. Do you have QUBE in your home? „ 1. yes

2. no

18. Does anyone in your .household have a checking !• yes account at Bank One? 2. no

19. Does anyone in your household have a library card? 1. yes

2. no 268

20. In the past year have you or any member of your 1. yes household used any library? 2. no

21. If anyone in the houshold has used a library, how l.from 1-11 often would you say that he or she has used it times during the past year? 2.12 or more times.

22. We now need to ask you some personal questions. All of this information will be kept confidential.

Using the following chart, please give the education level of each member of your household.

you your other oldest next next spouse adult child oldest oldest at home child at child home home pre-school 1 1 1 1 1 1 grade school 2 2 2 2 2 2 some high school 3 3 3 3 3 3 high school grad 4 4 4 4 4 4 some college 5 5 5 5 5 5 two year college 6 6 6 6 6 6 grad four year college 7 7 7 7 7 7 grad post-graduate 8 8 8 8 8 8

23. Is there anyone not covered by the above chart? If so, please describe relationship and education level: 269

24. How many children in each of the following 1.no children age categories are living with you at home? (Please write the number on the space______under 5 yrs provided. If you have no children in a particular category, leave it blank). __5-9 yrs

10-llyrs

12-17yrs

18-22yrs

23 yrs or older

25. Into which income category does the total 1. under $10,000 income of your household fall: 2. $10,000-14,999 3. $15,000-19,999 4. $20,000-24,999 5. $25,000-29,999 6. $30,000-39,999 7. $40,000-49,999 8. $50,000-99,000 9.$100,000 or more

26. Is anyone in your household employed by l. OCLC Bank One or by OCLC or by the Public Libraries of Columbus and Franklin County ? 2. Bank One If so, please note which one(s). ______. 3. Public Library

27. In order to complete the research for this project, we shall need to send you materials. We need to know the names and ages of each member of your household, Please list only those 12 years old or older.

yourself:age:( ) name:______i spouse, if married: age: ( )name:______;______

Other adult(s): age: ( ) name:______

age: (___ ) name:______

Children living at home: oldest: age: (___)___ name:______

Next oldest: age: (___) name:______

Next oldest: age: (___) name:______

Next oldest: age: (_____ ) name:

next oldest: ,age: ( ) name: 270

28. finally, there are many types of telephones today. For the Channel 2000 project it is important that we know what the receiver looks like on the phone nearest your television set.

Below are several pictures of the receivers of telephones currently available. Please look at your phone nearest your television, then mark an X in the box that has a picture of your phone's receiver.

Which of the following does the receiver on the telephone nearest your TV n n V 1 i!f

The receiver on my phone nearest my tv is not like any of these.

(Please mark your type of phone with an X in the appropriate box). i Thank you very much for your cooperation.

Please take a moment right now to put your completed questionnaire in the stamped, pr addressed envelope provided, and to mail i to us.

Thank you. APPENDIX F

QUESTIONNAIRE COVER LETTERS Dear Channel 2000 Participant Thank you for your interest in Channel 2000. You have been selected as a participant family in the Channel 2000 test being sponsored by OCLC, BANK ONE, and the Public Libraries of Columbus and Franklin County. You should be pleased to know that you will be among only 200 households in the entire United States who will have the opportunity to participate in this exciting project. Your comments and opinions will have a significant impact on future electronic information and banking services. We want everyone to understand that Channel 2000 is a free service that will be tested for research purposes only. No one will attempt to sell you anything either before, during, or after the test. We ask only for your cooperation in this research effort. As a Channel 2000 participant, it is necessary for you to do only two things:

1. Distribute the enclosed questionnaires to your family according to the name labels attached in the upper right corner. These questionnaires must be completed and returned in the enclosed stamped envelope by September 21.

2. Attend a demonstration session on the use of Channel 2000. At the session you will be provided, at absolutely no cost, the information and equipment necessary to access the information available through the Channel 2000 service. Details about this session will be mailed to you within the next two weekB.

The test is scheduled to begin on October 1, 1980. We are looking forward to your participation in, and your opinions about, the Channel 2000 service. Thank you for your cooperation.

Decision Research Corporation

Decision Research Corporation Offices in Columbus and Cleveland 21 East State Street • Columbus • Ohio • 43215 • Phone. 614-224-0623 DRC 274

Dear Channel 2000 Participant: Now that you have had Channel 2000 in your home for several weeks, we are interested in your initial reactions. We appreciate your cooperation with our first questionnaire, and we have now enclosed a second questionnaire "Now What Do You Think?" especially designed so that you can tell us how you now feel about Channel 2000. Here's all you have to do:

• Distribute the enclosed questionnaires to the person named in the upper right-hand corner of each questionnaire.

• Carefully complete all the items on the questionnaire

• Collect the completed questionnaires and return them in the enclosed, stamped envelope that we have sent you by December 1.

About the questionnaire itself. We think you’ll find this second questionnaire interesting to complete since you have had a chance to use Channel 2000. Some of the questions and statements may seem somewhat unusual, but they are designed to help us better understand what you like and dislike about Channel 2000. Be sure to complete all of the items and don't spend too much time on any one question. We again thank you for your cooperation in this exciting new project.

Decision Research Corporation Offices in Columbus and Cleveland 21 East State Street • Columbus • Ohio • 432 1 5 • Phone' 614-224-0623 DRC 275

Dear Channel 2000 Participant:

The Channel 2000 test will be coming to its conclusion January 1. 1981. Each individual who has participated in this three month test has been an extremely valuable part of this research effort.

To conclude our experiment, we are interested in your final evaluation of Channel 2000. This should be considered the most important part of the entire test.

We have enclosed a $2.00 cash gift to signify the importance of this last questionnaire, and to also say "thank you" for cooperating with the final evaluation.

(Jpon receipt of your questionnaires), OCLC will be forwarding to your family an additional gift as a remembrance of your participation in the first (J.S. consumer test of Channel 2000 technology.

Here is all that you have to do:

• Distribute the enclosed questionnaire to the person named in the upper right-hand corner of each questionnaire.

• Carefully respond to each question.

• Collect the completed questionnaires and return them in the enclosed, stamped envelopes that we have sent you by January 14.

Your patience, understanding, and cooperation is greatly appreciated. We hope that Channel 2000 enriched your life in some meaningful way. Remember when we receive your completed questionnaire we will forward to your family your Channel 2000 participation gift.

Decision Research Corporation

Decision Research Corporation Offices in Columbus and Cleveland 21 East State Street • Columbus • Ohio • 43215 • Phone- 614-224-0623

i APPENDIX G

LIFE STYLE INTERCORRELATION MATRIX LIFE STYLE INTERCORRELATION MATRIX

PI B2 ■ B3 94 R5. H6 B7 99- 99 910 B1 ------trtwotro— -ortn-T6T— ■“ •0.-3299-’ — =0709279 ■■ ------07707-59 ------n :n i3 ? 5 — —0. 07767" 0 .0 4 9 3 ’/ ' 0 .0 6 4 1 7 ■ ' 11.0*176 6— 1 . 0 0 coo 0.1 6 3 5 0 • 0 .0 4 9 1 3 0 .03 32 0 0. 16399 • 0.04303 —0.2493 2 0 .0 7976 - 0 . ?78->5 B2 0 . OB 767 n. io 7fi6 B3 0 .3 2 0 0 9 0.16350 . 1.00000 O'. 05639 -0 .0 0 22 0 0.0824R . -0 .1 8 0 0 6 0.0 5 5 1 ? -0 .0 3 7 1 5 BA -0 .0 9 27? 0.04913 0 .0 5 6 39 1.00000 0 .0 8 1 3 3 -0.04R 9R - 0 . 12781 0 . 15861 -0 - 0 8576 0 .1 7 146 B5 ------0t T0~559----- 0.-03-320— . -07002-23 - - 0 .0 3 1 3 3 1 0 0 00 0 0. 15703 • 0 .^ 4 1 7 6 0 .1 1 u53 —*J. 04^6? M. I 1 Oh I B6 0 .0 1 3 2 5 0.16399 0.00248 -0 .0 9 8 9 8 • 0 .1 5 7 0 8 ’ 1.00000 0.23637 -0 .2 2 2 0 0 0 .0 6663 ‘ -0 .1 6 7 3 ? B7 -0 .0 7 7 6 7 0.04300 - 0 . 18 006 - 0 . 127(11 0.2417 6 0.23637 . 1.00000 -0 .0 2 A 0 4 / 0 .0 5517 -0 .3 0 0 5 6 BIJ 0 .0 6 9 8 3 -0 .2 4 932 0. 05513 0.15861 0 .1 1 0 5 3 -0 .2 2 2 0 0 . -0 .0 2 4 0 4 . 1.0 0 0 0 0 -0 .0 5 3 7 8 0.3007 0 ■B9------0 .0 6 AT?------0-T03OT4— -0 .0 3 7 1 5 -0 .0 8 576 —0. 264 (1.06668 u . 0 5 4 i I —u.uSii / 1 .('PI'IKI —9.C 1 ..’9 5 B10 0 . 09 36 6 -0 .2 7 9 2 5 0 .103 85 0. 12146 0 .1 1 0 6 1 ' -0 .1 6 7 3 9 -0 .3 0 0 5 6 0 .3 0 0 7 0 -0 .2 1 2 9 5 1 .00000 B ll -0 .1 3 4 3 5 0.'30 310 0 .043 39 0 .1 3 7 2 2 -0 .1 0 2 6 1 0.02301 -0 .1 5 4 0 9 -0 .1 9 9 2 3 -0 .0 3 0 2 1 -0 .0 5 7 6 1 B12 -0 .0 5 0 9 4 -0 .0 1 0 4 9 ' 0 .0 6 1 3 3 0 .0 0 8 8 3 -0 .4 6 39 2 -0 .1 3 6 7 2 -0 .2 7 5 3 3 0 .0 6 5 7 7 0 .0 4 4 4 3 0.04393 -813 ------0 . 12211----- “ 0.09740— 0 .0 4126 0.1 0 6 4 6 -0 .0 5 1 1 4 , -0 .1 1 8 9 9 —0.1 6 3 1 0 U | —u. 13(15 B14 0 .0 2 5 9 0 0 .0 3349 0 .0 3 2 6 0 .-0.06375-• -0 .1 4 1 0 8 0.0 7 2 0 5 . 0.0 3 6 3 6 - 0 . 14493: 0.0 3837 —0.14454 B15 0 . 139*7 5 - -0 .0 1 3 7 1 0. 12 1 26 , -0 .0 3 9 4 3 - 0 . 11327 ’ -0 .0 3 0 3 1 -0 .0 8 3 1 0 -0 .0 5 8 1 5 0.04904 -0 .0 6 1 9 1 B1 6 0.23(346 -0 .0 6 7 7 3 0. 02416 -0 .0 1 9 5 8 0.12151 '-0.04000 . - 0 .0 3 9 ° 2 • 0 .2 3 6 6 4 ' 0 .0 3519 , 0 .1 3 0 1 7 BIT -Or26?7-9 ----- — crroi^no —0. 36 169 0 .0 5 3 B 2 "' 0 . Oti V4 5 0. I*t4‘-i4 0. 36 704 n. 1 6 (49 — l1. 1 5i)<* BIB -0 .1 3 7 3 9 0.407d4 0.03142 • 0 .0 0 6 0 4 -0 .0 3 069 P. 20053 0 .0 3 67? -0 .3 3 1 6 7 0.1311? -0 .3 2 6 4 7 B19 0 .0 4 1 1 4 -0 .0 7 3 9 1 -0 .0 0 6 2 ? 0.32373 - 0 . 01779 -0 .2 5 3 1 5 - 0 . 07071 0 .1 4 9 7 9 0.03164 0.05551 B20 -0 .2 3 6 7 5 -0 .1 1 2 2 7 -0 .1 3 0 2 0 0 .1 4623 -0 .0 8 34] 0.05943 0.-OA38 3 O. 10783 0.01013 0 .13574 B21------Ot-1394-O— -*frroir286— ------0.-22T53-----— “0774708 ' “0. 12 103 —0.1 0253 - 0 . 1781 4“ 0 .2 8 8 5 8 —I). 1 1 1 I f> 0. 3 1 12 5 ‘ B22 0. 12787 0.03 430 ■ ' -0 .0 0 8 0 1 0.02198 0.17 309 0.03849 0.16951 : 0 .1 6241 -0 .1 3 0 7 3 0. 13236 B23 0 .0 1 7 3 9 -0 .1 5 1 0 7 0 .2 5 6 0 7 0 .2 4 4 4 3 , 0.02651 -0 .1 5 1 6 4 0.04161 0.2916? 0.06741 0. 05747 B24 -0 .2 5 4 9 2 -0 .0 7 4 1 9 . -0 .2 5 5 7 0 0 .1 7 9 8 2 -0 .1 0 670 0 .0 3 2 3 0 . .0 .2 3 0 9 0 0.00294 0 .0 5 2 3 9 -0 .0 5 4 1 8 B25 ------0TO653"2------0 .1 7 0 7 4 ------07-03372 ----- "■ 0 /2 6 4 3 7 0 . UO ^06 -0 .1 7 1 9 6 - 0 . 1599 5 0 .3 3 7 7 c - > > • ' 1 B26 -0 .2 2 8 3 6 -0 .0 9 0 7 1 -0 .1 2 6 3 4 0 .0 2 2 1 4 0 .0 6 6 9 8 0 .1 0 7 6 4 0. 19925 -0 .0 2 3 3 0 0 .0 6 3 2 4 -0 .1 9 1 4 8

B ll B12 / . /.B13 B14 Bt 5 R16 ' ■ B17 B18 B19 820 O:’02 598 U. ' -0 . 26879 - 0 . 1M 1 -4 9 h .04114 B1 - - 0 ; 13435 -O.OS.T^A" 0. 12211 -0 .0 7 3 9 1 R 2 0 .3 0 3 1 0 -0 .0 1 C 4 9 -0 .0 9 7 4 0 0 .0 3 3 4 9 -0 .0 1 3 7 1 -0 .0 6 7 7 3 0 .0 1636 0 .4 0 7 8 4 -0 .1 1 7 7 7 B3 0.0433? 0.06133 0 .0 4 1 2 6 0.0 3 2 6 0 0. 12 12 6 0.0 2 4 1 6 -0 .3 5 1 6 9 0 .0 3 142 -0 .0 0 6 2 9 -0 .1 3 3 2 8 BA 0 .13 72 2 0.00803 0 .1 0 6 4 6 -0 .0 6 3 7 5 -0 .0 3 9 4 3 -0 .0 1 053 0.05332 0.00604 0.3 237 3 0. 1442(1 85 ------= 0 n tr2 frl------0 .4 6 3 9 2 ---- =0705114 ■ -0 ; 1410,5 -0. 11 327" 0772151 0 .0 8 9 4 5 —0.0 -U. 1.1 1 f /•+ B6 0.02 30 1 -0 .1 3 6 7 2 - 0 .1 ie 9 9 0 .0 7 2 0 5 -0 .0 3 0 3 1 -0 .0 4 0 0 8 0 . 14999 0.2005.8 • -0 :2 5 8 1 5 0 .0 5968 .-0 .1 3 3 1 0 0 .0 3 6 3 6 -0 .0 3 3 1 0 -0 .0 3 9 9 2 0.36704 0 .0 3 6 7 3 -0 .0 7 0 7 1 0 .0 4 3 8 3 B7 • -0 .1 5 4 0 9 -0 .2 7 5 3 3 0 . 1 40?9 BB - 0 . 1 9 °2 3 0.05577 -0 .0 1 7 6 3 -0 .1 4 4 9 3 -0 .0 5 8 1 5 0.2 3 6 6 4 -0 .0 4 0 3 9 -0 .3 3 1 6 7 0.1038 3 ------=U703021 ------0715715 0 . 033B7 0. Ot u04 I). 0851** l'. 1(5 (44 n. i •* i l'J 0 . U 'i I 44 •89 — -- 0.-0445^ 0.05451 O. 13524 BIO -0 .0 5 7 6 1 0 .0 4 5j93 0.2 7 9 9 5 -0 .1 4 4 5 4 - 0 .0 4 I ' l l 0 .1 3 0 1 7 . -0 .2 1 4 0 2 -0 .3 2 6 4 7

--j SJ LIFE STYLE INTERCORRELATION MATRIX

' B ll B12 Bl? . Bl 4 Bl 5 ' R16 Rt 7 RIB RIO R20 B ll 1.00000 0 .0 7 557 0 .1 4 1 0 2 -0 .0 2 0 4 6 0 .0 0 521 P .05953 -0 .0 4 6 0 6 0 .2 8 8 9 3 O .lo ? i7 - 0 . 13267 B12 1 0 .0 7 5 5 7 , 1.00000 0.00062 0.1 5 0 7 2 - 0 .0 2 0 4 2 -0 .0 3 1 4 3 0.08319 0 .0 0 2 4 3 0 .0 5 4 3 7 -0 .0 0 1 7 1 B 13 0 .1 4102 0 .0 0 062 . 1.00000 0. 13996 . 0 .2 0 4 0 2 0.07883 -0 .1 0 3 9 8 - 0 . 19?26 . 0.0 0 3 4 5 - P . 02892 B 14 ■ ------0.02046 " 0 .1 5 u7 j; 0 .13996 . 1 . ooono ■' 0.4029 7' " 0. 12 35 0 " 0. 1544U U. Ul'UAi, U.(Jfc'->4 5 -O'. 77505 R 15 0 .0 0 521 0 .0 2 0 4 2 0.20402 0 .4 0 2 9 7 ■ 1.00000 0.14691 ’ -0 .2 4 2 6 7 -0 .0 4 7 3 3 0.16R01 -0.24879 616 0 .0 5 9 5 3 -0 .0 3 1 4 3 0. 073 33 0.12350 0.1469 1 1.00000 -0 .0 7 6 7 4 O .O O ?^ 0 .2 9 7 5 6 -0 .0 7 R I1 B 17 -0 .0 4 606 0.08318 -0 .1 0 3 9 0 0 . 154(10 —0.2 4 2 6 7 -0 .0 7 6 7 4 1.00000 0. 10379 - 0 . 0 4447 0.0774 6 B1R ' ' 0 .2 8 8 9 ? " — 0;00243 - 0 . 1 9 2 2 6 ' 0 .0 0 0 5 4 ' ' - O ; 04733 0. 0(1 H i (J. 10379 I .OO0CTO- -0.044 f A —O'.TVfi 115 B19 0 .1 9 7 1 7 0.05437 . 0.00345 0.06945- 0 .16 501 0.29756 -0 .0 4 4 4 7 -0 .0 4 9 7 5 1.00000 -0 .0 1 6 7 4 B20 -0 .1 3 2 6 7 -0 .0 0 1 7 1 -0 .0 2 8 9 2 -0 .2 7 5 0 5 -0 .2 4 8 7 9 , , -0 .0 7 3 1 3 0.07245 . -0 .0 6 1 1 5 -0 .0 1 6 7 4 , 1. noooo B21 0 .1 7 9 6 6 0 .0 4 3 2 4 , 0 .2 3 °9 6 -0 .0 2 0 5 9 0.1616? . » 0.2 1 3 4 5 - 0 . 16734 -0 .0 6 7 0 2 0 .7 9347 0.04471 B22 —0.0 6 6 4 7 -0710533 u. 06 035 , —0.05 505 —U. 1 / 4 ' 0. 78 97 7 ' II. 1 2 // i (1. 001196 9 .1 9 J/ (i 0 .0 3 /49 B23 0 .0 3 2 3 3 0 .1 0 9 7 9 0 .1 2 1 6 7 -0 .0 3 0 4 2 0 . 1749 5 0.0 7 1 0 9 -0 .0 7 2 9 4 -0 .0 6 4 3 0 0.14561 - 0 .0 5 046 B24 0 .0 2 1 0 2 0.06911 -0 .0 5 4 57 -0 .0 9 7 9 8 -0 .0 5 889 -0 .1 5 6 1 8 0.18711 0 .0 6 6 3 0 ■ 0 .0 1 7 6 5 0 .23101 B25 -0 .0 0 0 1 2 0 .01255 0. 21R89 -0 .0 7 2 0 7 0. 13 02 9 0.38 182 -0 .0 3 3 5 6 -0 .0 9 4 4 7 0 .2 5 7 1 7 -0 .0 7 8 ° 6 ‘ 826 —0*1580 6 '-O .10271- ' —0. 24681 , -0 .0 1 5 1 5 —0. 25'865 '-0.'0T596 01. 33027 U.U5414 -u .u y 9 /4 0.10173” " : i ' 1 . i i . 1 , 1 1 1 1 ' B21 B22 R23 ' B24 R25 R26 Bl 0.13040 0.12787 0 .0 1 7 3 9 - -0 .2 5 4 9 2 0 .0 6 552 -0 .2 2 8 3 6 B2 ...... -0 .0 1 2 9 6 0 .0 3 480 - 0 . L5107 -----=0-."07419 —0. 17 67 4 -0 .0 9 0 7 1 " ,B3 > - 0 .2 2 7 5 3 -0 .0 0 3 0 1 , 0 .2 5 6 0 7 • -0 .2 5 5 7 0 ' 0 . 03 372 - 0 . 12634 :b 4 0 .3 4 7 0 8 - 0 .0 2 1 9 8 , ,0 . 2 4 4 4 3 0.1798? 0 .2 6 4 3 7 0 .0 2 2 1 4 ' i,B5 0 .1 2 1 0 5 . 0 .1 7 3 0 9 . , 0 .0 2651 -0 .1 0 6 7 0 0 .0 0 906 0 .0 6 6 9 8 BA * ...... —0. 10233' ' 0.'03 349 ■" -0 .1 5 1 6 4 ' " o:03230 ' -0 .1 7 1 9 6 0. 10 / A 4 87 -0 .1 7 3 1 4 0 .16951 0.04161 0 .2 3 0 9 0 -0 .1 5 9 9 5 0 . 19926 - B8 0.2 8 8 5 0 0.16241 0 .2 9 1 6 9 0 .0 0 2 9 4 0 .3 3 7 7 2 -0 .0 2 3 3 0 B9 -0 .1 1 3 7 6 - 0 . 13 073 0 .067 41 0.0 5 2 3 9 -0 .0 6 5 5 7 0 .0 6 3 2 4 B 10 0 .3 1 1 2 5 ' 0 .1 3 2 0 6 0 . 05747 .—0. 0541B ..... Q."33"5R9 ' - P . I y 148 B ll 0 . 17966 -0 .0 6 6 4 7 0 .0 8 2 0 3 0 .0 2 1 0 2 -0 .0 0 0 1 2 -0 .1 5 3 0 6 ■ B12 0 .0 4 3 2 4 -0 .1 0 5 5 3 0. 10979' • ‘ 0. 06911 ‘ -0 .0 1 2 5 5 , -0 .1 0 2 7 1 . - BIT 0 .2 3 9 9 6 0.06035 0 .1 2 1 6 7 . -0 .0 5 4 5 7 0 .2 1 8 8 9 -0 .2 4 6 B 1 B14‘ ■ ' -0 .0 2 0 5 9 -0 .0 5 5'.j5 , — 1 lm ' -0.0970 8 .- 0 .0 / 2 0 i -0 .0 1 51 5 '■ B15 0 .1 6 1 6 3 -0 .1 7 6 5 4 0.1 7 4 9 5 - - 0 . 050.39 0. 13029 -0 .2 6 8 6 5 B 16 0 .2 1 3 4 5 0.23977 0 .0 7 1 0 9 - 0 . 15613 ' 0 .3 8 1 3 ? -0 .0 1 5 9 6 B17 -0 .1 6 7 3 4 0.12773 -0 .0 7 2 9 4 0.18711 -0 .0 3 3 5 6 0. 33027 BIB' " —l*. 0670 2 O'; 00 09 6" ■" - 0 . 064 80 u.06ft8G '-0r0944 / 1' U. l)rt4 14 B19 0 .2 9 8 4 7 . 0 . 1037B 0. 14561 , 0 .0 1 7 6 5 0 .2 5 7 1 7 -0 .0 9 9 7 4 • B20 0 .0 4 4 7 1 ' 0.03749 - 0 . 05046 0 .2 3 1 0 1 - 0 .0 2 896 0 .1 0 1 7 3 . 1 .00000 -O .'05 096 0.27941 -0 .1 1 4 7 0 0 .1 9 024 -0 .2 6 2 1 3 iB22 l l . —0.0 5 3 9 6 1 . ooonri" ■ —0. 046 51 ■ ' -L 'l. 02 519 " 0 .3 4 0 1 9 0. 1794T B23 0^2794.1 -0 .0 4 6 5 1 _ 1.00000 ,. „■ -Q.Q4957 0.14.941 0 . n l 943 - B24 -0 .1 1 4 7 0 - 0 .0 2 519 -0 .0 4 9 57 1.00000 0 .1 1 1 6 1 0 .0 0 1 0 7 B25 0 .1 9 0 2 4 0.34019- ' 0.-14941 0 . 11161 1.00 000 0.11408 B26 -0 .2 6 2 1 3 0.17941 0 .0 1 9 4 3 0 .0 8 1 0 7 0 . 11403 1.00000 ' 1 DETERMINANT OP CPP.REL AT-ION MATRIX = 0.00156811 0 . 156 600300-0.?) t

to 00 APPENDIX H

ACTIVITIES INTERCORRELATION MATRIX ACTIVITIES INTERCORRELATION MATRIX

B68 B69 B70 B71 B72 B73 874 375 B76 877 I B60 1.00000 0.35406 0.08150 0 . 12B92 0.19103 0 .0 4 0 1 3 0 .2 6 8 0 7 0 .0 2 9 7 5 0 .1 0 6 5 8 -0 .1 4 5 3 3 B69 0 .3 5 4 8 6 1 .0 0 0 0 0 .0 .1 9 4 5 5 -0 .0 3 3 9 4 0.13335 0.12074 0.38960 0.26509 0 .1 8 4 3 2 0 .0 4 9 4 6 B70 0.00150 0.19455 1.00000 0.23371 0.36768 0 .1 9 8 3 3 0.37480 0.14359 0.38894 0 .2 4 3 2 7 B71 0 .1 2 8 9 2 ^ 0 .0 3 3 9 4 0 .2 3 3 7 1 1.00000 0.1B419 0.04R 54 0.16601 0 .0 5 4 6 3 _ 0-?2Rin _ 0.1H44n B72 0.19103 0.13335 0 .3 6 7 6 8 0 . 18419 1 .00000 0 .1 2 7 6 4 0 .2 3 2 4 3 0 .0 0 3 7 8 0.19478 0.13159 B73 0 .0 4 0 1 3 0.12074 0.19033 0.04854 0.12764 l;00000 0 .0 9 3 0 6 0 .0 0 6 0 7 0.1B731 0 .2 1 6 7 2 B74 0 .2 6 8 0 7 0.38960 0.37480 0.16601 0.23243 0 .0 9 3 0 6 1 .0 0 0 0 0 0 .1 5 8 6 2 0.42111 0.14227 H 7 5 0.02975 0.26509 0.14359 0.05463 Cl.0037ft 0-00607 0-15R62 1 -nnnnn n-O^O?? B76 0.10658 0.18432 0.3B894 0.22810 0.19478 0.18731 0 .4 2 1 1 1 0.07257 1.00000 0 .2 4 4 5 4 B77 -0 .1 4 5 ? 3 0 .0 4 9 4 6 0 .2 4 3 2 7 0 .1 8 4 5 0 0 .1 3159 0 .2 1 6 7 2 0 .1 4 2 2 7 0.09022 . 0.24454 1 .0 0 0 0 0 B78 0 .1 0 1 6 2 0 .0 1 0 6 0 0 .1 6 3 2 2 0 .0 8 4 0 9 0.02272 -0.02912 0 .1 7 3 3 1 -0 .0 2 8 8 8 0.31385 -0.04458 B79 0 .4 7 0 8 4 '0.44837 0.30184 0. 13123 0.27001 -0 .0 0 5 1 ft (1-36676 n.nnfti5 n.?ft4?l -n .0 2 0 4 ? B80 0 .5 0 8 6 6 0 .3 0 9 1 1 0 .1 4 3 4 8 0.14314 0.11931 -0 .0 0 4 5 4 0 .3 0 0 4 2 0.05115 0.15820 0.01828 B81 0 .1 6 1 6 2 0.05388 0.07607 0.23933 0.10502 -0.06045 0 .1 1 5 9 5 0 .0 2 7 8 1 0.14682 -0.11163 B82 0 .2 1 8 2 3 0.1B 335 0.14102 0.08005 0.21764 0 .0 2 9 7 4 0 .1 9 4 0 3 -0 .0 5 5 8 4 0 .0 6 5 8 0 -0 .0 0 6 4 8

B78 B79 RPO RBI RR2 B6B 0.1B 162 0 .4 7 0 8 4 0.50866 0il6l62 ' 0.21823 , B69 0 .0 1 0 6 0 0 .441337 0 .3 0 9 1 1 0.05388 0.18335 B70 - 0.16322 0.301R4 • 0.14348 0.07607 O.14102 B71 0 .0 3 4 0 9 0.13123 0.14314 0 .2 3 9 3 3 0 .0 8 0 0 5 B72 0 .0 2 2 7 2 0.270B 1 0.11931 0.10502 0.21764 B73 . -0.02912. -0 .0 0 5 1 3 -0.00454 -0.06045 0.02974 B74 0.17331 0.36676 0.30042 0.11595 0.19403 B75 -0.023BB 0.00315 0.05115 0.027B1 -0 .0 5 5 8 4 B76 0 .3 1 3 8 5 0 .2 3 5 2 1 0.15820 0.14682 0 .0 6 5 8 0 0 7 7 -0.04458 -0,02847 0.01828 -0.11163 -0.00648 B7B l.n nn n n 0.26851 0 .2 8 7 0 2 0.3351 6 0 . 1044ft * B79 0 .2 6 8 5 1 I.00000 0.75201 0.12538 0.33268 BBO 0 .2 8 7 0 2 0.75201 1.00000 0. 13210 0.24355 I ' B81 0 .3 3 5 1 6 0.12538 0.13210 1.00000 0.16386 RR? 0.10498 0.3326R 0.24355 O.163R6 i.o o n n o

DETERM INANT OF CORRELATION MATRIX = 0.02932941 0.293293860 -0 1 )

M CO O