Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Adoption of information technology in a small business setting 1992

You don't seem to have a PDF reader installed, try download the pdf

Item Metadata

Download

Media
ubc_1992_spring_laktin_richard_s.pdf
ubc_1992_spring_laktin_richard_s.pdf [ 3.41MB ]
Metadata
JSON: 1.0086714.json
JSON-LD: 1.0086714+ld.json
RDF/XML (Pretty): 1.0086714.xml
RDF/JSON: 1.0086714+rdf.json
Turtle: 1.0086714+rdf-turtle.txt
N-Triples: 1.0086714+rdf-ntriples.txt
Citation
1.0086714.ris

Full Text

ADOPTION OF iNFORMATION TECHNOLOGY IN A SMALL BUSINESS SETTING by RICHARD S. LAKTTN B. Comm., University of British Columbia, 1982 CA, Canadian Institute of Chartered Accountants, 1991 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in THE FACULTY OF GRADUATE STUDIES Department of Commerce and Business Administration We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April 1992 © Richard S. Laktin, 1992 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Commerce and Business Administration The University of British Columbia 1956 Main Mall Vancouver, British Columbia Canada V6T 1Y3 Date: 30 April, 1992 11 AB STRACT Many small businesses are turning to Information Technology as a means of competitive advantage and survival in today’s tougher business climate. The Public Accounting profession portrays itself in the role of Information Consultant to small business when it comes to information technology. The role that Public Accountants play in the information technology adoption process is poorly understood. The purpose of this research was to examine more closely the role that information consultants play in the adoption process, with particular emphasis on the public accountant. The Dffusion of Information Technology model (Moore, 1989) was used as the theoretical foundation for this study. The Diffusion of Information Technology model is ll grounded in theory and is supported by Moore’s research results. The major research questions answered are: 1. What role do independent information consultants such as accounting firms play in the Dffiision ofInformation Technology process? 2. Is the Diffiis/on ofInformation Technology model a general model? A cross-sectional survey using a questionnaire was issued to small business clients of public accounting firms. Profiles of information technology users and non-users were generated from questionnaire data. These profiles were subject to regression analysis and structural equation modelling using PLS (Partial Least Squares). The analysis provided some answers to the role accountants play in the information technology adoption process as well as supporting the Diffusion of Information Technology model in a small business domain. 111 TABLE OF CONTENTS ABSTRACT ii TABLE OF CONTENTS iii LIST OF TABLES iv LIST OFFIGURES v ACKNOWLEDGEMENTS viii CHAPTER 1: INTRODUCTION AND OVERVIEW OF RESEARCH 1 1.1 RESEARCH STUDY RATIONALE 1 1.2 RESEARCH DIRECTION 3 1.3 THE COMPUTERIZED ACCOUNTING SYSTEM 4 1.4 TOWARDS A SMALL BUSINESS ORIENTATION 5 1.5 THE ROLE OF INFORMATION TECHNOLOGY IN ORGANIZATIONS 7 CHAPTER 2: LITERATURE REVIEW 10 CHAPTER 3: ADOPTION OF INFORMATION TECHNOLOGY 13 3.1 DIFFUSION OF INNOVATIONS 14 3.2 THE THEORY OF REASONED ACTION 15 3.3 DIFFUSION OF INFORMATION TECHNOLOGY 15 3.4 SIGNIFICANCE OF THE DIFFUSION OF INFORMATION TECHNOLOGY MODEL 15 CHAPTER 4: TECHNOLOGY TRANSFER - RESEARCH QUESTIONS 17 4.1 RESEARCH HYPOTHESES 18 CHAPTER 5: INSTRUMENT DEVELOPMENT 22 SECTION A - INTRODUCTION 22 5.1 GENERAL 22 5.1.1 RELIABILITY 22 5.1.2 VALIDITY 23 5.1.3 QUESTIONNAIRE SELECTION 24 SECTION B: QUESTIONNAIRE DESIGN - PILOT STUDY 24 5.2 CLIENT/DIFFUSION OF INFORMATION TECHNOLOGY QUESTIONNAIRE 24 5.2.1 PERCEIVED CHARACTERISTICS OF INNOVATIONS 24 5.2.2 SYSTEM USAGE 26 5.2.3 CLIENT COMPUTERIZED ACCOUNTING SYSTEM SUPPORT 30 SECTION C: FINAL SURVEYS - SCALE RELIABILITIES 30 5.4 GENERAL 30 5.5RESULTS 31 SECTION D: QUESTIONNAIRE DESIGN 31 5.6 GENERAL 31 5.7 FORMAT 32 5.7.1 PAMPHLET 32 5.7.2 QUESTION LAYOUT 33 5.7.3 COVERING LETTER 33 CHAPTER 6: DATA COLLECTION AND ANALYSIS 34 SECTION A: DATA COLLECTION AND CONDITIONING 34 6.1 INTRODUCTION 34 6.2 SURVEY SAMPLE 34 6.2.1 TARGET POPULATION SELECTION 34 6.2.2 PROBLEMS ENCOUNTERED 36 6.2.3 RESPONSE RATES 37 6.3 CLIENT FIRM’S SURVEY 38 6.3.1 RESULT DEMONSTRABILITY 39 6.4 CONDITIONING THE DATA 40 6.4.1 GENERAL 40 6.4.2 ACCURACY OF INPUT DATA 40 6.4.3 MISSING DATA 40 iv 6.4.4 OUTLIERS AN]) SKEWNESS .41 6.4.5 NON-LINEARITY AND HOMOSCEDASTICITY 41 SECTION B: DESCRIPTIVE STATISTICS 42 6.5 GENERAL 42 6.6 DEMOGRAPHICS 43 6.7 ATTITUDE TOWARDS INNOVATING 44 6.8 PERCEIVED CHARACTERISTICS OF INNOVATING 45 6.9 SUBJECTIVE NORMS 46 6.10 INNOVATIVENESS MEASURES 46 6.11 COMPUTERIZED ACCOUNTING SYSTEM SUPPORT 49 SECTION C: REGRESSION ANALYSIS 49 6.12 GENERAL 49 6.13 THE EFFECT OF PERCEIVED CHARACTERISTICS OF INNOVATIVNESS AND VOLUNTARINESS ON ATTITUDE 50 6.14 THE EFFECT OF ATTITUDE. SUBJECTIVE NORM. PERCEIVED CHARACTERISTICS OF INNOVATIVENESS. VOLUNTARINESS AN]) SUPPORT ON INNOVATIVENESS 53 6.14.1 GENERAL 53 6.14.2 ATTITUDE. SUBJECTIVE NORM AND VOLUNTARINESS ON INNOVATIVENESS 53 6.14.3 PERCEIVED CHARACTERISTICS OF INNOVATING. SUBJECTIVE NORM, AND VOLUNTARINESS ON INNOVATIVENESS 55 6.14.4 OTHER REGRESSIONS 57 SECTION D: PATH MODELING 59 6.15 CHOICE OF PATH MODEL COMPUTER IMPLEMENTATION - LISREL vsPLS 60 6.15.1 DESIGN OF PLS PATH MODEL 61 6.15.2 ANALYSIS OF SAMPLE SIZE REQUIREMENTS 62 6.15.3 GOODNESS OF FIT DETERMINATION 64 6.15.4 ASSESSMENT OF HYPOTHESES TESTING 66 6.16 SUMMARY OF RESULTS: PATH ANALYSIS 69 SECTION F: SUMMARY OF DATA ANALYSIS 70 6.17 GENERAL 70 6.18 SUMMARY OF DESCRIPTIVE STATISTICS 70 6.19 SUMMARY OF HYPOTHESES TESTING 70 CHAPTER 7: CONTRIBUTIONS. IMPLICATIONS AND LIMITATIONS 73 7.1 INTRODUCTION 73 7.2 SUMMARY OF THE RESEARCH PROCESS 73 7.3 THE RESEARCH QUESTIONS ANSWERED 74 7.3.1 QUESTION TWO 74 7.3.2 QUESTION ONE 75 7.4 CONTRIBUTIONS 76 7.5 LIMITATIONS OF THE STUDY 77 7.6 CONCLUSION 78 TABLES 80 FIGURES 100 BIBLIOGRAPHY 108 APPENDICES 117 VLIST OF TABLES TABLE 1 - RELIABILITY COEFFICIENT: PILOT TEST (SPSS) 81 TABLE 2 - RELIABILITY COEFFICIENT: ACTUAL STUDY (SPSS) 81 TABLE 3 - RELIABILITY COEFFICIENT: ACTUAL STUDY (SPSS) USERS VS NON-USERS 82 TABLE 4- RELIABILITY COEFFICIENT: MOORE 82 TABLE 5(a) - DEMOGRAPHIC BACKGROU1D OF SURVEY RESPONDENTS 83 TABLE 5(b) - DEMOGRAPHIC BACKGROU1D OF SURVEY RESPONDENTS 84 TABLE 6(a) - SURVEY VARIABLES - DESCRIPTIVE STATISTICS 85 TABLE 6(b) - SURVEY VARIABLES - DESCRIPTIVE STATISTICS 86 TABLE 7(a) - USERS VERSUS NON-USERS 87 TABLE 7(b) - USERS VERSUS NON-USERS 88 TABLE 8- REGRESSION RESULTS 89 TABLE 9- REGRESSION RESULTS 90 TABLE 10(a) - REGRESSION RESULTS 91 TABLE 10(b) - REGRESSION RESULTS 92 TABLE 11(a) - REGRESSION RESULTS 93 TABLE 11(b) - REGRESSION RESULTS 94 TABLE 12(a) - REGRESSION RESULTS 95 TABLE 12(b) - REGRESSION RESULTS 96 TABLE 13(a) - SUIVIMARY RESULTS OF HYPOTHESES TESTING 97 TABLE 13(b) - SUMMARY RESULTS OF HYPOTHESES TESTING 98 TABLE 14 - GENERAL PLS STATISTICS FOR TESTED MODELS 99 vi LIST OF FIGURES FIGURE 1 - DIFFUSION OF INFORMATION TECHNOLOGY MODEL 101 FIGURE 2- DIFFUSION OF INNOVATIONS MODEL 102 FIGURE 3- INNOVATION DECISION MODEL 103 FIGURE 4- STAGES OF TNE INNOVATION DECISION PROCESS MODEL 104 FIGURE 5- NON-CAS USERS: RESULT DEMONSTRABILITY 105 FIGURE 6- DIFFUSION OF INFORMATION TECHNOLOGY MODEL: PLS LOADINGS ON 106 ORIGINAL MODEL FIGURE 7- DIFFUSION OF INFORMATION TECHNOLOGY MODEL: PLS LOADINGS ON 107 EXTENDED MODEL vii ACKNOWLEDGEMENT S The completion of this dissertation required the help, support and frequent encouragement by several individuals, to whom I owe a great deal of gratitude. Special recognition is directed at my thesis supervisor, Professor Al Dexter, who was always there to direct, encourage, and refocus my efforts at producing this paper. I would like to also thank the other members of my committee, Chino Rao and Gary Moore, for their helpful criticism and comments. I would also like to thank my fellow graduate students at UBC and the faculty members in the Management Information Systems area for their support and encouragement. I would like to express my appreciation to Wynne Chin from the University of Calgary for his help with the PLS analysis. Without his timely intervention, a major part of the statistical analysis could not have been completed. I am very grateful to my sister, Tamara, who put up with the irritations, grumpiness, and frequent late night disturbances resulting from the printing out of various drafts of this paper. In addition she was instrumental in helping out with the data verification process. Special thanks go out to my cousin Ann who was very helpful with data verification and reviewing this paper for grammatical errors. Thank you Lindsay, for your encouragement and insightful suggestions on the final versions of this paper. Finally, I would like to thank my parents, Cyril and Doris, for their financial and material support provided at various times throughout the time spent completing the graduate program. Without their help this thesis would not have been completed. 1CHAPTER 1: JNTRODUCTION AND OVERVIEW OF RESEARCH 1.1 RESEARCH STuDY RATIONALE Most small firms have limited access to information sources on information technology (IT). As a result, information is often sought from an external information consultant (Goodson, 1990). The role of the external information consultant as an information source to small firms is an important research area. In the role of information consultants professional accountants have been involved with many computer systems that have been considered successful by their users, and several that have been considered failures. To the accountant as well as the small business client they serve, the success or failure of the introduction of an information technology may seem to be not only a product of planning but a product of fortune as well. To an accountant, working in a profession that sells information and methods of generating information as products, unsuccessful implementation of computerized accounting systems is to be avoided. Maintaining good client relations is the bottom line to professional accounting organizations and failures (perceived or otherwise) are unacceptable, as small businesses cannot afford the emotional and monetary costs of an unsuccessfully implemented computerized accounting system. Research that can illuminate the interaction between small firms and their public accountant may provide the accounting profession with an understanding of how to better deliver the current information technology services it already provides to small business clients. Equally as important to small businesses are suggestions for coping with information technology and finding ways to increase productivity given the scarcity of trained and skilled specialists. There is a growing belief that information technology will be the most important technology to change business and society in the 1990’s as Canada moves from an economy that is resource based to one that is service based (Gunning, 1992). Small businesses may end up in the unenviable position of relying on information technology much more than they currently are, and unable to find ready assistance (in the form of skilled labour) to implement and manage the information technology they require. 2For the public accountant, this research should help reinforce the need to be adequately trained in areas that will be called upon increasingly more often by current and future clients, such as information technology. Public accountants are finding themselves more and more in the position of being Information Consultants to their small business clients. The respective institutes (CICA and CGA) are portraying their members as computer (IT) professionals in national ads. This research should provide results that show if the message is getting through to the public as well as to the professional accountant. For the purposes of this paper, the term information consultants is broadly defined as “professionals who use their knowledge of inforniation technology to help individuals (i.e. clients/customers) obtain sufficient knowledge/skill level in the use of an information technology to become independent of further extensive professional aid in using the information technology”. This definition includes information centers, DP departments, computer consultants, and pjjlic accountants. The public accountant is often relied upon by the small business manager for help in installing computerized accounting services to ensure that the system will meet the accountant’s requirements as well as the manager’s. This expectation arises from the public perception of the accountant’s expertise with information technology. Public accountants now find that some 95% of their audit clients have information technology installed (Walker, 1991). Often, however, accountants are not familiar enough with automated systems and treat them as automated manual systems (which they are not), resulting in potential disservice to the client (Overbey et al, 1987). To avoid the public and private humiliation that adverse headlines tend to bring, as well as the subsequent lawsuits and loss of business, research is required that will aid the professional accountant in helping his client successfully adopt any new information technology. Despite the good reputation of information consultants, failures still occur. Practical advice based on solid research, designed to minimize the risk of failure, would be very welcome. Also, new types of information technologies are continually being developed. Inevitably, the new information technology will find its way into business. The skills to cope .3 with the introduction of the information technology need to be defined in an attempt to avoid any trepidation on the part of the client, based on past experience, that may otherwise occur. There is a general reluctance to adopt new information technology in the public accounting profession (Batch et al, 1989) as well as in other professions (Newman, 1990). If these information technology specialists are resistant to learning and adopting newer information technology, it should be no wonder that the information technology specialists experience user resistance to the introduction of even basic information technology. This research should provide motivation for the information consultant to continue on the arduous task of bringing his clients into the 1990’s by introducing a theory backed approach on how to successfully introduce new information technology into an organization. 1.2 RESEARCH DIRECTION A review of any major MIS publication will show that the majority of research in MIS is carried out on large organizations (Attewell, 1989). The result is similar for studies on how information technology affects organizations as well. It can be easy to fall into the trap of thinking that results from these studies apply equally well to small organizations. However, it has been shown that small firms differ from large firms in many areas, including job creation and growth which in turn affect many other organizational characteristics (Attewell, 1989). For example, research on the role of information consultants, such as the Information Center (IC), is generally carried out on large firms (for a typical large firm study see Brancheau & Wetherbe, 1990). However, there are few (if any) IC’s or similar entities in small firms. There has been little empirical research that has looked at the role of information consultants in the adoption of information technology in a small business setting. The role of the information consultant in the diffusion of innovations process will be examined. For small business managers this is an important issue as small firms usually lack the resources to develop necessary expertise in-house. These businesses often look to their professional accountant for advice on their information requirements. For professional accountants this is also an important issue as their associations are attempting to transform 4their members into information specialists to meet the needs of their clients. For example, the Canadian Institute of Chartered Accountants (CICA) is currently considering recognition of areas of specialization (if not accreditation) amongst CA’s, one such area being information technology (Brown, 1992, Goodson, 1990; Luscombe, 1990). 1.3 THE COMPUTERIZED ACCOUNHNG SYSTEM The Computerized Accounting System is the specific information technology of interest to the accounting profession and small business in general. The Computerized Accounting System is a special subset of the Personal Work Station which Moore studied. The Personal Work Station as defined by Moore consists of a set of computerized tools designed for an individual; is used on a microcomputer or terminal connected to a minicomputer or mainframe; is accompanied by appropriate software; and is used directly (hands on) (Moore & Benbasat, 1991). The Personal Work Station is general and not function dependent. A Personal Work Station can be used in marketing, finance, production or any other area of an organization. The choice of tools (hardware/software) comprising the Personal Work Station is usually up to the individual. A Computerized Accounting System for the purposes of this research is defined as a set of computerized tools for an individual, and usually consists of a personal or microcomputer with one or more software packages, including an accounting program and/or other software such as a spreadsheet, database, word-processing, etc. in support of the accounting function. A Computerized Accounting System is similar to the Personal Work Station defined by Moore. The major differences between a Computerized Accounting System and Personal Work Station are that the use of a Computerized Accounting System (hardware/software) is usually an organizational decision and a Computerized Accounting System supports the accounting function primarily. 1.4 TOWARDS A SMALL BUSiNESS ORIENTATION Research into information technology, now entering its third decade, has primarily focused on large organizations. Although there are several issues regarding whether or not it is necessary to study small businesses separately from other businesses, the main issue is whether the organizational factors found in small firms are sufficiently similar to those of larger firms. If the main factors of interest are common across firms then it is appropriate and economically prudent to limit research studies to large firms and extrapolate the results to all other firms, given the difficulty in obtaining results from small firms. If these factors are dissimilar, then we as researchers have been omitting a significant group of organizations from our studies and we cannot claim with confidence that our results are generalizable across all firms. This orientation towards big business is natural, as larger firms tend to operate in complex conditions. Understanding the environmental and internal factors that influence how a firm will behave is important to the enterprise and to society. This understanding is necessary because large firms have high public profiles, are large employers, and make large contributions to local economies, research institutes, and governments in the form of taxes or donations. Large firms are properly viewed as being very important to our economy. Small businesses are also important to the economy. A study on small businesses in Canada, commissioned by the Federal Business Development Bank (FBDB) in 1986 and released in 1987, found some unexpected results. Small businesses (defined as firms with sales under $2 million and typically with less than 20 employees) accounted for 25% of our GNP, 96% of all business organizations (over 700,000), created the greatest employment opportunities for women and young people (under 25 years old), had less of a wage gap between men and women, employed 32% of all workers (excluding farm, professionals, fishing and commission sales people) and over the period 1978-1982 created over 52% of all new jobs (FBDB, 1987). More recent data confirms the impact of small firms on job creation, as a study commissioned by the Canadian Organization of Small Business found small businesses created over 98% of the new jobs in the period 1984-1987 (Small Business 6Magazine, October 1989). The increasing importance of small businesses can be shown in B.C., where small businesses employed almost 60% of B.C. workers by the end of 1988 compared to under 45% in 1986 (Smith, 1989), represented 92% of all businesses (Richmond Business, 1990) and created 96% of net new jobs (Richmond Business, 1990). Similar growth has occurred all across Canada during this time. In the USA, small businesses in the late 1970’s and early 1980’s accounted for 98% of all non-farm business organizations; 39% of the GNP; and 48% of non-farm, non-government employment (DeLone, 1988). In the U.K., small firms were found to contribute 35% of all technological innovations during the period 1970-1979 (Pavitt et al, 1989) and the portion of innovating small firms (under 200 employees) has been increasing significantly over the period 1945-1983 (Pavitt et a!, 1989). The importance of small firms to the economies of Western countries is obvious. The above statistics hide the sensitivity of small firms to economic fluctuations. Even in boom times many small firms experience a rocky road. The Canadian experience in the period 1978-82, for firms employing 5 or less full time employees, indicated that for every 100 net new jobs created: 52 were in currently existing firms; 106 were for newly created firms which survived; and 58 were lost for new firms that didn’t survive (FBDB, 1987). Due to this sensitivity to the economic environment, smaller firms are often perceived to be more risky, subject to higher failure rates, have more problems collecting receivables, have more difficulty keeping adequate records (DeLone, 1988).’ It is also evident that small firms are very important to public accountants, and vice versa. There is a special, symbiotic relationship between these two groups. This relationship, while acknowledged, is not well understood and varies from country to country. It appears that many small businesses in Canada rely on their public accountants for more than their accounting and tax knowledge (Goodson, 1990; Delente et al, 1990; Hamilton, 1989), while most small firms in Australia still seek mainly year end accounting and tax services from their accountants (Holmes & Nicholls, 1989). A recent Canadian study on small firm’s relationship with their accountants found that one of the reasons small firms initially engaged their 1The researcher has encountered several small firms that have experienced most, if not all, of the above problems through his own involvement in accounting public practice. 7accountant was to install a computer system (ranked 8th on the top 10 list), a response provided by 21% of the survey firms. However, when asked about ongoing work performed by their accountants, “advice on computers” did not make the top 10 list. A significant portion of firms requested that more services, including computer systems advice, be provided by their accountant (Hamilton, 1989). 1.5 THE ROLE OF INFORMATION TECHNOLOGY 1N ORGANIZATIONS Unplanned and uncontrolled adoption of information technology are major problems for any firm (Miller, 1988). These problems could include loss of data and programs (Stulberg, 1991) and poor decisions based on unreliable information systems (CICA, 1986; Alavi & Weiss, 1986; Gremillion & Pyburn, 1983; and Davis, 1981). Any or all of these problems could lead to possible cessation of operations (Rosen et al, 1986; Allen, 1982). More recently, sabotage via computer viruses has become a real concern (Jenish, 1992; BYTE, August 1991; Rockburn, 1990; Kunz & Maingot, 1989). While most larger firms have internal resources to help overcome these problems (in-house expertise, financial resources to acquire adequate information technology) most small firms remain at risk due to their lack of resources. Factors contributing to the problem of unmanaged information technology include ignorance of the full potential of the information technology by the information consultant (Cox, 1990; Batch et al, 1989) or the user (Benson, 1983) with the user often being more concerned about the information technology’s impact on himself (Baronas & Louis, 1988); management ignorance of the skills the organization has available for using the information technology (Benson, 1983); and management reluctance or inability to provide adequate user training (Buckler, 1990 and others). For large and small firms the information technology user is often unsophisticated because the technology is new to the firm and personnel familiar with it would be relatively few (Lees & Lees, 1987). To learn to use the information technology the user has the options of relying on information consultants (Melone & Bayer, 1990; Stieren, 1990), other staff (Melon & Bayer, 1990; Nilakanta & Scamell, 1990; McFarlan & 8McKenney, 1983), or on the user’s own abilities. The extent of reliance on other skill sources depends on the individuals own skills and the organizations resources. The ability of large firms to cope with the above problems of information technology are generally better than for small firms. A problem faced by many small business managers is that they attempt to manage information technology based on practices that they are familiar with, strategies aimed at obtaining or maintaining stability. Such practices are not conducive to coping with the major change information technology tends to impose on an organization (Miller, 1988) as is the case with the initial introduction of an information technology. Most large firms have experienced these major changes several years (or decades) ago and will be more familiar in dealing with change than their smaller counterparts. In large firms users often have skilled resources to fall back on such as an EDP department or personnel who had recently come from a firm with the information technology. With the increasing complexity of computer technology even these traditional sources are finding it increasingly difficult to keep up (Geliman, 1991; Gotleib, 1990) with the result that large firms will turn to specialists (consultants) if necessary (Gotleib, 1990; Boynton & Zmud, 1987). Users in small businesses on the other hand have much fewer resources to fall back on (Willits, 1990; Delone, 1988; Lees & Lees, 1987). Often they must rely on external skilled specialists, helpful friends, or themselves (Lefebvre & Lefebvre, 1990; Gable, 1989; Delone, 1988; Lees, 1987). In many cases hiring the external information specialist is much cheaper than hiring full time EDP staff (Arter, 1988) with the result that external information consultants are commonly used by small firms (Bracker & Pearson, 1985). For the small business the specialist is often their professional advisor - their public accountant (Delente et. al, 1990; Peat et al, 1984). Recent studies show that in Canada there is a growing shortage of skilled information technology specialists (Buechert, 1992). While this shortage poses problems from businesses in general, it provides an opportunity for public accountants to fill this void. Partly in response to this trend, organizations such as the CICA have exhibited plans to expand their involvement in information technology on a large scale (Brown, 1992). 9It has been suggested that the reasons a small firm seeks outside help for managing information technology are similar to those used for seeking outside help in business planning (Gable, 1989). If this is true, then the professional accountant is the person to whom the business manager will turn as the accountant often has provided the business planning advice initially. However, success in providing a business plan doesnt ensure success regarding the adoption of information technology. The failures of information systems installed with the help of information consultants have been well documented in the media. This is particularly true for accountants (e.g. see Babcock, 1986) and the fear of lawsuits over malpractice for providing information systems or advice is a real and growing threat (Dragich, 1989; Walton & Durham, 1988). While there is research to support the claim that external accounting services help small firms to be successful (Bracker & Pearson, 1985), there are also research results that claim using external information consultants, including accountants, provide less than satisfactory results for a small business (Hamilton, 1989; Baker, 1987; Lees, 1987; Lees & Lees, 1987; Bracker & Pearson, 1985). Some of these studies indicated that higher satisfaction could be achieved if the consultant provided a full range of support and services. 10 CHAPTER 2: LITERATURE REVIEW ACCOUNTANTS: THEN “Observe that much of the difficulty in the conception of profit, taxes, costs, and so on, can be seen to come from the professionalization of the accountants as a group. They are the ones who force upon the industrial situation the concern with numbers, with exchangeable money, with tangibles rather than intangibles, with exactness, with predictability, with control, with law and order generally, etc. ... Andy Kay [then president of the company] pointed out that the accountants have the lowest vocabulary scores of any of the professional groups. I added that the psychiatrists think of them as being the most obsessional of any group. From what I know of them, they also attract to the schools of accounting those who are number bound, those who are interested in small details, those who are tradition bound, and the like.” [Maslow, 1965 quoted by Davidson, 1991]. A1D NOW “My own research ... found that members of professional accounting firms are very bright, with an average intelligence level at the 84th percentile of the general population.... For starters, accountants tend to be more assertive, independent-minded, unconventional, cheerful, enthusiastic, rebellious, experimenting, liberal, self-sufficient, careless of social rules and standards, nonconforming, anxious, independent and impulsive.” [Davidson & Dalby, 1991]. From an organizational perspective, there is a growing realization that information can be considered as an asset (Frarnel, 1990; Ahituv, 1989), albeit an intangible asset. Many firms (large and small) are turning to information technology due to the increasingly complex and competitive business environment and the recent technological and software trends making it feasible and less costly to acquire an information technology, allowing firms to better manage and protect their information (Huber, 1990; McGill, 1990). The concept of information as an asset is not new to large firms or to public accountants, but to many small businesses it is a novel idea with the result (as many public accountants can attest) that little is done to protect (Bradbard et al, 1990; Alavi & Weiss, 1986) or exploit their data. A recent BYTE survey of its readers (including large and small firms) found that 53% of respondents had suffered loss of critical data costing an average of $14,000 (BYTE, August 1991). The wide spread diffusion of information technology has left many firms open to the issue of security. Many small firms appear to be ignorant of the necessity of information 11 technology security (Pendegraft et a!, 1987). For the small business it has been suggested that security is even more important than for large firms due to the high degree of reliance on information technology (Pendegraft et a!, 1987) and to the high degree of {unrelated] third party knowledge about the use of the information technology, particularly microcomputer software (Bradbard et al, 1990; Overbey et al, 1987). Large firms tend to experience less security problems as they tend to use large computers and more restricted software. Prior experience with larger computer systems also provides large firms with an advantage in safeguarding their data and micro computer systems. There are often several reasons that firms acquire information technology. Initiative to introduce information technology is either due to a PUSH (organizational) environment or a PULL (individual) environment. A PUSH environment exists when events external to the user (or the firm) force information technology on the user. Firms acquire information technology due to competitive pressures such as improved business value indicators like return on investment (ROT) figures or net income (Alavi et a!, 1988; Kauffman & Weill, 1989) and competitive advantage (Framel, 1990; Clemons & Row, 1989; Alavi et a!, 1988; Reich & Benbasat, 1988; Boynton & Zmud, 1987). A PULL environment exists when the user finds it necessary to acquire information technology due to his own work environment. Employees may acquire information technology for higher job satisfaction (Kraut et al, 1989; Pentland, 1989; Millman & Hartwick, 1987). Additionally, non-IS employees may acquire information technology due to frustration with the IS department for delays in developing user required systems (Gremillion & Pyburn, 1983; Rockart & Flannery, 1983; Davis, 1981; McLean, 1979) or due to the inability of specialists to understand user requirements (Gremillion & Pyburn, 1983, Rockart & Flannery, 1983; Davis, 1981). The user is not only more likely to be satisfied with a system that he developed himself (Gremillion & Pyburn, 1983); but he is also responsible for the implementation (Davis, 1981). 12 Research of successful adoption of information technology has focused on measurable attributes associated with success. Over the past decade or so, the definition of success has evolved from a one dimensional point of view (i.e. see studies by DeLone, 1988; Sein et al, 1987; Raymond, 1985; McKeen, 1983; Ein-Dor & Segev, 1982; Lucas, 1978 and others) to a perspective with complex multi-dimensional features (i.e. see studies by Melone & Bayer, 1990; Rivard & Huff, 1988; Sanders & Courtney, 1985; Barki & Huff, 1984; Bailey & Pearson, 1983, Ives et al, 1983; McKeen, 1983; Zmud, 1979 and others). As a result of the increasing knowledge on information technology adoption processes, currently success is viewed as a relative term (Gallupe, 1989). In other words, success is dependent on how well there is a match between the user’s expectation of what the information technology is supposed to accomplish, and what the information technology actually does. User attitudes have increasingly been seen as an important indicator of the success of information technology adoption (Lin & Ashcrafi, 1990; Melone, 1990; Thompson, 1989; Goodhue, 1986). The research focus on user attitudes and behaviour is due to the increased emphasis on theory based constructs such as attitudes (from the social and cognitive psychology domain - for an overview of current thought on attitudes, see Pratkanis et al, 1989), where in the IS domain the concept of user attitude encompasses the success attribute of user satisfaction (Melone, 1990) as well as several other of the single dimensional attributes (Goodhue, 1986). Recent research has begun to look more closely at the process of adoption of information technology, also called technology transfer (Bouldin, 1989). The study of information technology using an adoption of innovations approach has been persued in the MIS field (Cooper & Zmud 1990; Alexander 1989; Moore 1989; Brancheau 1987; and others) and in the psychology domain (Hill et al., 1987). Much of this work has drawn from the literature on diffusion of innovations which was pioneered by Rogers (1983), and from the work on attitudes and beliefs by Ajzen and Fishbein (1980) as well as other psychologists. 13 CHAPTER 3: ADOPTION OF INFORMATION TECHNOLOGY It must be considered that there is nothing more difficult to carry out, nor more doubtful of success, nor more dangerous to handle, than to initiate a new order of things. For the reformer has enemies in all those who profit by the older order, and only lukewarm defenders in all those who could profit by the new order. This lukewarmness arises partly from fear of their adversaries, who have the laws in their favor, and partly from the incredulity of mankind, who do not truly believe in anything new until they have had an actual experience of it. (Machiavielli, Niccolo [1500’s], The Prince, Translated by Luigi Rice, Rev. E. R. P. Vincent, New York: New American Library, 1952; cited in Foundations of Business Systems (Flaaten et al, 1989, pg. 37)). A general criticism about information systems research has been the lack of an adequate theory of IS (Goodhue, 1986). There is considerable confusion on the issue of what a successful ny’brmation system is (Goodhue, 1986). The recent research on information system attitudes and adoption of innovations has begun to clear up this confusion. The current view of information system which incorporates these concepts have been described by Boon & Pienaar (1989, pp. 122): “Technology is not an end in itself but merely a means to an end, the end being to help knowledge workers to do their jobs effectively and efficiently. Knowing these knowledge workers and what they are doing, as well as the information technology, would result in appropriate and successful application of the technology’ (emphasis added). The issue of knowing what the knowledge workers are doing is addressed by Moore in his study. Moore (1989) has developed a general model, the Diffusion of Information Technology model (see Figure 1 - Diffusion of Information Technology), that explains the adoption and use of information technologies by individuals. This model was integrated from concepts contained in the Diffusion of Innovations model by Rogers (1983) (see Figure 2) and the Theory of Reasoned Action by Ajzen and Fishbein (1980) (see Figure 3), to explain the adoption of information technology by individuals. In developing this model, Moore has attempted to overcome the previously noted weaknesses (ie. lack of theory, measuring information system success) in research in this area. The Moore model is the most comprehensive and theory backed work to date on information technology diffusion and adoption. 14 3.1 DIFFUSION OF INNOVATIONS The Diffusion of Innovations work by Rogers is well supported by research. Rogers’ Dffusion ofInnovations model is used to explain the rate of adoption of innovations (Rogers, 1983), which included five perceived attributes of innovations; type of innovation-decision (individual or collective decision); communication channels (media or interpersonal contact); nature of the social system (social norms, interconnectedness of the communication network); and extent of change agent’s (product champion or opinion leader) promotion efforts - see Figure 2. Based on a synthesis of the literature and research on innovations Rogers (1983) has determined that there were five attributes of innovation that are all conceptually distinct from each other (Relative Advantage - the degree to which an innovation is perceived as being better than its precursor; Compatibility - the degree to which an innovation is perceived as being consistent with the existing values, needs, and past experiences of potential adopters; Complexity - the degree to which an innovation is perceived as being difficult to use; Trialability - the degree to which an innovation may be experimented with before adoption; and Observahility - the degree to which the results of an innovation are observable to others). Moore added an additional two attributes (Image - the degree to which use of an innovation is perceived to enhance one’s image or status in one’s social system; and Visibility - the degree to which the innovation is apparent to the sense of sight) and called the resulting seven attributes Perceived Characteristics ofInnovation . An additional related variable, Voluntariness of use (the degree to which use of the innovation is perceived as being voluntary, or of free will), was also added by Moore and the variable Complexity was renamed Ease of Use. Finally, Moore added some scales to measure Computer Avoidance (a term which he did not define). Appendix I-A contains a summary of a complete list of Perceived Characteristics of Innovativeness variables and Voluntariness definitions. 15 3.2 THE THEORY OF REASONED ACTION In the Theory ofReasoned Action, which is well supported by research studies, Ajzen & Fishbein identified the relationship between intentions, beliefs, attitudes, and behaviours (Ajzen & Fishbein, 1980). The basic premise is that an individual’s behaviour is determined by his decision or intention (which is reasoned) to perform that behaviour. The attitude toward the specific behaviour (an individual’s personal attitude towards the behaviour) and his Subjective Norms (the individual’s perception of what other people think about the behaviour) determine the individual’s behavioural intention. These attitudes and subjective norms are a function of the individual’s belieJ. The basic premises of the Theory of Reasoned Action are illustrated in Figure 3. 3.3 DIFFUSION OF iNFORMATION TECHNOLOGY The link between the Diffusion of Innovations model and the Theory of Reasoned Action can be seen in Figure 1. The synthesized Diffusion of In!hrmation Technology model developed by Moore can be described as follows (Moore & Benbasat, 1990, pp. 3): “Innovations diffuse because of the cumulative decisions of individuals to adopt them. Thus, it is not the potential adopters’ perceptions of the innovation itself, but rather their perceptions of using the innovation that are key to whether an innovation diffuses.” To test the Diffusion of Information Technology model, a questionnaire was developed and administered in a cross sectional study involving individuals in six Canadian organizations. The questionnaire results supported all eight Perceived Characteristics of Innovation variables as being factors in explaining the diffusion of Personal Work Stations, which was the particular innovation being investigated (Moore & Benbasat, 1990). 3.4 SIGNIFICANCE OF THE DIFFUSION OF INFORMATION TECHNOLOGY MODEL The Diffusion of Information Technology model attempts to predict, explain and influence individual behaviour towards the adoption of information technology. The Diffusion of Information Technology model is also designed to be a general model (Moore, 1989). As a general model, the DiffiTsion of Information Technology model should apply to a 16 specific information technology other than the Personal Work Station, such as Computerized Accounting Systems. The Diffusion of Information Technology model should also apply equally well to small businesses and large businesses. These observations about the Diffusion of Information Technology model arise from an inspection of the theory on which the model is based. Because the Diffusion of Information Technology model is based on theoretical models, it will contain the characteristics of the underlying models. An important characteristic of the Theory of Reasoned Action is the ability to predict, explain and influence individual behaviour (Ajzen & Fishbein, 1980). The Theory of Reasoned Action is generalizable and is applicable to all people. The Diffusion of Innovations model focuses on the adoption of innovations. The Diffision of Innovations model should be generalizable across all innovations, including information technology. It is important to examine whether the Diffusion of Information Technology model is sufficiently robust to include small businesses as part of the population it encompasses. Research models that are of help to small firms are few and far between. This model could provide a means for explaining why a particular innovation, such as installation of a computerized accounting system, succeeds in one firm and not another. It could also be used for predicting if the innovation is likely to succeed, before significant time and resources are committed to a project, by determining if the firm has an adequate mix of similar attitudes and beliefs as those found for the successful adopters. 17 CHAPTER 4: TECHNOLOGY TRANSFER - RESEARCH QUESTIONS “We have lived through a bewildering array of new hardware and software technologies, most of which are supposed to increase the productivity of the average programmer and systems analyst by a factor of ten or more; yet these technologies are not even being used in many DP organizations and have achieved only modest results in many others.... It took the military 75 years to go from the technology of muskets to the technology of rifles, so we should not be too discouraged to learn that it takes 14-15 years ... for new software technologies to be accepted” (Yourdon, found in Bouldin, 1989, pg. xiii). As discussed in the previous section, the D/jiision of Information Technology model has the ability to predict and explain individual behaviour towards the adoption of information technology. An underlying reason for this current study was to verif’ the robustness of Moore’s results for the Diffusion of Information Technology model. Equally important was to determine the potential usefulness of this model to the public accountant (information consultant) as well as to their small business clients. For the small business manager or information technology specialist, it is expected that by understanding the factors that lead to successful adoption of information technology a systematic approach can be developed to influence individual behaviour to adopt new information technology in the future. Besides attempting to extend the Diffusion of Information Technology model to the small business domain, this study attempted to obtain new knowledge regarding the appropriateness of using the theory based work of Rogers and Fishbein & Ajzen in the MIS domain. The Communication Channels section of Roger’s model, coupled with the Extent of Change Agent’s Promotion Efforts section (Figure 2) and the Connnunications Network section of Fishbein & Ajzen’s model (Figure 3) essentially represent the same concept - information gathering/exchange (for convenience the term Communication Channels will be used in the remainder of this thesis). Moore’s research did not investigate this area. However, the importance of communications channels in the adoption process should not be underestimated, as it has been pointed out that 18 “before a business unit can adopt and use a technology, members of the business unit must become knowledgeable of the technology and be able to propose ideas for its use. This awareness results from communication behaviors ... whereby a ‘technology provider’ familiar with the technology interacts with a potential ‘technology user’ not familiar with the technology” (Lind & Zmud, 1990, pg. 4). For a small business, the technology provider, likely an external consultant (Gable, 1989), is often the accountant (Goodson, 1990; Hamilton, 1989). The role of external consultants as information sources on information technology has not been well established in the Diffusion of Information Technology literature (Gable, 1989). Unlike most external consultants, accountants are often considered to be an integral part of their client’s management team (Delente et al., 1990; Goodson, 1990). For many small business managers, the opinion of their accountants are highly regarded and persuasive (Goodson, 1990). 4.1 RESEARCH HYPOTHESES The Diffusion of Information Technology model provides a means to determine just what the characteristics of a successful interaction between the user and a specific information technology are. Interactions between users and an information technology are registered by means of a questionnaire that Moore has developed and validated. Moore’s questionnaire did not focus on the role of external information consultants, probably as a result of his focus on large business adoption of information technology where the necessary expertise would be available in-house through the Information Centre or similar department. As small firms do not have a similar body of in-house information expertise, the role of external information consultant becomes more important. This specific item may provide an important research area for small firms. As validating Moore’s results regarding the Diffusion of Information Technology model is one goal of this study, a summary of the Moore hypotheses (modified to reflect the Computerized Accounting System) is provided below. Hj: One attitude towards using a Computerized Accounting System will influence one’s innovativeness with respect to Computerized Accounting System usage. 19 H2: Relative Advantage will have a contribution more than any other Perceived Characteristics of Innovation on one’s attitude towards adopting Computerized Accounting Systems. H3: Computer Avoidance will have a contribution less than any other Perceived Characteristics of Innovation on one’s attitude towards adopting Coniputerized Accounting Systems. 114: The Subjective Norm will influence one’s innovativeness with respect to Computerized Accounting System usage. H5. The Subjective Norni will influence one’s attitude toward adopting the Computerized Accounting System. H6: Voluntariness is negatively related to one’s innovativeness with respect to Computerized Accounting System usage. H7: Voluntariness i’ill be negatively related to one’s attitude towards using Computerized Accounting System. An important research question for small business managers arises concerning the role that Support groups, especially external information consultants such as accounting professionals, play in the process of information technology diffusion. The research hypotheses related to this question are developed in the following paragraphs. It has been shown, in Chapter 1, that small and medium firms rely on external consultants more than large firms. Because small firms have little in-house expertise in information technology, especially for an important information technology such as a Computerized Accounting System, the involvement of an external source of information and guidance should contribute to the success of the introduction and adoption of the Computerized Accounting System. H8. The involvement of a Support Group i’ili contribute to a successful adoption of Computerized Accounting Systems. 20 As a Support Group is made up of different components, it follows that each of these components should contribute to a successful Computerized Accounting System. For the purposes of this study, the Support Group is comprised of Friends, other Employees, external Accountant, and external Consultant. This group generate the following hypotheses: H9. The involvement ofa Friend will contribute to a successful Computerized Accounting System. H10. The involvement of other Employees will contribute to a successful Computerized Accounting System. H1]• The involvement of an external Accountant will contribute to a successful Computerized Accounting System. Hp. The involvement qf an external Consultant will contribute to a successful Computerized Accounting System. An investigation of the Moore, Fishbein & Ajzen, and Rogers models indicate that the presence of a communications channel will influence other areas of the Diffusion of Information Technology model as well as Innovativeness. In this study, communications channels is represented by the Support Group. The Fishbein & Azjen model (Figure 3) shows direct links from Communications Network to Subjective Norm and Attitude. These links suggest the following two hypotheses: H13. The involvement of a Support Group will have a positive influence on Subjective Norm. H14. The involvement ofa Support Group will have a positive influence on Attitude. Also, while not explicit in the Fishbein & Ajzen model, it is possible that the Perceived Characteristics of Innovation variables may also be influenced by the communications channels. This link is suggested from a review of the adoption process indicated in Roger’s Stages of the Innovation Decision Process model (Figure 4), where the Knot ‘ledge/Persuasion cycle (incorporating the communications channels) impacts the 21 Decision cycle (which incorporates the behavioural intention, which are shaped by Perceived Characteristics of Innovation variables). Finally, the perceptions of several Perceived Characteristics of Innovation variables in Figure 1 (ie. Trialability, Visibility, Relative Advantage and Image) can be influenced by how other people (eg. Support Group) perceive or present information technology. From these observations an additional hypothesis can be generated. H15. The involvement of a Support Group will have a positive influence on Perceived characteristics ofInnovation variables. 22 CHAPTER 5: 1NSTRUMENT DEVELOPMENT SECTION A - TNTRODUCTION 5.1 GENERAL In this section the development of the two questionnaires used in the study will be discussed. Reliability results for both the pilot study and the final study will be presented. The use of questionnaires as a method of gathering research data is both common and controversial. It is common because it is convenient and often the only feasible way a researcher can obtain sufficient volume of data in an economical manner. It is controversial as the method is susceptible to a number of sources of error that could render any results suspect. A good questionnaire must therefore strike a balance between its length and complexity, presenting to respondents a form that isn’t intimidating, while obtaining data that is reliable and valid. Moore spent a considerable amount of time establishing the reliability and validity of his questionnaire. The changes to the Diffusion of Information Technology questionnaire discussed in the next section were of a type to potentially call into question its reliability but not its validity. The changes made were generally cosmetic, substituting Computerized Accounting System for Personal Work Station and cleaning up terminology to be more consistent with a small business environment. These changes were not expected to affect the focus of the questionnaire from the underlying theoretical foundations, therefore the validity of the questionnaire should not have been affected. Changing the wording of individual questions may have affected how they were interpreted, which is a reliability issue. As a result reliability issues will be dealt with in more detail than validity issues. 5.1.1 RELIABILITY Reliability is defined as “the degree to which the results of measurement are free of error” (Stone, 1978). Generally, there are two components to any measurement, a “true” component and an “error” component. A reliable measurement instrument is one that has a 23 low error component. In other words, repeated use of the instrument gives consistent results. Also, a measure is considered to be reliable when independent but comparable measures for the item of interest provide similar results (Churchill, 1979). The appropriate level of reliability is a factor of the goals of the researcher and published criteria for the type of research being done. Reliability numbers range from 0 to 1 and are usually presented as decimal fractions, where the higher the fraction the better the reliability. The general rule of thumb for a reliability outcome is .80 (Bryman & Cramer, 1990). For the purposes of this study, a reliability figure of .70 will be used as this level of reliabilty is appropriate for a study that is in the early stage of theory testing (Nunnally, 1978) and is also an acceptable itt/c of thumb level for PLS analysis (Barclay et. al., 1991). It should be stressed that .70 is the lower bound for an acceptable reliability level. 5.1.2 VALIDITY Validity is defined as ‘the degree to which a measure actually measures what it purports to” (Nunnally, 1967, pp. 75). In other words, the differences observed are true differences for the characteristics being investigated and not a result of some other source (Churchill, 1979). There are several items comprising validity which are summarized in Appendix I-C. It should be noted that not all of these factors may be an important issue with any given questionnaire, but they should at least be considered upon preparation. Validity is not considered to be a problem in this research as the questionnaire used was previously validated by Moore. Changes made to the questionnaire for this study did not fundamentaly alter what the questions were meant to measure. For example, questions meant to measure Image still measured Image, only the Image being measured was for a Computerized Accounting System (Modified question U-6: Using a CAS improves my image within the organization) and not a Personal Work Station (Original question U-6: Using a PWS improves my image within the organization ). This substitution of CAS for PWS occured for all 39 questions. 24 5.1.3 QUESTIONNAIRE SELECTION The research issues being investigated indicated that two separate questionnaires were required. One questionnaire to test the Diffusion of Information Technology model and simultaneously gather data on users (clients) information technology information sources, the second questionnaire to elicit data from accountants. The development of each of these questionnaires is discussed in the following sections. SECTION B: QUESTIONNAIRE DESIGN - PILOT STUDY All references to questions in this section refer to the Pilot Study questionnaire. 5.2 CLIENT/DIFFUSION OF INFORMATION TECHNOLOGY QUESTIONNAIRE 5.2.1 PERCEIVED CHARACTERISTICS OF INNOVATIONS One goal of this research study is to replicate the results Gary Moore obtained validating his Diffusion ofInformation Technology model. Moore spent considerable time and effort in developing a questionnaire that met suitable reliability and validity criteria (see Moore & Benbasat, 1991). It was determined that redeveloping an alternate questionnaire would be redundant, fitile, and not contribute to a cumulative discipline. Therefore, Moore’s questionnaire was adopted with some minor modifications which are discussed below. In Moore’s study, the measurement of Perceived Characteristics of Innovations was obtained through the use of an interval scale (ranging from 1 to 7) consisting of 50 questions. These 50 questions were used to measure the 9 Perceived Characteristics of Innovation variables that Moore considered integral to the Diffusion of Information Technology model. Based on subsequent analysis of the Diffusion of Information Technology model using LISREL (Linear Structural Relations Model), Moore was able to determine that only 8 Perceived Characteristics of Innovation variables were significant factors. Moore also was able to determine that the Perceived Characteristics of Innovation questions could be trimmed down to 38 from 50 without significantly affecting the results (Moore & Benbasat, 1991). In this paper, references to Moore’s questionnaire will refer to the 38 item instrument 25 unless otherwise noted. In this current study, Moore’s questionnaire was modified by changing all Personal Work Station references to Computerized Accounting System to focus the study on the information technology Computerized Accounting System. Altering the questionnaire introduced the risk that the instrument no longer measured what it was supposed to measure. The modified questionnaire was tested by a pilot study on a sample of small businesses and compared to Moore’s results to establish that the modifications did not fundamentally alter the reliability of the questionnaire in relation to Moore’s Diffusion of Information Technology model. The major risk inherent in this approach is if the pilot study does not produce statistically similar results, it will be difficult to determine if the results are from the changes to the instrument or from difference between large and small firms. Due to this potential problem, an additional pilot study was contemplated to be carried out on a relatively unmodified version of Moore’s questionnaire. The only modification to this questionnaire would be the substitution of Computerized Accounting System for Personal Work Station. The results from these two pilot studies would be compared to each other and to Moore’s results to ensure that the overall integrity of the questionnaire was not damaged. Any differences between the two pilot studies could be attributed to changes in the questionnaire, while differences between the pilot studies and Moore’s study could be attributed to differences between large and small firms. As it turned out the results for the pilot study were statistically similar enough to Moore’s findings to dispense with the second pilot study. The pilot study results were compared using the reported reliability figures (Cronbach’s alpha) for each Perceived Characteristics of Innovation variable to Moore’s results. Pilot scores of .60 and higher were considered as acceptable as reliability scores tend to increase with larger sample sizes (Nunnally, 1978). This pilot study had all Perceived Characteristics of Innovation variables except Visibilty (.28) and image (.59) reporting scores above .60 (see Table 1). The Perceived Characteristics of Innovation variable Visibility had a reliability score much lower than the minimum acceptable and was examined more closely. Upon reviewing Moore’s 26 rationale for using a subset of his original questionnaire it was decided that Visibility could be improved by adding an additional question to the questionnaire, bringing the Diffusion of Information Technology subset of the questionnaire up to 39 questions. This additional question had originally been dropped, by Moore, from the 50 item questionnaire in developing the 38 item questionnaire. On the whole, the reliability results were encouraging. It was decided that the modified pilot study questionnaire would be used in the actual study. The Perceived Characteristics of Innovation questions were labeled U-i to U-39 for Computerized Accounting System users and N-i to N-39 for non-Computerized Accounting System users (for Pilot Study questionnaire see Appendix IT-A, for final questionnaire see Appendix IT-B). 5.2.2 SYSTEM USAGE The adoption of a Computerized Accounting System is the dependent variable of interest in this study. Like other success measures, measurement of adoption is difficult and surrogate items are often used, such as system usage. After reviewing the literature it became evident that usage was commonly measured by using one or two items. This is disturbing as reliability is impossible to establish based on a measure of one and difficult for two items. Even in Moore’s study this practice was followed (Moore, 1989). However, as Moore argues in his thesis, with dependent variables this is not as major a drawback as it is for independent variables. As validating Moore’s model is an important part of this study, it was determined to use similar usage measures as those used by Moore. Adoption is measured by determining the usage of the information technology (the Personal Work Station). The usage measures are called Innovativeness. There are three aspects of innovative behaviour that were measured in his study, these are Adoptive Innovativeness - degree to which an individual is relatively early in adopting an innovation, Use Innovativeness - degree to which an individual puts an innovation to use within a given use domain, and Implementation Innovativeness - degree to which an individual who has adopted the innovation uses it to solve novel problems, or in a new use domain (Moore, 1989). The Innovativeness measures and definitions are summarized in Appendix I-B. 27 Adoptive Innovativeness Adoptive Innovativeness was considered to be the time of first use of the Computerized Accounting System. Two questions were included in the pilot questionnaire to measure this item. These questions were day and month the CAS was first used (B—3) and the number of months the CAS was regularly used (B-8b). A reliability scale was developed by converting both questions to an interval scale from 1 to 7 (1=less than one month; 2=between 1 and 3 months; 3=between 3 and 6 months; 4between 6 and 12 months 5=between 12 and 18 months; 6=between 18 and 24 months; 7=more than 24 months). A reliability score of .97 was calculated (see Table 1). The results were encouraging enough to leave these questions unmodified. While it is preferable to use more than two items for reliablity testing, the resulting reliability score was high enough to indicate that a third question would not be required. Implementation Innovativeness Implementation Innovativeness was measured by asking questions on hours of use and frequency of use. The idea behind these questions was to determine the degree of use the Computerized Accounting System was currently receiving. There were two questions for hours of use in the Pilot study, overall weekly use of a CAS in hours (B-4) and weekly use, in hours, broken down by Jthiction (B-8a). Before a reliability score could be determined for these two scales, the question on hourly use broken down by function (B-8a) was converted to a single number by summing the hours of each function used, in order to make the two measures similar in nature. A reliability score of .80 (see Table 1) was achieved. While the reliability score was acceptable, a review of the questionnaires indicated that there were problems that some respondents had in answering these questions consistently. The basic problem was that the process of summing hours of Computerized Accounting System usage for functions in B-8a resulted in a single total that seldom equaled the hours reported in the overall weekly usage scale (B-4). Often th totals resulting from adding hours reported in question B-8a were considerably higher than the 28 overall number reported in question B-4. It was reasoned that individuals are probably more likely to accurately remember how much they use individual Computerized Accounting System functions than to quickly provide an overall estimate of their time using all Computerized Accounting System functions, therefore it was decided to drop the overall CAS usage measure (B-4: Overall, how many hours per week do you use a CAS?) from the questionnaire and to rely on the question measuring CAS usage byfunction (B-8a: On average how many hours per week do you spend using the CAS on the following functions? ...). It was also decided not to develop a replacement question for the item dropped as the best alternative would have been to obtain actual usage figures. This alternative was not feasible as the researcher had no access to the respondents’ place of work to measure usage due to confidentiality. Judging fiom the researcher’s own experience working with small businesses, it was also unlikely such records existed in small firms either. Frequency of Use in the Pilot study was measured by three questions. Two of the scales asked the same question, using slightly different wording. Both scales (B-5: How regularly do you now use a CAS?, B-Il: I have been using a CAS for ...) measured Computerized Accounting System usage in an overall manner. The third question measured frequency of use of individual CAS functions (B-7: On average, how frequently do you currently use the following functions? ...). All three questions used a seven point ranking scale (1= Not at all, 2Less than once per month, 3=About 1-3 times per month, 4=About once per week, 5=About 2 to 4 times per week, 6=About once per day, 7=More than once per day). After reviewing the responses to these questions it was decided not to use the question measuring use by individual Computerized Accounting System functions (B-7) in determining a reliability score due to problems in interpreting these responses. For example, a person could use several functions about once a week (indicated by a “4” on the scale for B-7) yet report using a CAS more than once per day (indicated by a “7” on the scale for B-5 or B-i 1). These different reponses could arise due to the timing of use of each function. This same problem was noted by Moore. 29 The two scales (B-5, B-Il) were thought to ask the same question, nleasuring Computerized Accounting Systeni usage in an overall manner, and were included to determine if respondents were answering consistently. A reliability score of .22 (see Table 1) was obtained for these two measures which was very surprising, given the similarity of these two questions. The responses were reviewed as were these two questions. A possible explanation is that respondents interpreted B-5 (How regularly do you now use a CAS?) in the present tense and B-i 1 (I have been using a CAS for ...) in the past tense. The inclusion of the same 7 point ranking scale (discussed above) should have caused respondents to answer the questions similarily. Dropping one of these questions (B-il) was considered; however it was decided to retain this question to see if similar results would occur in the full study. Use Innovativeness The Pilot study included four questions designed to measure system usage, called Use Innovativeness by Moore. These four questions were: did the f/rn; use a inamframe or micro (B-6), how frequently the functions were used each day (B-7), how many hours per week eachfunction ias used (B-8a), and how long the user had been using the function, in months (B-8b). An overall Use Innovativeness reliability score was calculated by taking the average number of functions used for each question. The reliability score was found to be low, .45 (see Table 1). Further reliability calculations were performed on a reduced subset of questions and it was found that by dropping the question did the firm use a mainframe or micro (B-6) the score improved to .77. The Pilot study indicated that all respondents only used microcomputers, which made sense for a small business environment. It was decided to drop B-6 from the final questionnaire for the above reasons. 30 5.2.3 CLIENT COMPUTERIZED ACCOUNTING SYSTEM SUPPORT General An important part of this study was to examine the role of the support group in Computerized Accounting System adoption. A series of questions were asked regarding the makeup of the support group and the role they play in helping the client with the use of the client’s Computerized Accounting System. Current Support For Computerized Accounting System users, there were five questions designed to measure the composition of the support group. These questions were currently receive continuing support (B-13), ...iast JO source(s,) of CAS support (B-14), ...where to go fneed Computerized Accounting System help (B-i 5), ...rating of satisfaction with support group (B-17), and ...rating qf effectiveness of support group (B-19). Because each question measured different aspects of support, the results were transformed to a binary measure for each support group (i=support, O=no support). This treatment resulted in a reliability score of .94. Based on follow up conversations with some respondents it appeared that B-i4 was confusing. A reliability measure of .93 was obtained on the other four questions. As there were several comments about the length of the questionnaire, it was decided to drop B-i4 from the final questionnaire, resulting in a shorter questionnaire and only a minor reduction on reliability. SECTION C: FINAL SURVEYS - SCALE RELIABILITIES 5.4 GENERAL Although full details of the full study are provided in the next chapter, the reliability scores for the various measures are summarized in Table 2 found at the end of this chapter. For the Perceived Characteristics of Innovation variables, all 75 respondents are included. For the scales measuring Innovativeness and client Computerized Accounting System support, 31 only the 53 Computerized Accounting System user questionnaires were included, as the 22 non-Computerized Accounting System user questionnaires did not capture any of this information. 5.5 RESULTS As shown in Table 2, all of the results are above the minimum 70 except for Result Demonstrability (.43), and Voluntariness (.69). The reliability scores generally indicate that the modifications made in the Pilot study achieved their intended purpose, to produce a questionnaire with acceptable reliability scores. The implications of these results, including Result Demonstrability, will be looked at in more detail in the next chapter. It can be concluded that the scales can be used with confidence across different domains (firm size) and different information technology. It was very encouraging to see that the use frequency was .97 (Table 2) compared to the pilot study results of .22 (Table 1). This improvement in reliability appears to be a result of the respondents in the final sample interpreting the two questions similarily (pilot study B-5, final study B-4: How regularly do you now use a CAS?; pilot study B-il, final study B-9: I have been using a CAS for ...), as they were intended to be, while the pilot study group generally interpreted the questions differently. The final reliability results include respondents from both the pilot study and final study. SECTION D: QUESTIONNAIRE DESIGN 5.6 GENERAL The original intention was to follow closely the design and layout used by Moore. This approach was considered the most appropriate as Moores questionnaire design was based on the Total Design Method which had been designed and tested by Diliman (1978). This method was reported to have resulted in very high response rates. There were some variations from Moore’s approach that were adopted due to a variety of reasons. These are discussed later in this section. 32 5.7 FORMAT 5.7.1 PAMPHLET The questionnaire was set up in booklet format, with coloured pages separating the major sections of the questionnaire. A covering letter from UBC was also attached to the front of the questionnaire. Moore had chosen this format in order to improve the overall appearance of the questionnaire in an attempt to make it appear more professional and worthy of a good response (Moore, 1989). After presenting a copy of the questionnaire to the Partners in one of the accounting firms participating in the study, and discussing the possible distribution of a similar questionnaire to their client base, it was determined that some changes would have to be made. The Partners considered the questionnaire too long in appearance and that many of their clients would simply not fill it out, even though the covering letter stated that not all of the questionnaire was to be filled out. It was decided to split the questionnaire into two parts, one part for Computerized Accounting System users and one for non-Computerized Accounting System users. This approach was used for a number of reasons. First, it was expected that there would be differences between Computerized Accounting System users and non-Computerized Accounting System users. Separating the questionnaire based on this consideration was consistent with the objectives of the research. Secondly, the researcher would not have direct access to the client base of participating Accounting firms. Because the Partners or someone knowledgeable in each Accounting firm were to do the distribution to their clients, they would know if the intended recipient was a user or non-user and distribute the appropriate questionnaire. Finally, the questionnaire each potential respondent was to receive would be approximately half the size as originally designed which should enhance willingness to participate. These factors made the splitting of the questionnaire practical and desirable. 33 5.7.2 QUESTION LAYOUT The questionnaire layout was organized in a manner that emphasized reduction in the number of pages. This was done by rearranging the appearance of several of the questions so that they were horizontally oriented and not vertically oriented. This approach was taken because of early feedback received on the apparent length of the questionnaire, even after splitting it into two parts. The Partners used to review the questionnaire were very cognizant of how their clients would respond to lengthy questionnaires. Instructions on how to answer questions were placed at the start of each section. Additionally, embedded in each question were instructions on how to answer that specific question. At the end of each section encouragement was provided to complete the remaining part(s) of the questionnaire. 5.7.3 COVERING LETTER Two covering letters were prepared for distribution with each questionnaire. One was printed on IJBC letterhead and explained the purpose of the research as well as the confidentiality of the replies received. The second letter was prepared on the letterhead of the participating accounting firm and explained that the firm was not sponsoring the study but believed the results would be useful. Encouragement to participate and confidentiality were stressed in this letter also. Both of these covering letters (see Appendix IT-B) were designed after extensive consultation with Partners from different accounting firms and with the thesis supervisor. It was emphasized to the Partners that the wording of the second covering letter (the accounting firm letter) was a suggestion only and that they were free to make changes as they chose. The rationale behind this approach was to win Partner support for helping out in the survey by allowing them to participate in the design of a part of the questionnaire (the covering letter) and to present to the client a package that would encourage them to participate in the study. 34 CHAPTER 6: DATA COLLECTION AND ANALYSIS SECTION A: DATA COLLECTION AND CONDITIOMNG 6.1 INTRODUCTION This chapter will present the data collection and analysis on the final versions of the questionnaires used in this study. The reasoning behind the sample selection, the data integrity checks performed, the statistical analysis and results will be discussed in some detail. Before proceeding with this discussion a brief summary of the goals for this study are presented. The prime objective of this study is to establish the role that public accountants play in the introduction and adoption of information technology in small businesses. This type of information is vital, as several research studies have shown that public accountants are not getting the message out, to their members and to the small business community, that accountants are skilled information technology specialists (see Hamilton, 1989; Batch, 1989). As part of this analysis, the Diffusion of Information Technology model developed by Moore will be examined in a Small Business setting, using Computerized Accounting System as the information technology of interest. This will be dine in order to evaluate whether (i) the Dffusion of Information Technology model is generalizable across firm size and (ii) different information technologies than those examined by Moore when developing this model. Recall that the major differences between a Computerized Accounting System and Personal Work Station are that the use of a Computerized Accounting System is usually an organizational decision, and a Computerized Accounting System supports the accounting function primarily, whereas the use of a Personal Work Station is often a personal decision and a Personal Work Station may encompass any functional area in an organization. 6.2 SURVEY SAMPLE 6.2.1 TARGET POPULATION SELECTION The target population is the client base of public practice accounting firms. Most small businesses use a public accountant for tax purposes or for preparation of financial statements. 35 However, not all firms decide to use a public accountant. There may be differences between firms who use public accountants and those who don’t.2 The sample is drawn from the client lists of public accounting firms (CA and CGA). A convenience sample of small to medium size accounting firms in southwestern B.C. were contacted to elicit interest in the study as these accounting firms were the most likely to have large numbers of small business clients. Due to the method of selecting the sample certain biases may have been introduced that may affect the generalizability of the results. A potential regional bias may restrict the generalizability of the results to the rest of the province or outside of B.C. A regional bias may exist due to possible differences in individuals’ attitudes towards adoption of information technology, in southwestern B.C., relative to the rest of Canada. Because the Diffusion of Information Technology model measures individuals’ attitudes, any bias would affect the results. Other regional biases may exist at the firm level as southwestern B.C. may have a larger than average number of small businesses concentrated in specific industries. These industries could have their own peculiar rate of adoption, independent of an individual’s propensity to adopt. Also, there could be a bias between small businesses in large cities and small cities. Additional regional bias could be introduced at the public accounting firm level. B.C. public accounting firms could have different levels of knowledge or initiative towards introducing information technology to their clients. These potential biases inherent in this study should not greatly affect the objectives of this study (i.e. generalizability). One objective of this study is to provide a predictive instrument that can be used to help small firms successfully introduce an information technology such as a Computerized Accounting System. This objective would be met by successfully replicating Moore’s results. The Diffusion of Information Technology model should work as successfully in B.C. as any other province; therefore regional biases should not 2Although there is no reliable information on the number of firms that don’t use public accounting firms, this number is generally accepted to be small. Firms that fall into this category include inactive or nearly inactive companies. The inclusion of these firms in the study would cause misleading results as IT is not likely to be a priority with low activity firms. Public accounting firms are not likely to be interested in inactive businesses either, as these firms are not likely to become clients nor pay their accounting fees. 36 be an issue. Also, members of public accounting firms (CA and CGA) must all take Canada wide exams as well as continuing Professional Development courses. All of these professionals will have a similar educational exposure to information technology which should help reduce regional differences amongst public accounting firms level of knowledge about Computerized Accounting System. Data collection involved the use of survey instruments, with data analysis performed on self-reported data. Directed interviews were considered as a multi-method approach is considered appropriate for generating more assurance on the validity of the findings. However, the multi-method approach proved to not be feasible and the directed interview approach was abandoned. 6.2.2 PROBLEMS ENCOUNTERED No different than any other research project, this one had its share of problems from the onset. Due to the volume and variety of problems encountered it was considered justifiable to devote a seperate section discussing these problems and their impact on the study. The sample size of directed interviews could not be increased beyond five or six due to the promise of confidentiality made to all participants, especially clients. At one point, arrangements were made with selected accounting firm personnel for follow up interviews, but conditions in the working world interfered with the follow up process to a point where the whole process was abandoned. Initially, a couple of key people went on two to three week holidays shortly after agreeing to be interviewed, When they returned it was considered that too long a time period had elapsed to put confidence in their responses. Additionally, some participating accounting firms (along with participating personnel) decided to back out of their commitments. It was too late to recruit new participants as the remaining accounting firms had already distributed questionnaires to their personnel and clients. Coupled with the problems of holidays and attrition of participating accounting firms, an untimely mail strike hampered data collection efforts severely. It appears that many questionnaires that were delivered to clients during this time were either not filled out or 37 mailed in. As no facility to follow up on non respondents was available, these lost respondents could not be recovered. Also, by the time the strike was over, the participating accounting firms had entered the start of their busy season and distribution of questionnaires was given low priority. Regaining the initial enthusiasm exhibited by participating accounting firms proved to be difficult. Data collection became a tedious task as researcher phone calls would often not be returned and promised actions would not be delivered. 6.2.3 RESPONSE RATES A total of 283 questionnaires were distributed to accounting firms and other contact people for distribution. Of these, 120 were returned by contacts who had decided to end participation in the survey, resulting in a total of 163 questionnaires being distributed to various clients. A total of 56 usable questionnaires were returned (no breakdown is available on how many client firms responded) resulting in a response rate of 34%. This response rate was lower than expected. A higher response rate was expected as the contact people had agreed to solicit agreement to participate from their clients before distributing the questionnaires. Based on follow up discussions with some of these contacts it appears that some firms sent the questionnaires out without consulting with the clients, while others contacted the clients first and then sent out the questionnaires. It also seems that some clients did not fill out the questionnaires even though they had told their contact that they would. Additionally, some contacts may not have distributed all of the questionnaires allocated to them. This lower than expected response rate resulted in a change in approach to analysing the Diffusion of Information Technology model by using FLS instead of LISREL. It was decided that the 19 responses from the pilot study would be included in the data analysis in order to have enough questionnaires to use PLS. All results reported for the final survey, including reliability results, included the pilot questionnaires. The pilot questionnaires were included as there were only minor differences in the two questionnaires for the research issues in question. A convenience sample of clients of B.C. public accounting firms was used due to various constraints. Face to face contact with individuals of the participating public 38 accounting firms (and with selected clients) was required for purposes of cultivating interest in the study and to overcome potential concerns about confidentiality of the client data base. Questionnaires were sent to both Computerized Accounting System users and non- Computerized Accounting System users. It is important to include non-Computerized Accounting System users as it has been pointed out that one should not leave out the “zero value” or control group when exploring the effects of an intervention (Attewell, 1989). The intervention being controlled for in this case, the Computerized Accounting System, is consistent with Moore’s approach. However, it is difficult to control for the intervention of the public accountant by acquiring data from firms who do not use accountants for any reason. As discussed earlier, these firms may not exist or would be extremely difficult to locate. Due to these limitations, any results obtained for the validation of Moores’ model can only be generalized to firms that use Public Accountants. This limitation to the scope of generalizability is not severe, as it has been previously mentioned that most firms use Public Accountants. An initial sample size goal of 200-3 00 responses was set in order to accommodate the objective of testing Moore’s Diffusion of Information Technology model using LISREL. However, as stated earlier, several unexpected problems arose that dramatically reduced the number of questionnaires that could be expected to be returned. As a result of these data collection problems it was decided to use PLS instead of LISREL as PLS is widely considered an acceptable alternative to LISREL (Barclay et al, 1991). 6.3 CLIENT FIRM’S SURVEY The results from the full study indicated general reliability support for the scales used to describe the variables in Moore’s Diffusion of Information Technology model. All of the Perceived Characteristics of Innovating variables had reliability scores at the .70 level and higher except for Result Demonstrability, which dropped from a reliability score of .62 in the pilot study (Table 1) to a reliability score of .43 in the actual survey (Table 2), and Voluntariness which scored .69 (Table 2) dropping from .74 (Table 1). Except for Result 39 Demonstrability, the reliability results (Table 3) are comparable to those obtained by Moore (Table 4). The reliability scores would likely increase with a higher response rate. 6.3.1 RESULT DEMONSTRABILITY Moore used the following definition for Result Demonstrability, proposed by Zaitman: “The more amenable to demonstration the innovation is, the more visible its advantages are, the more likely it is to be adopted” (Moore, 1989, pp. 110). The reliability score of .43 (Table 2) for Result Demonstrability is considerably lower than Moore’s result of .79 (Table 4). This result could reflect a difference between sample domains (firm size) or result from use of a subset (3 questions) of the 4 questions used to originally define this Perceived Characteristics of Innovating variable. A closer look at the responses of non-Computerized Accounting System users indicates that the majority of these individuals perform non-accounting tasks. This visual analysis is substantiated by Mann- Whitney tests, which confirm that there is a statistical difference between Computerized Accounting System users and non-users for Result Demonstrability (discussed later in this chapter, also see Table 7(b)). Additional reliability figures were obtained by obtaining a breakdown between Computerized Accounting System users and non-users. Result Demonstrability reliability improves to . 71 (Table 3) when Computerized Accounting System user data only is used. A graph of Result Demonstrability non-Computerized Accounting System users was generated to determine why no reliability figure could be calculated for this variable. Inspection of this graph (Figure 5) shows that the three scales used to measure Result Demonstrability (U 15, U23, and U33) received very inconsistent responses. Normally, a graph with scales that are highly reliable would have the scores for each scale moving in the same direction for each response. The graph in Figure 5 shows that the scores for each scale move in opposite directions for each response, in most cases. Further inspection of Table 3 indicates that no other variable showed an obvious similar variability in responses by non-Computerized Accounting System users, although Voluntariness (alpha=.43) indicated that non-users did appear to have some difficulty with 40 this measure also. It is not clear why non-users would record responses that were as inconsistent as those observed for Result Demonstrability (and possibly Voluntariness). 6.4 CONDITIOMNG THE DATA 6.4.1 GENERAL Before the data could be analysed, several steps were required to ensure that the results would be meaningful. These include checking the data for accuracy, dealing with missing data, and dealing with outliers. These are discussed below. 6.4.2 ACCURACY OF INPUT DATA The data was originally input into a spreadsheet program by the researcher, who then rechecked large sections of each questionnaire. A printed copy of the input was then compared to the original questionnaire by two independent persons (the data checkers). Differences between the two were noted by each data checker on the print-out. The researcher then compared the items identified as being incorrect to the corresponding questionnaire and made appropriate corrections to the spreadsheet. Very few errors were detected by the data checkers. With a relatively small sample it is unlikely that there would be many undetected errors. After these error checking steps the accuracy of the data was considered to be very high. 6.4.3 MISSING DATA Due to the variety in the types of questions, it was not possible to adopt one global approach in treating the data for missing values. Questionnaires that were missing data for large sections of the questionnaire were not used at all (there were 2 of these). Multi-item scales, such as those used to define a Perceived Characteristics of Innovating variable, would have the scale mean inserted if only one item was missing, otherwise the item was coded as missing. 41 6.4.4 OUTLIERS AND SKEWNESS Typical regression analysis assumes normal distribution of the data. Outliers (data with extreme values) can unduly influence regression results due to their effect on the regression equation. The regression equations of interest in this study are those relating to the Diffusion of Information Technology model. Data relating to Perceived Characteristics of Innovating variables and Subjective Norm variables were reviewed for obvious nonsensical responses. One questionnaire was rejected as all Perceived Characteristics of Innovating questions were marked neutral (4 on a 7 point interval scale) indicating the respondent had not taken time to understand or read the questions. Descriptive statistics were also reviewed to determine if there were any other cases of outliers. Except for the non-usei responses to Result Demonstrability (discussed in a previous section), no others were found. A search for skewness is usually done to determine if the data distribution is normal as well as whether there may be more outliers. Moore found that his data was generally skewed but that transformations were not practical due to the design of the questionnaire (Moore, 1989). Transforming his data did not provide results different from the original data (Moore, 1989). Given the small sample size and the relatively large impact removing any questionnaires could have, whether they were outliers or not, it was determined that there would be little benefit in performing skewness tests. It should be noted that normal distribution is an underlying assumption of regression analysis and for LISREL. However, PLS does not assume data is multivariate normal (Barclay etal, 1991). 6.4.5 NON-LINEARITY ANI HOMOSCEDASTICITY An examination of scattergrams is used to reveal if the relationship between two variables show linearity (straight line) and homoscedasticity (variability in scores are approximately equivalent for all values of the two variables). Both of these, revealed by the presence of an oval shaped scattergram, are required assumptions for multivariate regression. 42 Scattergrams were produced for the variables of interest and no significant violations of these two assumptions were detected. Thus the data appeared to be of good quality for further analysis. The accuracy was found to be high and missing data was minimal. SECTION B: DESCRIPTIVE STATISTICS 6.5 GENERAL As well as the demographic data generated (Table 5), various descriptive statistics were generated for the research variables including the mean, standard deviation, and maximum and minimum reported values. These are summarized in Table 6. A comparison of Computerized Accounting System users vs. non-Computerized Accounting System users is provided in Table 7(a) and 7(b), including Mann-Whitney U test results. These results will be discussed in detail later in this chapter. The Mann-Whitney U test is used to determine if there are differences between Computerized Accounting System users and non-Computerized Accounting System users. The Mann-Whitney (M-W) test is used in order to avoid relying upon the t-test and its assumptions (normal distribution). The M-W test is a conservative test. This test was also used by Moore as part of his data analysis. Regression analysis results on the variables of interest are provided in Table 8 through Table 11. Regression results are discussed in the following sections. General comparisons will be made to Moore’s study, based on whether the results support the hypothesis that is being tested. Specific numerical comparisons will be made to Moore’s study where the results from this study differ from Moore’s. A summary of results from hypothesis testing for this study can be found in Table 13(a), and for Moore’s study in Table 13(b). References to question numbers will refer to the final questionnaire (Appendix Il-Al) unless otherwise noted. 43 6.6 DEMOGRAPHICS Demographic data is summarized in Table 5 with Adjusted Frequency figures used (these are corrected for missing data). The general categories reported on include Department of Employment; Organization Level; Education; Age; and Sex. Where relevant, comparisons are made to Moore’s survey. The main focus of the data gathering effort was the accounting/finance function. A total of 51.5% of respondents were engaged full time in the accounting area. The remaining 48.5% were distributed throughout other departments, including Administration (19%) and Other (29%) - “Other” consisted of areas not falling into Accounting or Administration. In many small and medium size firms, the concept of “department” is not well developed, resulting in difficulty classifying many of the respondents. Over 54% of respondents operated at Supervisory or higher levels of management. Another 23% were from specialized positions (Professional/Technical). The remaining 23% of respondents were at the clerical level. This response rate seems to indicate that the targeted individuals in the client firms were reached. There is a surprisingly high level of respondents that did not obtain education beyond high school (18%) while another 10% received some training from a trade school. The remaining 72% received some College/University education, including 8% with Postgraduate degrees. [From Moore: High School= 12%; Trade S chool= 19%; College/University69%; Postgraduate 18%]. It appears that for small/medium sized firms that the level of education is not as important as it is for larger firms. The majority of respondents are under 30 (39%), followed by 30-39 (36%); 40-49 (19%); and 50+ (6%). These differ from Moore’s considerably [Under 30=16%; 30-39=45%; 40-49=27%; 50+=12%]. It would be dangerous to generalize to any large extent as the methods of gathering the above information differ and may cause the perceived differences noted (ie. Moore had respondents gathered into a room to fill out the questionnaire, some potential respondents may have had to stay behind to “run the shop” and these may have been the younger employees). However, there appear to be definite differences in the age groups of 44 employees working in smaller firms. The SEX profile is also in sharp contrast to Moores study. This study had 33% male respondents and 67% female, while Moore had 63% males and 37% females respond. Again, definite differences in smaller firms. The smaller sample size in this study may contribute to some of this difference. The overall demographic profile of this study indicates sharp differences from Moor&s survey. Respondents are generally younger, more likely to be female, and have less formal education than in larger firms. These findings generally support earlier studies (discussed in Chapter 2) on demographic characteristics of people employed in small to medium sized firms. 6.7 ATTITUDE TOWARDS INNOVATTNG The dependent variable Attitude was generated from a four item semantic differential scale (good-bad; harmful-beneficial; wise-foolish; and negative-positive) in response to the question Overall, my using a C’AS in my job is (B-i). Various descriptive statistics were gathered on Attitude. These statistics are based on all 75 questionnaires. On a seven point scale (lmost positive, 7=most pessimistic) an overall average of 2.2 (Table 6(a)) indicates that attitudes are generally quite positive towards the Computerized Accounting System. Results of M-W tests on Attitude were also generated (Table 7(b)) and show that there is a significant difference (M-W = -3.7, p=.000) between Computerized Accounting System users (mean 1.8) and non-users (mean = 3.3). These M-W results provide a method of determining to what extent the overall mean of 2.2 is influenced by users and non-users. The descriptive statistics results in general, and M-W results for users specifically, indicate support for Hj [One attitude towards i/sing Computerized Accounting Systems will influence one’s innovativeness with respect to Computerized Accounting System usagej. The claim for support of Hj is based on the assertion that the more positive the attitude the more a Computerized Accounting System will be used. Since users (ie. people who are innovative) have more positive attitudes than non-users (ie. non-innovative people), the hypothesis is supported. This is similar to Moore’s findings. 45 6.8 PERCEIVED CHARACTERISTICS OF INNOVAHNG Perceived Characteristics of Innovating scales were recorded so that higher numbers reflected a higher degree of agreement with the perception associated with that variable. All of the Perceived Characteristics of Innovating variables except for Voluntariness (3.1) had a mean score of 4 (neutral) or higher (Table 6(a)). The most positive Perceived Characteristics of Innovating variables are Relative Advantage (5.50), Compatibility (5.43), and Result Denionstrability (5.19). Based on the M-W test, all of the Perceived Characteristics of Innovating variables are significantly different between Computerized Accounting System users and non-users at the p<.O5 level (Table 7(b)) except Ease of Use (.14). All of Moore’s Perceived Characteristics of Innovating variables were significantly different at p<.O5. The uniformity of scores for the variable Ease of Use amongst all small/medium firm respondents may be a result of the closer working relationship amongst users and non-users contributing to common opinions about Computerized Accounting Systems. There is support for H2 LfRelative Advantage will have a contribution more than any other Perceived Characteristics of Innovating on one’s attitude towards adopting Computerized Accounting SystemsJ based on the Mean Scores descriptives (Table 6(a)) and the M-W scores (USERs column, Table (7(b)) As Result Demonstrability (-5.59) and Compatibility (-4.86) have higher U-test z-scores than Relative Advantage (-4.81) (Table 7(b)) only moderate support to H2 is provided as Relative Advantage is expected to have the highest z-score. These findings are the same as Moore’s. Voluntariness is measured on a ranking scale (from 1 to 7) in a manner similar to the method used for Perceived Characteristics of Innovating, with higher scores indicating a more positive response. As discussed previously in this section, Voluntariness had a mean score of 3. 13 (Table 6(a)), which indicates a more unfavorable (negative) response than the seven Perceived characteristics of Innovating variables. The M-W test (Table 7(b)) shows that Computerized Accounting System users means (2.72) are significantly lower than non- Computerized Accounting System users means (4.13), indicating support for H6 46 [Voluntariness is negatively related to one’s innovativeness with respect to Computerized Accounting System usagej. This finding is the same as Moore’s. 6.9 SUBJECTIVE NORMS Values for Subjective Norm scores were calculated by multiplying the Normative Belief (ranging from 1 to 7) by the Motivation to Comply (ranging from -3 to +3). The range of scores could vary from -21 to +21. The mean scores reported in this study (Table 6(a)) are mixed and range near zero which is neutral (-2.8 to 1.5). Moore’s ranged from 1.7 to 7.3. Based on the M-W tests (Table 7(b)), the only significant differences between Computerized Accounting System users and non-Computerized Accounting System users, at p.O5, arise from Senior Management (.0i9) and Subordinates (.003). In general, H4 [The Subjective Norm it//I injinence oiies innovativeness with respect to Coniputerized Accounting System usage] is not supported. This differs from Moore’s study where H4 was supported (all of Moore’s Subjective Norm variables showed significant differences between users and non- users). These M-W results are quite different from Moore’s and again seem to indicate differences between large and small firms. In smaller offices, employees are more likely to interact with people in other functional areas (cross-pollination of ideas) and the influence of reference groups would be more uniform. Large firms would likely have less uniform opinions about reference groups due to the lack of interaction with people in other functional areas. 6.10 INNOVATIVENESS MEASURES Innoi.’ativeness was previously discussed in chapter 5. The item usage, the surrogate for adoption, was measured in four different ways: months since first use of Computerized Accounting System, hours of use, frequency of use, and number of functions used. Descriptive statistics for each of these measures can be found in Table 6(b). Because Innovativeness information was only gathered for Computerized Accounting System users, M-W tests could not be run on Innovativeness variables and N’A appears for the boxes where statistics are not applicable in Table 7(b). As a result of this data gathering approach, mean scores reported in 47 Table 6(b) and Table 7(b) for all Innovativeness variables are identical. The Innovative measure Months elapsed since Computerized Accounting System adoption was calculated by taking the average of the two measures time offirst CAS use (B 3) and (‘AS use by function, in months (B-6(b)). An average of 56 months (Table 7(b)) was calculated. This is approximately 4 3/4 years and is higher than Moore’s 40.3 month average (3 1/3 years). The reasons for this difference are not readily apparent. Traditionally it has been held that larger firms adopt information technology before smaller firms. Perhaps the particular information technology of interest, Computerized Accounting Systems, diffuse earlier than the other Personal Work Station items that Moore examined. It should be noted that no statistical tests were done to determine if the values for both studies were significantly different. If such tests were run it is possible that they could show no statistical difference in adoption periods between the two studies. Hours of use of Computerized Accounting System per week is calculated by using a single question which asked how many hours per week each (‘AS/unction was used (B-6(a)). The hours for each application were summed for each Computerized Accounting System user and an overall average was calculated from the total hours calculated, for all Computerized Accounting System users. An average of 21.6 hours per week (Table 7(b)) is more than the 15.9 hours reported by Moore. This average indicates that accounting/finance employees spend a good deal of their time with Computerized Accounting System. No statistical tests were performed to determine if the values for both studies were significantly different. Frequency qf Computerized Accounting System use is calculated in two ways. In the first method, a general frequency of use is calculated by taking an average of the results for the two questions which ask how long the GAS user has been using the CAS (B-4 and B-9) as these two items ask the same question. Both items consist of a seven point scale, and an average of 6 (about once per day, Table 7(b)) was calculated. In the second method, frequency values for a question that asked for frequency of use by function (B-5), were obtained by summing the coded values from a seven point scale (1=not at all, 4about once per week, 7=more than once per day), for each of the eight applications. Ranges of values for 48 an individual Computerized Accounting System user could vary from 53 (didn’t use any Computerized Accounting System applications) to 56 (used all eight applications more than once per day). An average value of 27.8 functions was calculated using the second method. This was lower than Moore’s result of 35; however Moore’s Personal Work Station listed 12 functions to the 8 Computerized Accounting System items identified in this study. As noted in the footnote, the average value reported in this study may be understated as well. No statistical tests were performed to determine if the values for both studies were significantly different. The number qf functions used is calculated by averaging the responses to the questions asking the frequency (?f use byjiinction (B-5), how many hours per week each CAS function was used (B-6(a)), and how many months each CAS function ivas used (B-6(b)). Theoretically, if one of these questions received an answer then all three questions should have had an answer. Each question was coded a zero (0) for no response or a one (1) for a response. By averaging the responses to each function for the three questions, effects from missing data was likely to be minimized. An average of 4.5 Computerized Accounting System functions (Table 7(b)) are used, compared to 5.9 Personal Work Station functions for Moore. There is a higher proportion of Computerized Accounting System functions used (4.5/8) than Personal Work Station functions used (5.9/12). It is not clear if this difference is due to the selection of functions. The Computerized Accounting System functions are basically a subset of the Personal Work Station functions and the most popular functions may have been chosen. Alternatively, the nature of the task, accounting, may contribute to heavier use of information technology. No statistical tests were performed to determine if the values for both studies were significantly different. There were no specific hypotheses associated with Innovativeness. 3Respondents tended to only fill in part of the question. creating a potential ‘missing data” problem. The method chosen to record the responses resulted in a “Minimum Score” of 4 instead of the theoretical 8 discussed. This approach may result in understated Frequency of use results. 49 6.11 COMPUTERIZED ACCOUNTING SYSTEM SUPPORT This area was not explored by Moore, therefore no comparative statistics are available. The statistics in Table 7(b) are for Computerized Accounting System users only, as no data is available from non-Computerized Accounting System users, therefore N/A appears in the other columns. Mean scores tabulated in Table 6(b) are identical to those in Table 7(b), as no data from non-Computerized Accounting System users was gathered for Computerized Accounting Sysieni Support. Scores are tabulated on a seven point scale (1not at all, 4=once per week, 7=more than once per day). All four sources of Computerized Accounting System Support range from 2.2 to 2.6 (2=zless than once per month, 3=about 1-3 times per month). This suggests that there is generally not very much reliance on the Support Group. Also, no one group appears to be more dominant than any other. These results do not support the following hypothesis: H8 [The involvement ofa Support Group will contribute to a successful adoption of a Computerized Accounting SystemJ, H9 [The involvement of a Friend will contribute to a successful Computerized Accounting SystemJ, H10 [The involvement of other Employees will contribute to a .s’uccessfui Computerized Accounting SystemJ, Hjj [The involvement of an external Accountant will contribute to a successfiui Computerized Accounting SystemJ; and Hp [The involvement ofan external Consultant will contribute to a successful Computerized Accounting SysteniJ. SECTION C: REGRESSION ANALYSIS 6.12 GENERAL Moore’s research hypotheses and Diffusion of Information Technology model, as well as Computerized Accounting System Support, were tested using multiple regression and PLS. This was done by examining the effects of the different independent variables (Perceived Characteristics of Innovating, Voluntariness, Attitude, Subjective Norm, and Computerized Accounting System Support) on each of the Innovativeness measures. The results are discussed in this section. PLS results are discussed in the next section. 50 6.13 THE EFFECT OF PERCEIVED CHARACTERISTICS OF INNOVATIVNESS AND VOLUNTARINESS ON ATTITUIE The initial regression model analysed was the seven Perceived Characteristics of Innovating variables and Voluntariness on Attitude. The procedure followed paralleled that of Moore (1989). A STEPWISE regression was run, with the F-value probability set at p<.O5 for entry and p>. 10 for removal of a variable once in the equation. Following this regression, a second regression was run where all variables were forced into the equation in the same order as the STEPWISE regression. The end result of the forced entry procedure is to produce a regression with all variables in the equation, but the stepped entry allows the direct contribution of each variable to R2 to be examined. The regression results on the full equation provide an R2.776 and an adjusted R2=.749 (Table 8), indicating that the Perceived Characteristics of Innovating variables are significant in the formation on Attitude towards using computerized accounting systems.. The regression results indicate that the various Perceived Characteristics of Innovating variables have different effects on Attitude. Only Re/alive Advantage is highly significant (p.OO, R2=.73). Visibility (p=.O4, incremental R2=03 [=76-73]) is marginally significant. None of the other variables contribute to R2 in any meaningful way. These results are summarized in Table 8 where part I lists the results for the forced step regression and part II lists the results for the regression on the full equation. [Moore’s study had an adjustedR2.677, and more of the Perceived Characteristics of Innovating variables were significant]. The regression results support H2 [Re/alive Advantage will have a contribution more than any other Perceived characteristics of Innovation on one’s attitude trniards adopting Computerized Accounting Systems], as Relative Advantage’s contribution to R2=.73 while the full equation had anR2=.78. This result was similar to Moore’s. There was no support for H7 [Voluntariness will be negatively related to one’s attitude towards using Computerized Accounting Systems], as Voluntariness was not significant. Moore’s results showed a negative Beta for Voluntariness (which was significant), 51 and supported H7. Voluntariness results may be affected by the low reliability of non-users responses (Table 3), which was discussed in Chapter 5. A second regression, including variables for Perceived Characteristics of Innovating, Voluntariness, and Subjective Norm, was run on Attitude. The purpose of this regression was to examine the influence of Subjective Norm on Attitude. A composite score for Subjective Norm was computed by summing the scores of the individual Subjective Norms for each measure. The regression results are provided in Table 9, PCI and SN column. There is very little change from the results of the regression without Subjective Norm as Subjective Norm is not statistically significant in this regression. Thus H5 [The Subjective Norm i4ll influence one’s attitude toward adopting the Computerized Accounting Systenzj is not supported. In Moore’s study H5 was supported as Subjective Norm was significant. The difference in results between this study and Moore’s may be due to the differences in sample size as well as sample selection. The individual reference groups comprising the composite Subjective Norm are Friends, Peers, Superiors, Subordinates and Senior Management. Individual scores for each reference group could range from -21 to 21 (this was discussed in Chapter 5). The composite measure calculated for Subjective Norm will therefore be a neutral value (near zero) if the reference groups are not that important to the respondent or if the scores are extreme on either side of zero. A small sample size may not be able to differentiate between these two possible explanations. A larger sample size would indicate significance, if there was a trend in the scores in the same direction for the reference groups, or if one (or more) reference group was clearly dominant and the remaining group scores were near neutral. A larger sample size would not help if two reference groups with opposite scores were dominant. Sample selection, and possibly firm size may be a factor as larger firms tend to have established cultures and prevailing opinions on information technology use. Moore’s sample 52 consisted of individuals from six large firms, whereas this study contained responses from many more (smaller) firms, possibly 20 or more. Moore’s respondents would likely show a more dominant culture affecting Attitude as, at most, there are six different cultures and possibly less. A dominant culture could emphasize one reference group or combination over another. It would be difficult to determine if Moore’s significant results for Subjective Norm are due to the small sample size of large firms or the presence of a corporate culture that is common to large firms. The current study could have up to 20 or more cultures, which may result in no clearly dominant reference group. As no significant results for Subjective Norm were found in this study, there is a likelihood that there is no dominant corporate culture common to small businesses in general. The current regression involving Subjective Norm does not provide enough information to determine if the hypothesis being tested is being correctly rejected (or accepted). A further regression was run with Computerized Accounting System Support, Perceived Characteristics of Innovating, Vohintariness and Subjective Norm on Attitude. This equation is similar to the previous regression with the addition of Computerized Accounting System Support. The purpose of this regression was to determine the influence of Computerized Accounting System Support on Attitude in order to examine the extensions made to the Diffusion of Information Technology model. This regression was run with Computerized Accounting System Support included as a composite score (individual support groups consisted of Friend, Employee, Accountant, and Consultant). The regression results indicated that the composite Computerized Accounting System Support score was not significant in the regression equation (p=.3 1). These results are summarized on Table 9, PCL SN & SUPPORT column. H14 [The involvement of a Support Group wi/i have a positive influence on Attitude], was not supported. The regression results on Support clearly indicate that this variable has no effect on Attitude. This result is unexpected and may be an artifact of the variable Attitude (discussed below). 53 The results of the regressions on Attitude, while not as supportive of the Diffusion of Information Technology model as Moore achieved, raise the same issue - the operationalization of Attitude. Only Relative Advantage (p.OO), Visibility (p=.O4) and Ease of Use (p. 10) at p.lO were statistically significant (Table 8). Relative Advantage by itself provides an R2 of .73 while the full equation has an R2 of .78. The finding that Attitude captures Relative Advantage and not the Perceived Characteristics of Innovating variables in general, seems convincing and supports Moore’s conclusion that this appears to be the case (Moore, 1989). The unexpected lack of significance of Support on Attitude may be a result of the poor operationalization of Attitude. 6.14 THE EFFECT OF ATTITUDE, SUBJECTIVE NORM, PERCEIVED CHARACTERISTICS OF INNOVATIVENESS, VOLUNTARINESS AND SUPPORT ON INNOVATIVENESS 6.14.1 GENERAL Regression analysis was also run on the independent variables and Innovativeness. An initial regression used Attitude, Subjective Norm, and Voluntariness as independent variables. An additional independent variable, Support, was added in a subsequent regression run to examine the influence of this variable on Innovativeness. A different regression on individual Perceived characteristics of Innovating variables and Subjective Norm variables was also run. This second regression omitted the intervening variables AttItude and the overall Subjective Nonii measure. Once again Support was added in a subsequent run to measure its impact on Innovativeness. 6.14.2 ATTITUDE, SUBJECTIVE NORM AND VOLUNTARINESS ON INNOVATIVENES S Four regressions were run, one for each of the dependent Innovativeness measures (Number of Functions Used, Frequency of Use, Months Since Adopted, and Hours of Use Per Week). The dependent variables for each regression run were Attitude, Subjective Norm 54 and Voluntariness. The results of the regressions, summarized in Table 10(a), will be discussed in the following paragraphs. The regressions were done with all independent variables and each dependent variable entering the equation at once. Subjective Norm is not significant in any of the regressions. H4 [The Subjective Norm will influence one’s innovaliveness i’ith respect to Computerized Accounting System usage], is not supported. This is consistent with Moore’s findings. Voluntariness is not significant in any of the regressions, thus rejecting H7 [Voluntariness will be negatively related to one’s attitude towards using Computerized Accounting System]. Moore’s study supported H7 for all Innovative measures. Attitude is significant for all Innovativeness variables, although the betas are all negative, rejecting Hj [One’s attitude towards using Computerized Accounting Systems will influence one’s innovativeness with respect to Computerized Accounting System usage]. Moore found that Attitude was significant for all Innovativeness variables and all had positive betas, supporting H]. As Voluntariness was not significant for any of the Innovativeness variables, H6 [Voluntariness is negatively related to one’s innovaliveness i’ith respect to Computerized Accounting System usagej, is not supported. Moore’s results supported H6. The adjusted R2 values for the four regression equations range from a low of .193 to a high of .223. The variations in R2 indicate that the independent variables capture different degrees of the variance in the different forms of innovativeness. This low range of adjusted R2 values indicate that these independent variables do not explain Innovativeness very well. These results are fairly comparable to Moore’s findings. An additional series of four regressions were run including Computerized Accounting System Support as an independent variable in the aboye regression equation. The results from these regressions are summarized in Table 10(b). The composite Computerized Accounting System Support measure was used in this regression. Computerized Accounting System Support was significant for all Innovativeness variables. This result supports H8 [The involvement of a Support Group will contribute to a successful adoption of a Computerized Accounting Systenij. With the addition of Computerized Accounting System Support, the influence of the other independent variables on the various Innovativeness variables changed. With the addition of the Support Group variable (compared to the regressions without the Support Group variable), Voluntariness became significant for Hours (p=.O5 vs p=. 14), while Attitude became less significant for all Innovativeness variables [Functions (p=.O4 vs p.OO); Frequency (p=.O5 vs p=.OO); Months (p=.O9O vs p=.O4O); Hours (p=.l 1 vs p.OO)]. Thus, the inclusion of Computerized Accounting System Support has weakened the importance of Attitude on Innovativeness (ie. weaker support for Hj). This effect indicates that Computerized Accounting System Support has an influence in the Diffusion of Information Technology model. 6.14.3 PERCEIVED CHARACTERISTICS OF INNOVATING, SUBJECTIVE NORM, ANT) VOLUNTARINESS ON INNOVATIVENESS Individual Perceived Characteristics of Innovating variables and individual Subjective Norm variables were regressed on the dependent Innovativeness variables in order to measure the magnitude of their direct effects on adoptive behaviour. As each individual Perceived Characteristics of Innovating and Subjective Norm measure were expected to have different influences on the dependent variables, STEPWISE regression was used with the probability to enter a variable into the equation set at p.O5. As in the previous regression runs, Computerized Accounting System Support was added as an independent variable to subsequent regression runs. In this case, the individual Support Group measures were used instead of the composite scale in order to examine the influence each scale has on Innovativeness. Regression statistics without Support Group are included in Table 11 (a) and with Support Group in Table 11(b). For the regression without Support Group, the Subjective Norm variables were significant only for the Innovativeness variable Number Of Functions Used (Peers p=.OO, Subordinates p=.OO) and weakly significant for Frequency of Use (Subordinates p.O7). The 56 overall Subjective Norm results weakly support Hq [The Subjective Norm wi/i influence one’s innovativeness i’ith respect to Computerized Accounting System usagej. This result is similar to Moore’s where Subjective Norm variables found to be significant were Functions Used and Frequency. Result Demonstrability is significant for all Innovativeness measures (Functions p=.OO; Frequency p=.O1, Months p=.OO; Hours p=’.O3). Moore had similar results. Voluntariness is only weakly significant for Months of Use (p=.O7), thus not supporting H6 [Voluntariness is negatively related to one’s innovativeness with respect to Computerized Accounting System usagej. Moore’s results supported H6 for all Innovativeness measures. Relative Advantage was significant for Functions (p=.O5), Frequency (p’.OO) and weakly significant for Months (p=.O6). Moore’s results only found Months significant. The addition of individual Group Support variables to the regression equations, discussed in the preceeding paragraphs, indicated that these variables did have an affect on Innovativeness. The regression results are summarized in Table 11(b). Consultant was significant for all Innovativeness variables, [Functions (p=.OO), Frequency (p=.OO), Months (p=.O4) and Hours (p’.OO)] with all betas positive. This result supports H12 [The involvement of an external Consultant will contribute to a successful Computerized Accounting Systemj. Personnel was significant for Functions (p=.OO) and Frequency (p.OO), moderately supporting H10 [I he i ivoivement of other Employees till contribute to a successful Computerized Accounting SystemJ. Accountant was not significant, rejecting Hjj [The involvement of an external Accountant will contribute to a successful Computerized Accounting Systenij. Friend was not significant, rejecting H9 [The invoii.’ement of a Friend will contribute to a successful Computerized Accounting SystemJ. 57 The overall regression results for the equations including Support Group variables provide support for H8 [The involvement of a Support Group vili contribute to a successful adoption of Computerized Accounting SystemJ. Voluntariness became more significant for Hours (p=.Ol) than without the presence of Support Group while Visibility became not significant for Hours. There were less Weakly Signficant variables once Support Group was added to the regression equation. 6.14.4 OTHER REGRESSIONS A series of regressions of Support Group, the independent variable on Subjective Norm, Attitude and Perceived Characteristics of Innovating, as dependent variables, were run. None of these regressions were done by Moore, therefore no comparative statistics are available. The results from these regressions are summarized in Table 12(a) and Table 12(b). The regression Support Group on Subjective Norm was suggested by the Fishbein & Azjen model. One of the links in this model is between Communications Network (Support Group in this study) and Subjective Norm (see Figure 3). The individual variables of Support Group were regressed against the composite Subjective Norm. Regression results indicated that the individual Support Group variables were not significant and H13 [The involvement of a Support Group will have a positive influence on Subjective Norinj was not supported. The regression Support Group on the individual Subjective Norm variables was then run to see if the composite Subjective Norm was masking its individual components. While Support Group Personnel was the only significant variable, the low adjusted R2’s and F’s indicate that the relationship between Support Group and Subjective Norm is weak at best. The regression Support Group on Perceived Characteristics of Innovating was suggested by both the Fishbein & Ajzen model and Moore’s model. The Fishbein & Azjen model indicate a link between Communications Network and Attitude Towards Adopting 58 (Figure 3). Moore’s model indicates that Attitude Towards Adopting is determined by the Perceived Characteristics of Innovating variables (Figure 1 ).4 It seems logical to regress Communications Network (Support Group) on the individual Perceived Characteristics of Innovating variables if they have become a surrogate for Attitude. Regression of the individual Support Group variables were run against the Perceived Characteristics ofInnovating (individual) dependent variables. The Support Group variable Accountant was significant for Compatibility (p=.O1), Ease of Use (p=.O3), Relative Advantage (p=.OO) and Result Demonstrability (p=.OO) The Support Group variable Consultant was significant for Visibility (p=.OO). The Support Group variable Friend was significant for Image (p.OO) and Trialability (p.O2) The Support Group variable Personnel was significant for Compatibility (p=.O2). These results provide general support for H15 [The im.’olvement of a Support Group i’ili have a positive influence on Perceived Characteristics ofInnovation variablesJ. A regression of Support Group on the individual Innovativeness variables was run without the presence of other independent variables in order to determine how large an effect the individual Support Group variables have on Innovativeness. The results are summarized in Table 12(b). Consultant and Personnel were both significant for all Innovativeness varibles. The adjusted R2 values and F values are generally high, indicating that Support Group has a significant effect on Innovativeness. The regression results support H8 [The involvement of a Support Group will contribute to a success!;’,! adoption of computerized accounting systemJ, H10 [The involvement of other Employees will contribute to a successfiui computerized accounting systemJ and Hp [The im.’olvement of an external Consultant u ‘ill contribute to a successful computerized accounting systemj. These results were consistent to the earlier regression results reported (Table 11(b)). 4As reported in a previous section of this paper. regression results indicated that some of the Perceived Characteristics of Innovating (PCI) variables better represented the concept “Attitude” than did the composite scales designed to represent “Attitude”. When Moore used LISREL to test his model, he used these PCI variables to represent the concept “Attitude” instead of the original Attitude scales. 59 A regression of Support Group on Voluntariness was run to determine if there was any effect between these variables. The results are summarized in Table 12(b). Personnel (p:=.O2) was the only significant Support Group variable, however the R2 and F values were fairly low. The rationale for running this regression was to see if the perception of Voluntariness was influenced by the Support Group. The results do not support this. SECTION D: PATH MODELING Since there were some problems with simple regression analysis that caused difficulty interpreting results for the Di/flision of Information Technology model for both studies the Attitude variables generated may not have captured the construct Attitude. It also appeared that the construct Subjective Norm (Subjective Norm) may not have been appropriately specified, because the regression analysis indicated that individual Subjective Norm variables accounted for more variance on the dependent variables than did Subjective Norm itself Additionally, the construct Innovativeness could not be generated using normal regression techniques due to the differences in the scales of the Innovativeness variables (eg. months vs hours vs functions used). These factors indicated that an alternate method to regression analysis would assist in constructing Subjective Norm, Attitude and Innovativeness from their individual components. One such method is known alternatively as causal modeling (Barclay et a!, 1991; Bagozzi, 1982), structural equation modeling (Fournell et a!, 1982) or path modeling (Wold, 1985). For convenience the term path modeling will be used throughout this paper. 60 Path modeling utilizes second generation5multivariate analysis techniques in order to obtain statistical information that cannot generally be obtained by first generation statistical techniques (Barclay et al, 1991; Dimnik, 1986). Path modeling is a method of research, and can be used to determine internal consistency, reliability, construct validity, and for hypothesis testing (Bagozzi, 1982). Borrowed from econometrics (path models and manifest variables) and psychometrics (latent variables) (Wold, 1985), all path models have in common the traits of latent variables linked to manifest variables by paths. 6.15 CHOICE OF PATH MODEL COMPUTER IMPLEMENTATION - LISREL vsPLS The two most common computer implementations of path modeling are LIS]?EL and Partial Least Squares (PLSI. Both of these programs have their strengths and weaknesses for analysing models. The choice of which program to use depends on the stage of theory development being tested and the goals of the researcher. PLS is generally used in the early stages of theory development while LISI?EL is better suited to models based on well developed theory. LISREL is based on assumptions of multivariate normality in data whereas PLS requires no such assumptions. LISREL requires large sample sizes while PLS can be used with much smaller sample sizes (Barclay et al, 1991). Afier analysing the various characteristics of this study it was determined that the use of PLS would be most appropriate. This study is examining the Di/,/iision of Information Technology model first developed by Moore. While the theories underlying this model have become established in other fields, the synthesis of Rogers and Fishbein & Ajzens theories has not been tried before. The applicability of this approach has yet to be firmly established. Additionally, preliminary results from regression analysis indicated that multivariate normality 5The term “second generation” is used to denote the use of more sophisticated mathematical models and statistical computer programs. A second generation multivariate technique must meet four requirements (Fournell, C., A Second Generation of Multivariate Analysis Methods, 1982. cited in Barclay et. a!.. 1991): the technique must 1) analyze multiple criterion and predictor variables: 2) analyze unobservable theoretical variables; 3) analyze errors in measurement: and 4) be applicable in a confirmatory (ie. hypothesis testing) context. Additionally. “first generation multivariate analysis procedures are special cases of second generation techniques. Multiple regression. multiple discriminant analysis. analysis of variance and covariance, and principal components analysis are all special cases of canonical correlation ... which itself is a special case of PLS ...“ (Barclay et. al. 1991, pg. 4). 61 assumptions may not apply to the data in this sample. Finally, a relatively small sample size was obtained which indicated that the use of LISREL would not be feasible. 6.15.1 DESIGN OF PLS PATH MODEL PLS path models are comprised of manifest variables (indicators) and latent variables (constructs). The arrangement of manifest variables (MVs) and latent variables (LVs) determine the framework for the PLS model. Each PLS model has two sets of equations: one equation describing the path (links) of each MV to each LV (outer design matrix); the second equation describing the path connecting LVs to each other (inner design matrix). The Diffusion of Information Technology model, with iWVs’ and LVs identified, can be seen in Figure 6. This model design is comparable with the LISREL model used by Moore (see Figure 3). The outer design matrix for the Diffusion of Information Technology model is illustrated in Figure 6. Latent Variables are linked to Manifest Variables by paths. Paths can flow in either direction (indicated by arrows), depending on the underlying theory supporting the model. (1) LV Subjective Norm has as its indicators MVs Supervisors, Peers, Senior Ivianagement, Subordinates, Friends and Perceived Voluntariness. The MV, Friends, which was omitted from the Moore’s LISREL analysis, was included in the PLS analysis in order to fully analyse the Df/iision (?f Information Technology model The MV indicator Perceived Vohintariness has been included because the regression results indicated a strong interaction with Subjective Norm. The inclusion of this All7 is consistent with the approach used by Moore. (2) LV Voluntariness has as its indicator All’ Perceived Voluntariness. (3) LVAttitude has as its indicator AlV, Relative Advantage, Image, Compatibility, Ease of Use, Trialability, Visibility, and Result Demonstrability. These MVs are actually Perceived Characteristics ofInnovating indicators and not the original indicators derived 62 for Attitude. As discussed in the regression analysis section, the Perceived Characteristics of innovating indicators had a greater direct effect on Innovativeness than Attitude alone, indicating that Attitude had not been operationalized appropriately. As PLS uses the related MVs to synthesize the LVs it was decided that LV Attitude could be derived from the Perceived Characteristics ofInnovating indicators. (4) LV Innovativeness has as its indicators MVs Hours, Months, and Functions. One indicator, Frequency of Use, was not included in the model due to the difficulty in interpreting the composite scale. Although Moore’s analysis indicated that MV Functions and Frequency “may be tapping the same dimension of Innovativeness” (Moore, 1989, pg. 186), results from this study indicate that they may actually be tapping different dimensions (see Table 11(a) and Table 11(b), noting the differences between these variables regression results). The inner design matrix has the path structure indicated in Figure 6. Path links in the inner design are from one Latent Variable to another. When using standardized scales, path loadings represent correlations (Barclay, 1991). The interpretation of non-standardized loadings are different and depend on the underlying premise of the model as set out by the model builder (Lohmoller, 1984). In the current model the scales are standardized and path loadings between LVc and MVs’ represent the relative importance of the composite scale score to the LV. The path loadings between LVs can also range from zero to one. The higher the loading the more important the relationship/link. Loadings greater than .3 are considered to be acceptable (Chin, 1992). 6.15.2 ANALYSIS OF SAMPLE SIZE REQUIREMENTS With PLS, a more modest sample size than with LISREL is used, because the less rigorous statistical assumptions require a minimum sample size often times: (1) the number of indicators from the most complex /brmative construct; or, (2) the largest number of predictors 63 leading to an endogenous construct (Barclay et al, 1991). A forniative construct (or Latent Variable in PLS terminology) is an LV that is a construction, or composite, of its MVs (Barclay et. al., 1991; Lohmoller, 1984). [Reflective construct’s on the other hand are L Vs with lviVs that are products or reflect the underlying construct of the LV (Barclay et. al., 1991; Lohmoller,1984)]. An endogenous construct is an LV that is predicted by other LVs. The construct for an LV that is not predicted by other LVs is call an exogenous construct. “The use of small samples ... seems to violate a traditional concern with sample size versus parameters to estimate. PLS can deal with this situation because ... the iterative algorithm behind PLS estimates parameters in only small subsets of a model during any given iteration. The subset estimation process consists of simple and multiple regressions so that the sample required is that which would support the most complex multiple regression encountered.” (Barclay et al, 1991, pp. 15). The determination of whether MVs are formative or reflective in regards to their associated LV depends on the researchers prior experience with the model and the understanding of the real world situation being studied. If the constructs are not well developed then the IvIVs for that construct are considered formative. For the purposes of FLS analysis of the Dffusion of Information Technology model (Figure 6) only the Subjective Norm MVs will be treated as formative indicators, while Subjective Norm and Communication Network will be treated as formative in the extended model (Figure 7). The MV for the other LVs will be treated as reflective indicators. In Figure 6, Subjective Norm and Voluntariness are exogenous L Vs, while Attitude and Innovativeness are endogenous LVs. The largest number of formative indicators is five (Subjective Norm) while the endogenous LV with the largest number of predictor LVs is Innovativeness with three (Subjective Norm, Voluntariness, and Attitude). This would indicate a minimum sample size of 50 (10 times the 5 Subjective Norm MVs). In Figure 7, Voluntariness and Communications hannei are exogenous L Vs, while Subjective Norm, Attitude and innovativeness are endogenous L Vs. The largest number of formative indicators is five (Subjective Norm) while the endogenous LV with the largest number of predictor LVs is Innovativeness with four (Communications Channel, Subjective Norm, Voluntariness, and Attitude). This would indicate a minimum sample size of 50 (10 64 times the 5 Subjective Norm MVs). A sample size of 50 is much less than the sample size of 500-600 Moore required for his initial development of the Diffusion of Information Technology model using LISREL. The total number of usable questionnaires available for analysis is 75, which exceeds the minimal required sample size. 6.15.3 GOODNESS OF FIT DETERMTNATION The Diffusion of Inforniation Technology model illustrated in Figure 6 was assessed by comparing PLS statistical results to various reduced versions of this model. This approach was used as PLS does not have any single goodness of fit” measure. Three common diagnostics used for PLS analysis are based on root mean square (RIVIS) covariances. These are Multiple R2 (R2), Communality (H2), and Redundancy (F2). R2 is the explained variance in the endogenous constructs (LVs). H2 is the proportion of variance the MV have in common with the principal component. This is a predictability measure. Mathematically, H2 is calculated (Lobmoller, 1984): H27, = 1 SSEn / SSO,, where SSE is the sum of squared prediction errors; SSO is the sum of squared observation errors; and n is the sample size. F2 measures the average squared multiple correlation between each endogenous construct and all exogenous constructs. The redundancy is the proportion of the variance that can be predicted by the predictors of the LV This is a test of predictive relevance. Mathematically the formula is similar to Communality except that it applies to LVs whereas Communality applies to MVc (Lohmoller, 1984): F27, = 1 - SSEJ, / SSO/1 65 Generally, the fit of the inner model is satisfactory if R2 is high enough; the fit of the outer model is satisfactory if H2 is high enough; and the fit of the total model is satisfactory if F2 is high enough (Lohmoller, 1984). Determinig if there is a fit or not is clearly a judgement call. The models which were compared to the full Diffusion of Information Technology model in Figure 6 were determined by eliminating the exogenous LV Voluntariness, then both exogenous LVs (Voluntariness and Subjective Norm). Finally, a model that eliminated only low scoring Perceived Characteristics of Innovating variables was generated. The same models were run again, this time using only data points from Computerized Accounting System users. One final model was run which extended the model in Figure 6 with the added LV Communications Channel (see Figure 7). The results from these comparisons are contained in Table 14. The Diffusion of Information Technology model was run using PLS, with a number of different configurations. The full model shown in Figure 6 was run (R2=.18,H2=.53,F2=.11) and compared to a similar model minus LV Voluntariness (R2=.13, H2=.46, F2=.06); minus both Voluntariness and Subjective Norm (R2=. 14, H2=.28, F2=.04); minus four Perceived Characteristics of Innovating variables (R2=.20, H2=.56, F2=. 13) [Result Demonstrability, Ease of Use, Trialability, and Image, all of which had low individual scores on several indices (path loading values, R2,H2,F2)]. The same models were run again, this time using Computerized Accounting System user data only (53 subjects). This set of PLS runs was generated because previous SPSS analysis had indicated that there were significant differences between Computerized Accounting System users and non-users for several of the MV (refer to Table 7(b)). While the sample size of 53 was somewhat less than the rule-of-thumb requirement for a minimum sample of 70, it was expected that this would not greatly affect results. Results were not that different from the full data set for the full model (R2=. 10, H2=.42, F2.05); minus LV Voluntariness (R2=. 10, H2=.36, F2=.04); minus both Voluntariness and Subjective Norm 66 (R2=. 15, H2=.37, F2.05); minus four Perceived Characteristics of Innovating variables (R2=. 12, H=.46,F2=.06) [Result Demonstrability and Image, all of which had low individual scores on several indices (path loading values, R2,H2,F2)]. A final FLS comparison on the Diffusion of Information Technology model (Figure 6) was made to an extended model which included the LV Communications Network (Figure 7). The extended model was included as Communications Channel was shown by regression analysis to have some influence on the other L Vs. However, the results with the full data set (R2=.20, H2=.54, F2.12) were similar to the original, full model. Further analysis on individual LVs indicated that the extended model increased R2s for the LVs it loaded on (Subjective Norm = .07; Attitude = .41; and Innovativeness = .34) compared to the original model (Attitude .28; and Innovativeness = .28; not calculated for Subjective Norm). The differences in the “fit” indicators for the various models are not that large and would probably not be statistically different from the original model. The extended model may have more explanatory power than the original model due to the presence of indirect affects that Communications Channel has on the other LV. The overall small numbers for the fit indicators suggest that the increased explanatory power may also not be statistically significant. Based on these analysis, no alternative model was shown to be superior (on a qualitative basis) to the original Diffusion of Information Technology model represented in Figure 6. A model with four Perceived Characteristics of Innovating variables removed indicated higher scores on the indices examined, however there were no theoretical grounds for removing these MVs. Compared to the original model, an extended model had slightly improved direct explanatory power and additional indirect explanatory power. 6.15.4 ASSESSMENT OF HYPOTHESES TESTIIJG As no tests for statistical significance between the various models have been done it is not possible to quantitatively evaluate each hypothesis. Qualitative interpretations are possible 67 however, based on analysis of path loadings and R2 results. The first seven hypotheses described in Chapter 4 are based on the full model (Figure 6) while the remaining eight hypotheses are based on the extended model (Figure 7). The various hypotheses and conclusions are presented below: Hi to H7 are analyzed based on the original Diffusion of Information Technology model described in Figure 6 and the extended model described in Figure 7. Based on qualitative analysis the conclusions for some of these hypotheses change depending on which model is used. Hj. One’s attitude towards using Computerized Accounting System i’ill influence one’s innovativeness t’ith re5pect to Computerized Accounting Systeni usage. The hypothesis indicates that a positive coefficient is required to increase innovativeness. For the original model (.624) and the extended model (.440) the path coefficient is positive thus Hi is supported H2: Relative Advantage wi/i have a contribution more than any other Perceived Characteristics oJ Innovating on one’s attitude towards adopting ‘omputerized Accounting Systems. Path loading for MV Relative Advantage on LV Attitude is the largest for the original model (.9141) and the extended model (.9180), supporting H2. H3: C’omputer Avoidance i’iii have a contribution less than any other Perceived Characteristics of Innovating on one’s attitude towards adopting Computerized Accounting Systems. This hypothesis is not explored in this study. 114: The Subjective Norm will influence one innovativeness with respect to Computerized Accounting System usage. For the original model (-.06 1) and for the extended model (.056) the path coefficient have a very small loading value indicating no support for H4, H5: The Subjective Norm i’iii influence one’s attitude toward adopting the Computerized Accounting System. The original model (.122) and extended model (.053) have very small path coefficients, which indicate no support for H5. 68 “6• Voluntariness is negatively related to one’s innovativeness iith respect to Computerized Accounting System usage. The path coefficient of for the original model (.324) and for the extended model (.3 13) indicates rejection ofH6. H7: Voluntariness will be negatively related to one attitude towards using Computerized Accounting Systems. The path coefficient for the original model (- .469) and for the extended model (-.406) indicates that the hypothesis is supported. Hg to H15 are based on the extended Diffusion of Information Technology model (see Figure 7). ‘‘8 The involvenient of a Support Group will contribute to a successful Computerized Accounting System. A positive path coefficient of .321 (Innovativeness) indicates support for Hg. Hypothesis H9-H12 are indirectly tested as they are A.1V and contribute to the overall Support Group (Communications Channel) LV path loading. H9. The involvement ofa Friend i ill contribute to a success/lu Computerized Accounting Sstem. As the path coefficient is small (.0694), this suggests that the hypothesis is not supported. Hj . The involvement of other Employees will contribute to a successful Computerized Accounting System. As the path coefficient of Personnel is positive (.3837), this suggests that the hypothesis is weakly supported as the indirect effect (.3837*.321.123) is small. Hjj. The involvement qf an external Accountant will contribute to a successful Computerized Accounting System. As the path coefficient is small and negative (- .0209), this suggests that the hypothesis is not supported. .11J2. The involvement of an external Consultant i’iii contribute to a successful Computerized Accounting System. As the path coefficient is large (.7121), this suggests that the hypothesis is supported as the indirect effect (.7121*321=229) is moderate. 69 H13. The involvement of a Support Group wi/i have a positive influence on Subjective Norm. As the path coefficient is .27 1, this suggests that the hypothesis is supported. H14. The involvement of a Support Group will have a positive influence on Attitude. As the path coefficient is .381, this suggests that the hypothesis is supported. H15. The involvement of a Support Group will have a positive influence on Perceived characteristics of Innovating variables. This hypothesis was tested indirectly using PLS, via following the paths. The indirect score of .350 (.381*. 9180) supports the hypothesis. [These results are summarized in Table 13(a).] 6.16 SUIVIIVL4RY OF RESULTS: PATH ANALYSIS Using PLS statistical results, it was shown that a reduced alternative model with four Perceived Chaiacteiistics of Innovating variables removed, provided a marginally (possibly not statistically different) better fit indicators than the original Diffusion of Information Technology model. However, the better indicator scores did not appear to be different enough to justify adopting the reduced model. An extended Diffusion of Information Technology model, including the LV Communications Network, did not provide any better fit indicators than the original model either. Qualitative analysis of individual LV and MV indicators suggest that improved predictive power may result when using the extended model. The path loadings in both the inner and outer model change to varying degrees when LV Communication Channel is added to the original model (compare Figure 6 to Figure 7). The introduction of this LV into the model may have removed some of the “noise” from the model which may have previously skewed the loading values. As no statistical analysis has been performed on the changes in loading values, no significance is claimed for the observed minor changes in loading values. 70 SECTION E: SUMMARY OF DATA ANALYSIS 6.17 GENERAL Three different techniques were used to analyse the data. The initial analysis included a comparison of the descriptive statistics between users and non-users of computerized accounting system adopters. Next, regression analysis was performed on the data to examine the effect of various independent variables on the dependent variables. Finally, path analysis was used to examine the theoretically derived Diffusion of Information Technology model developed by Moore and compare this model to other versions of this model to determine which model had the best fit to the data. 6.18 SUMIVIARY OF DESCRIPTIVE STATISTICS While there were significant differences between Computerized Accounting System users and non-users on several of the variables, there were fairly uniform Subjective Norms by all respondents. Overall, 71% of the sample were identified as Computerized Accounting System users, which indicates that non-users have a large impact on the overall results. The average time elapsed since initial adoption is just under five years. Computerized Accounting Systems are used fairly often, with over 4 computerized accounting system functions being used for 22 hours per week. 6.19 SUMMARY OF HYPOTHESES TESTTNG Hj. One attitude towards using a Computerized Accounting 5steni will influence one innovatii’eness i i/h respect to Computerized Accounting System usage. This hypothesis was supported by descriptive statistics, not supported by regression analysis, and partially supported by PLS. Regression analysis indicated that while Attitude was significant in the adoption process for all Innovativeness variables, all of the betas were also negative. PLS results indicated that the loadings were negative in value for the standard model and positive for the extended model. The confusing Attitude results (Table 10(b)) may have been an artifact of the scales used to measure Attitude. Substituting the Perceived Characteristics ofInnovating variables for Altitude resulted in significant (positive) Perceived 71 Characteristics ofInnovating variables for all Innovativeness variables (Table 11(a)). 112: Relative Advantage will have a contribution more than any other Perceived Characteristics of Innovating on one ‘c attitude towards adopting computerized accounting system. This hypothesis was generally supported using all methods. The regression results indicated that Relative Advantage was generally the most significant Perceived Characteristics of Innovating variable. PLS analysis was mixed with the original model supporting the hypothesis and the extended model rejecting the hypothesis. H3: Computer Avoidance will have a contribution less than any other Perceived Characteristics of Innovating on one’s attitude towards adopting Computerized Accounting Systems. This hypothesis is not explored in this study. H4: The Subjective Norm i’ill influence one innovativeness with respect to Computerized Accounting System usage. This hypothesis is only supported by PLS. H5: The Subjective Norm will influence one c attitude toward adopting the Computerized Accounting System. This hypothesis is supported only at the PLS stage. 116: Voluntariness is negatively related to one’s innovativeness with respect to Computerized Accounting System usage. This hypothesis was not supported by any method. H7: Voluntariness will be negatively related to one attitude towards using Computerized Accounting Systems. This hypothesis was generally not supported, except for the extended model using PLS. 118. The involvement of a Support Group ui/i contribute to a successful Computerized Accounting System. This hypothesis was supported by regression and PLS analysis. 72 H9. The involvement ofa Friend wi/i contribute to a successful Computerized Accounting System. This hypothesis was supported by regression and PLS analysis. Hj . The involvement of other Employees ii’iii contribute to a successfid Computerized Accounting System. This hypothesis was supported using PLS analysis. Hjj. The involvement of an external Accountant will contribute to a successful Computerized Accounting System. This hypothesis was supported using PLS analysis. H12. The involvement of an external Consultant w’ii/ contribute to a successful Computerized Accounting System. This hypothesis was supported using regression analysis and PLS. Hj . The involvement of a Support Group will have a positive influence on Subjective Norm. This hypothesis was supported using PLS only. H14. The involvement oja Support Group iiii have a positive influence on Attitude. This hypothesis was partially supported by regression analysis and supported by PLS. H15. The involvement of a Support Group i’iil have a positive influence on Perceived Characteristics ofInnovating variables. This hypothesis was not tested using descriptive statistics or PLS. It was supported using regression analysis. These results are summarized in Table 13(b). 73 CHAPTER 7: CONTRIBUTIONS, IMPLICATIONS ANI LIMITATIONS 7.1 INTRODUCTION There has been much written on the topic of diffusion of innovations and the impact of Information Technology on organizations. While this information may be generally useful, very little of it seems to apply to the small business domain. Research done on large corporations is usually concerned with big business problems and big business solutions. The relevance of these solutions to smaller firms is questionable, as small firms usually have different problems than large firms, or experience large firm problems in ways that are unique to the small business domain. More research addressing real world problems from a small businesss perspective is required which was the goal of this study. 7.2 SUMMARY OF THE RESEARCH PROCESS The motivation for this research came, in part from the lack of useful information available to Public Accountants (and other information consultants) on how to prepare their clients for the successful introduction (diffusion) of new Information Technologies. The particular information technology of interest was the Computerized Accounting System. Afier researching the information system literature it was determined that the most effective tool for obtaining the type of information that information consultants required was from a model on diffusion of information technology developed by Moore. This study has looked at the diffusion of information technology model first developed by Moore, in order to evaluate its robustness and generalizability to a small business, accounting domain. Quantitative and qualitative analysis were done using general descriptive data, regression analysis, and path analysis. As part of this analysis, the role of the information consultant in the diffusion process was examined. 74 The major research questions answered are: 1. What role do independent information consultants, such as accounting firms, play in the Dffusion ofInformation Technology process? 2. Is the DfJiision ofInformation Technology model a general model? Before the first question could be answered, the second question had to be addressed, as the solutions to both are related. In order to determine if the Diffusion of Information Technology was a general model, three different statistical approaches were applied. These included analysing the differences between computerized accounting system users and non- users by general descriptive tests, performing regression analysis of the independent variables on the dependent variables, and finally, applying path analysis using PLS. 7.3 THE RESEARCH QUESTIONS ANSWERED 7.3.1 QUESTION TWO Is the Dffiis/on ofInformation Technology model a general model? Results showed overall support for the general model. The role of information consultants was not very significant when applied to the general model but did show some effect on individual components of the model. The answer, therefore, is a qualified ‘yes”. Based on the results of hypotheses testing for H1 to H7 (excluding H3 which was not tested), no individual hypothesis was fully supported across all three statistical tests applied (H1 received some support using all three methods). However, each hypothesis received either partial support to definite support from at least one of the tests. The regression results indicated that a larger sample size may have obtained more significant results for some of the variables. What is clear from the results is that there are some statistically significant differences between large firms and small firms. In the large firm study where the Diffusion of Information Technology model was first developed, basically all of the variables were significant and provided support for all of the hypotheses. In this study most of the variables were not significant and at best, moderate support was provided to the 75 hypotheses. While a larger sample size would make more of the variables significant, it appears that, based on regression results, several variables would probably not become significant. 7.3.2 QUESTION ONE What role do independent information consultants such as accounting firms play in the Dffusion qfInformation Technology process? Hypotheses H8 to H15 directly addressed this question. Statistical analysis indicate that there is a relationship between the presence of outside support and adoption of computerized accounting systems. Regression analysis indicate that the Computerized Accounting System Support composite variable is significant for all four Innovativeness variables (number of Functions used, Frequency of use, Months since adopted, and Hours of use per week). The individual CAS Support variables had different levels of significance for each of the Innovativeness variables. The CAS Support group Consultant was significant when regressed on number of functions used (p=.000),frequency of use (p=.OO1), months since (‘AS first adopted (p.035), hours of use per week (p=.000), and Visibility (p.OOi). The (‘AS Support group Accountant was significant when regressed on the Perceived Characteristics of Innovating variables Compatibility (p=.Ol2), Ease of Use (p=.026), Relative Advantage (p=.000), and Result Demonstrability (p=.000). The (AS Support group Friend was significant when regressed on the Perceived Characteristics ofInnovating variables Image (p=, 004) and Ti/a/ability (p=.Ol 7). The CAS Support group Personnel was significant when regressed on the Innovativeness variables number offunctions used (p=.OO1) and frequency of use (p.000); on the Perceived Characteristics of Innovating variable Compatibility (p.023); Voluntariness (p=.O2O), and on the Subjective Norm variables Friend (p=.O85), Senior Management (p.O44) and Subordinate (p. 042). 76 The results indicate that CAS Support variables Information Consultants and Personnel have both a direct and an indirect affect on the Computerized Accounting System adoption process. The direct affect can be seen from regression analysis and the indirect affect comes from both regression and path analysis, where the Perceived (Jharacteristics of Innovating variables are used to synthesize Attitude which has a direct affect on adoption (Innovativeness). The role of the Public Accountant is significant indirectly on Innovativeness through its influence on the construct Attitude. The Accountant was shown to have significant influence on the Perceived Characteristics qf Innovating variables Compatibility, Ease of Use, Relative Advantage and Result Demonstrability. PLS analysis indicated the direction and magnitude of this influence on Innovativeness through the intervening variable Attitude. 7.4 CONTRIBUTIONS This study has shown that the Diffusion ofInformation Technology model can be used across different information technology domains and for large or small organizations. The strength of this model is that once the attitudinal and societal characteristics of information technology adoption are understood, information consultants will have the ability to predict if an information technology will be adopted for a given organization. They will also be able to recommend to clients a methodology to maximize the success of the introduction and adoption of an information technology. The reduced Dffrsion of Information Technology questionnaire (39 Questions) has sufficient reliability to be used in similar research. This questionnaire captures Perceived Characteristics of innovating variables adequately, but does not capture the construct Attitude. Although the development of suitable Attitude scales would normally be a recommendation, the use of path analysis programs like PLS to indirectly synthesize this variable suggests that further scale development for Altitude may not be warranted. The inclusion of Communications Channel (ie. Support Group) as an extension to the Dffusion of Information Technology model is an attempt to improve the robustness of this 77 model. Statistical analysis show that this extension does reveal some interactions in the Diffusion of Information Technology model not previously evident. However, the explanatory power of the model has only been modestly improved. This study will add to the small but growing body of research literature specifically oriented towards smaller organizations. While the Dffusion ofInformation Technology model is generalizable and robust in both large and small business domains, it seems that a subset (in PLS terminology) of manifest variables would be more appropriate to the small firm domain. Selection of variables to include in a subset of MVs is problematic, given the lack of quantitative analysis using PLS. 7.5 LIMITATIONS OF THE STUDY This study, while providing some interesting results about the information technology adoption process, has some underlying limitations which must be kept in mind. Sample size is always of some concern, as researchers are almost never satisfied with their sample. While the sample size of 75 was adequate for purposes of PLS analysis, there were several regression results that would probably have become significant with a larger sample. It was difficult to directly compare results with those obtained by Moore as he had a much larger sample size (600) and nearly everything was significant for his study. It becomes difficult to claim differences are due to firm size, or type of injormation technology examined, when results may be due to very large sample sizes. The sample selection methodology, while ensuring a high degree of confidentiality to respondents, resulted in a loss of control of sample selection. The sample should be called a convenience sample because there was very little randomness in the selection process. With the problems encountered in collecting completed questionnaires it is most likely that participating accounting firms selected potential respondents on the basis of individuals who would be most likely to fill out the questionnaire, The demographic’s data for the sample indicate that approximately one third of the respondents were non-users. There were significant differences between users and non-users 78 for several categories. The relatively large proportion of non-users may have swamped or masked some of the results on computerized accounting system adoption. Also, a large portion (65%) of respondents were female, whereas the Moore study had the reverse ratio. The data collection method, self-reported data using a questionnaire, is controversial. Use of self-reported data is criticized because it is often unreliable. While some attempts at improving reliablity were made (asking the same question more than once), the general problems with questionnaire data still remain. While the Dfji’sion of Information Technology model is purported to be a generalizable model, and statements are made about the generalizability of results from this study, it should be kept in mind that the results are applicable to this study only. While it is human nature to make inferences and extrapolate results it should be noted that such inferences and extrapolations are made at the risk of over-interpreting the results from an individual study. 7.6 CONCLUSION The role that information consultants currently play in the adoption process for small and medium size firms is understood a little better. While there is a strong association between the presence of information consultants and the successful adoption of a computerized accounting systems, many small businesses do not rely on this support group to help them with new information technology. It appears that computerized accounting system users who have used a computerized accounting system for a long period of time, and/or use many computerized accounting system functions, are more likely to rely on a support group. It is not clear if the presence of the support group leads to long and versatile computerized accounting system use, or if the experience gained due to the passage of time and/or heavy use has convinced users to seek outside help. If the latter case is true, then Public Accountants have a lot of work to do to get the message out that experience doesn’t have to come the hard way. 79 The extension to the Di/jitsion / Inftrniation Technology model, by including Support Group, has provided a modest improvement to the explanatory and predictive power of the model. Clearly, there is much that is still not known about the factors that can lead to the successful adoption of information technology. However, the extended Diffusion of Information Technology model does provide some insight into this process. TABLES 80 o c H C H H tn (1) C) m It •i j - I j H H C H C r1 1 — r 1 1 H H rn > H > > Z H C) C) — C c H — tTl c C — z 0 CD 0 0 CD 0 — H < t n c m C o - r n 1 r n > c H r - (I D t> c ,) > c i > C )r ’i H H < tn < Cl ) — m ‘ H c i) z (I ) (/ ) H Z H > > C ) pm H .-< z — H * CD Cl) Cl) CD CD 0 -t CD CD -t (ID 0 Cl) CD Cl, C) C) C) -t C) Li . - t C) -t C) 0 0 -q , - t CD C,) 0 Cl) C) C,) H 4- C ) LI Li L i 00 4 . . - . — (ID — H C H O H (ID H 2 H C C 4 - Li . LI . L. ) 00 4 - C ) 4 - 11 ) 2 C Z 0 0 0 0 0 0 L C (ID II 00 LM — C ) C ) L tJ C ) C ) C C ) L i C L. ) (J) Li . — L .J \C ) L — C ) C ( (ID - — — - - — - — — - — - C ô ’ > - . ( I D C ) 0 O L C 0 C J II L 0 C 4 - — C k ) C J Li C ) — C 4- C C ) (ID J L i. L. J LM — J t C ) (I D t. — — - — C C tJ C ) 4 - C 2 (ID (ID (ID (ID LI , 4 - • 00 C ) 00 00 LI , C LI . 4 - t) 4 - cc LI , 00 C ) (ID (ID (ID 00 TABLE 5(a) DEMOGRAPHIC BACKGROUND OF SURVEY RESPONDENTS Relative Ad justed Number Frequency Frequency DEPARTMENT OF EMPLOYMENT Administration 13 17. 3% 1 8.6% Accounting / Finance 37 49.3% 52.9% Other 20 26.7% 28.5% Missing 5 6.7% Total 75 100.0% 100.0% ORGANIZATION LEVEL Executive 15 20.0% 20.8% — Middle Management 13 17.3% 18.1% Supervisory 11 14.7% 15.3% Professional 12 16.0% 16.7% Technical 4 5.3% 5.6% Clerical/Support 15 20.0% 20.8% Other 2 2.7% 2.7% Missing 3 4.0% Total 75 100.0% 100.0% EDUCATION Some High School 3 4.(>% 4.0% High School Graduate 10 13.3% 13.3% Some Technical College 4 5.3% 5.3% Technical College Graduate 3 4.0% 4.0% Some Community College 7 9.4% 9.4% Community College Graduate 6 8.0% 8.0% Some University 14 18.7% 18.7% University Graduate 22 29.3% 29.3% Postgraduate 6 8.0% 8.0% Missing 0 0.0% Total 75 100.0% 100.0% 83 84 TABLE 5(b) DEMOGRAPHIC BACKGROUND OF SURVEY RESPONDENTS Relative Adjusted Number Frequency Frequency AGE Less that 30 years old 27 36.0% 38.0% 30 to 39 years old 27 36.0% 38.0% 40to49ycarsold 13 17.4% 18.3% 50 years old and older 4 5.3% 5.7% Missing 4 5.3% Total 75 100.0% 100.0% SEX Male 26 34.7% 34.7% Female 49 65.3% 65. 3% Missing 0 0.0Y Total 75 100.0% 100.0% OTHER Minimum Maximum Average Firm Size (Sales) $500k- <$250k >$ 10,000k $L000k Average Firm Size (Full Time Employees) 26 92 Avg. Accounting Staff (Full Time Employees) 3 12 85 TABLE 6(a) SURVEY VARIABLES - DESCRTPTWE STATISTICS MAXIMUM MINIMUM # SCALE MEAN STANDARD REPORTED REPORTED ITEMS SCORE DEVIATION SCORE SCORE PERCEIVED CHARACTERISTICS (Scale Range: 1 to 7) Compatibility 4 5.427 1.577 7.000 1.000 EaseofUse 6 5.118 1.003 7.000 2.167 Image 4 3.977 1.366 7.000 1.000 Relative Advantage 8 5.503 1.560 7.000 1.000 Result Demonstrability 3 5.187 1.179 7.000 2.667 Trialability 5 4.128 1.426 7.000 1.000 Visibility 5 4.856 1.484 7.000 1.000 Voluntariness 4 3.130 1.480 5.750 1.000 ATTITUDE 4 2.200 1.214 6.250 1.000 (Scale Range: I to 7) SUBJECTIVE NORMS (Scale Range: -21 to 21) Friends 1 -1.067 — 6.003 21.000 -18.000 Peers 1 1.467 5.512 — 21.000 -8.000 Supervisors 1 .547 5.194 21.000 -8.000 Senior Management 1 -2.320 5.403 21.000 -15.000 Subordnatcs 1 -2.773 5.562 21.000 -21.000 86 TABLE 6(b) SURVEY VARIABLES - DESCRIPTIVE STATISTICS MAXIMUM MINIMUM # SCALE MEAN STANDARD REPORTED REPORTED ITEMS SCORE DEVIATION SCORE SCORE INNOVATIVENFSS MEASURES Frcquency of Usc 1 27.792 8.725 49.000 4.000 (Scale range: 4 to 56) Months Since First Use 2 55.955 27.560 120.500 2.000 (Scale range: 1 to 199) Hours of Use per Week 1 21.584 12.606 40.000 3.000 (Scale range: ito 40) Number of Functions Used 3 4.397 1.790 7.667 1.000 (Scale_range:_0_to_8) CAS SUPPORT (Scale range: I to 7) Personnel from Firm 3 2.629 1.150 5.333 1.000 Friend 3 2.157 .993 5.000 1.000 Accountant 3 2.308 1.025 4.667 1.000 Consultant S 2.277 1.031 4.333 1.000 - — C) C - CD C) CD — r a CI ) — ) — C C — C z C rfj C.’) CI) Cd) C CI) CI) — Cl) CI ) Cl ) Cd ) C Cl) C.’) : - — — - - — - - — — - — CI ) — J — t. ) a — C - — — L1 t’ J — t) L. J C — — - - — — - - — - - — z — a C L. J c j, Cl ) k ) çj , — C D’ J — C C \ 00 88 TABLE 7(b) USERS VERSUS NON-USERS IL VARIABLE MEANS AND TESTS FOR DIFFERENCES (M-W TESTS) NON- U-TEST USERS USERS Z-SCORE SIGNIF PERCEIVED CHARACTERISTICS (M-W) Compatibility 6.10 3.80 -4.86 .0000 Ease of Use 5.25 4.81 -1.47 .1420 Image 4.35 3.08 -3,45 .0006 Relative Advantage 6.18 3.87 -4.81 .0000 Result Demonstrability 5.67 4.02 -5.59 .0000 Trialability 4.33 3.65 -1.9() .0575 Visibility 5.31 3.76 -3.25 .0011 Voluntariness 2.72 4.13 -3.74 .0002 ATTITUDE 1.76 3.25 -3.72 .0002 SUBJECTIVE NORMS Friends -.68 -2.00 -1.02 .3063 Peers .98 2.64 -.72 .4719 Supervisors .06 1.73 -.11 .9146 SeniorManagement -1.38 -4.59 -2.35_ .0189 Subordinates -1.74 -5.27 -2.95 .0032 INNOVATWENESS MEASURES Months elapsed since adoption 55.96 N/A N/A N/A Hours of use per week 21.58 N/A N/A N/A Frequency of use - general once/day N/A N/A N/A Frequency of use - detail 27.79 N/A N/A N/A Number of functions used 4.50 N/A N/A N/A CAS SUPPORT Personnel from Firm 2.63 N/A N/A N/A Friend 2.16 N/A N/A . N/A Accounting Firm 2.31 N/A N/A N/A Consultant 2.28 N/A N/A N/A 89 TABLE 8 REGRESSION RESULTS PERCEIVED CHARACTERISTICS AND VOLUNTARINESS ON ATTITUDE I. SUMMARY OF STEPPED FORCED ENTRY OF VARIABLES II. STA ‘ISTICS FOR VARIABLES IN TI-IF FINAL EOI ATION STEP VARIABLE IN BETA IN R2 F (EQN) SIG F (EQN) 1 Relative Advantage -.852 .726 193 .000 2 Visibility -.240 .755 111 .000 3 Voluntariness .113 .763 76 .000 4 EaseofUse -.095 .769 58 .000 5 Image -.073 .773 47 .000 6 Trialability .057 .775 39 .000 7 Compatiblity .076 .776 33 .000 8 Result Demonstrability -.019 .776 29 .000 I VARIABLE BETA STD ERR F (B) STG F BETA Relative Advantage -.632 .123 16.061 .000 Visibility -.2 10 .083 4.260 .043 Voluntariness .113 .067 1.924 .170 Ease of Usc -.123 .090 2.720 .104 Image -.079 .061 1.360 .248 Trialability .050 .069 .377 .541 Compatiblity .083 . 123 .269 .605 Result Demonstrability -.019 .078 .062 .805 R2=.776 Variance Explained Adjusted R2 .749 90 TABLE 9 REGRESSION RESULTS PCI’S, VOLUNTARINESS, SN AND SUPPORT ON ATTITUDE (IUatiOII PCI Only PCI and SN PCI,SN & SUPPORT Beta Weights Beta Sig. F Beta Sig. F Beta 1 Sig. F Relative Advantage -.63 2 .000 -.653 .000 -.681 .000 Visibility -210 .043 -.199 .054 -.200 .053 Voluntariness .113 .170 .092 .277 .083 .326 Ease of Use — -.123 .104 -.109 .155 — -.101 .188 Image -.079 .248 -.085 .216 -.087 .206 Trialability .050 .541 .057 .492 .060 .464 Compaliblity .083 .605 -.083 .603 .083 .603 Result Demonstrability -.019 .805 -.020 .793 -.048 .552 Subjective Norm -.069 .282 -.077 .235 CAS Support .074 .314 Variance Explained R2 =776 R2 =780 R2 =.783 Adj R2 =749 Adj R2 =750 Adj R2 =750 91 TABLE 10 (a) REGRESSION RESULTS ATTITUDE, SN, AND VOLUNTARINESS ON INNOVATIVENESS VARIABLES DEPENDENT INDEPENDENT Beta Sig. F Adj. F F Sig R2 NUMBER OF VoLuntariness -.023 .861 FUNCTIONS Attitude -.471 .000 USED Subjective Norm -.074 .503 .193 6.903 .0004 FREQUENCY OF Voluntariness -.092 .472 USE Attitude -.454 .00() Subjective Norm -.035 .746 .223 8.069 .0001 MONTHS SINCE Voluntariness . -. 107 .405 ADOPTED Attitude -.433 .001 Subjective Norm -.067 .538 .207 7.445 .0002 HOURS OF USE Voluntariness -. 191 .138 PER WEEK Attitude -.382 .003 Subjective Norm -.073 .500 .218 7.881 .0001 92 TABLE 10 (h) REGRESSION RESULTS ATTITUDE, SN, VOLUNTARINESS & SUPPORT ON INNOVATIVENESS VARIABLES DEPENDENT INDEPENDENT Beta Sig Adj. F F Sig R2 NUMBER OF Voluntariness -.052 .605 FUNCTIONS Attitude -.22 1 .035 USED Subjective Norm -.087 .302 Support .617 . .000 .524 21.338 .0000 FREQUENCV Voluntariness -. 122 .185 OF USE Attitude -.189 .050 Subjective Norm -.050 .524 Support .654 .000 .597 28.390 .0000 rvIONTHS Voluntariness -. 124 .304 SINCE Attitude -.289 .022 ADOPTED Subjective Norm -.075 .462 Support .354 .001 .309 9.260 .0000 HOURS OF USE Voluntariness -.2 14 .054 PER WEEK Attitude -.182 .110 Subjective Norm -.084 .366 Support .492 .000 .425 14.669 .0000 93 TABLE 11 (a) REGRESSION RESULTS PCI AND SUBJECTIVE NORMS ON INNOVATIVENESS EQUATION 1: DEPENDENT VARIABLE - NUMBER OF FUNCTIONS USED Entry Independent Variable Final Beta Sig. F Ad. R2 F Step 1 Result Demonstrability .447 .000 2 SN Peers -.297 .001 3 SN Subordinates .268 .004 4 Relative Advantage .217 .049 .504 20 Weakly Significant: Ease of Use -.186 .061 EQUATION 2: DEPENDENT VARIABLE - FREQUENCY OF USE Entry Independent Variable Final Beta Sig. F Ad. R2 F Step I Relative Advantage 403 .000 2 Result Demonstrability .365 .001 461 33 Weakly Significant: SN Subordinate .158 072 Ease of Use -.172 .091 EQUATION 3: DEPENDENT VARIABLE - MONTHS SINCE CAS FIRST ADOPTED Entry Independent Variable Final Bela Sig. F Adj. R2 F Step I Result Demonstrability .597 .000 .347 40 Weakly Significant: Relative Advantage .22-I .058 Voluntariness -. 179 .065 Coinpatibility .210 .078 EQUAT ON 4: DEPENDENT VARIABLE - HOURS OF USE PER WEEK Entry Independent Variable Final Beta Sig. F Adj. R2 F Step I Result Demonstrability 445 .025 2 Visibility .268 .013 .362 22 94 TABLE 11(h) REGRESSION RESULTS PCI, SUBJECTIVE NORMS, & SUPPORT ON INNOVATIVENESS EQUATION 1: DEPENDENT VARIABLE - NUMBER OF FUNCTIONS USED Entry Independent Variable Final Beta Sig. F Mj. R2 F Step I Result Demonstrability .349 .000 2 SN Peers -.281 .00() 3 SUPP Personnel .290 .001 4 SUPP Consultant .282 .001 5 SN Subordinate .199 .009 .660 30 EQUATION 2: DEPENDENT VARIABLE - FREQUENCY OF USE Entry Independent Variable Final Bela Sig. F Mj. R2 F Step 1 SIJPP Personnel .400 .000 2 SUPP Consultant .225 .001 3 Relative Advantage -.464 .006 4 Result Demonstrability .201 .014 5 SN Peers -. 135 .030 .727 41 Weakly Significant: SN Subordinate . 127 .062 EQUATION 3: DEPENDENT VARIABLE - MONTHS SINCE CAS FIRST ADOPTED Entry Independent Variable Final Beta Sig. F Adj. R2 F Step 1 Result Demonstrability .497 .000 2 Consultant .221 .035 .378 23 Weakly Significant: Voluntariness -. 169 .076 EQUATION 4: DEPENDENT VARIABLE - HOURS OF USE PER WEEK Entry Independent Variable Final Beta Sig. F Adj. R2 F Step 1 SUPP Consultant .436 .000 2 Result Demonstrability .3 12 .001 3 Voluntariness —.225 .010 .510 27 95 TABLE 12(a) REGRESSION RESULTS CAS SUPPORT ON OTHER DEPENDENT VARIABLES EQUATION 1: DEPENDENT VARIABLE - SUBJECTIVE NORM (COMPOSITE) Dep. Variable Independent Variable Final Beta Sig. F Adj. R2 F SNc No Significant Variables --- EQUATION 2: DEPENDENT VARIABLE - SUBJECTiVE NORM (COMPONENTS) + VOLUNTARINESS Dep. Variable Independent Variable Fiiial Beta Sig. F Adj. R2 F Friend SUPP Personnel .200 .085 .027 3 Peer No Significant Variables --- Supervisor No Significant Variables --- Senior Mgiut SUPP Personnel .234 .044 .042 4 Subordinate SUPP Personnel .236 .042 .043 4 EQUATION 3: DEPENDENT VARIABLE - PERCEIVED CHARACTERISTICS OF INNOVATING Dep. Variable Independent Variable Final Beta Sig. F Adj. R2 — F Compatibilty SLJPP Accountant .302 .012 “ SUPP Personnel .272 .023 .225 12 ********* Ease of Use SUPP Accountant .257 .026 .053 5 Image SUPP Friend .325 .004 .093 9 Rd. Advant. SUPP Accountant .487 .000 227 23 Res. Demon. SUPP Accountant .532 .00() .273 29 Trialability SIJPP Friend .275 .017 .063 6 Visibility SUPP Consultant .377 .001 . IS I 12 96 TABLE 12(b) REGRESSION RESULTS CAS SUPPORT ON OTHER DEPENDENT VARIABLES EQUATION 3: DEPENDENT VARIABLE - INNOVATIVE (USE) VARIABLES Dep. Variable Independent Variable Final Beta Sig. F Ad. R2 F Frequency SUPP Consultant .424 .000 SUPP Personnel .481 .000 .606 58 ********* Functions SUPP Consultant .445 .000 11 SUPP Personnel .383 .000 .503 38 * ** * *** * * Hours SUPP Consultant .522 .000 SUPP Personnel .181 .092 .384 24 ********* Months SUPP Consultant .327 .008 SUPP Personnel .234 .053 .218 11 EQUATION 5: DEPENDENT VARIABLE - VOLUNTARINESS Dep. Variable Independent Variable Final Beta Sig. F Mi. R2 F Voluntariness SUPP Personnel -. 267 .020 .059 6 97 TABLE 13 (a) — SUMMARY RESULTS OF HYPOTHESIS TESTING HYPOTHESES ADOPTERS VS. REGRESSION PLS NON-ADOPTERS ANALYSIS ANALYSIS Hi: ATTITUDE -> INNOVATIVENESS SUPPORTED NOT SUPPORTED SUPPORTED H2: RELATIVE ADV> OTHER PCI SUPPORTED SUPPORTED SUPPORTED H3: AVOIDANCE < OTHER PCI N/A N/A N/A H4: SN -> INNOVATIVENESS NOT NOT NOT SUPPORTED SUPPORTED SUPPORTED H5: SN -> ATTITUDE N/A NOT NOT SUPPORTED SUPPORTED H6: VOLUNTARY -> INNOVATIVENESS SUPPORTED NOT NOT SUPPORTED SUPPORTED H7: VOLUNTARY -> ATTITUDE N/A NOT SUPPORTED SUPPORTED H8: SUPPORT -> INNOVATIVENESS NOT SUPPORTED SUPPORTED SUPPORTED H9: FRIEND -> INNOVATIVENESS NOT NOT NOT SUPPORTED SUPPORTED SUPPORTED H1O: EMPLOYEE -> INNOVATIVENESS NOT SUPPORTED SUPPORTED SUPPORTED Hi 1: ACCOUNTANT -> INNOVATIVE NOT NOT NOT SUPPORTED SUPPORTED SUPPORTED H12: CONSULTANT -> INNOVATIVE NOT SUPPORTED SUPPORTED SUPPORTED Hi 3: SUPPORT -> SN N/A NOT SUPPORTED SUPPORTED H14: SUPPORT -> ATTITUDE N/A NOT SUPPORTED SUPPORTED HI 5: SUPPORT -> PCVS N/A SUPPORTED SUPPORTED 98 TABLE 13(b) SUMMARY RESULTS OF HYPOTHESIS TESTING (MOORE) HYPOTHESES ADOPTERS VS. REGRESSION PLS NON-ADOPTERS ANALYSIS ANALYSIS HI: ATTITUDE -> INNOVATIVENESS SUPPORTED SUPPORTED SUPPORTED H2: RELATIVE ADV> OTHER PCI NOT SUPPORTED NOT SUPPORTED SUPPORTED H3: AVOIDANCE < OTHER PCI SUPPORTED NOT SUPPORTED SUPPORTED H4: SN -> INNOVATIVENESS SUPPORTED NOT SUPPORTED SUPPORTED H5: SN -> ATTITUDE N/A SUPPORTED SUPPORTED H6: VOLUNTARY -> INNOVATIVE SUPPORTED SUPPORTED SUPPORTED H7: VOLUNTARY -> ATTITUDE N/A SUPPORTED SUPPORTED 99 TABLE 14 GENERAL PLS STATISTICS FOR TESTED MODELS MODEL MULTIPLE R AVG. COMMUN. AVG. REDUND. . (R2) (H2) (F2) All data points: Full Model .1837 .5275 .1097 Full - Voluntariness .1281 .4636 .0584 Full - Voluntariness - SN 1357 .2835 .0424 Full-4PCI’s .1992 .5558 .1252 Full + Communications Channel . 1960 .5389 .1238 CAS User data points: .0957 .4 173 .0470 Full Full - Voluntariness . 1012 .3593 .0379 Full - Voluntariness - SN .1501 .3745 .0470 Full - 4 PCIs . 1209 .4575 .0624 Full + Communications Channel . 1242 .4062 .0559 — FI G U R E 1 D iff us io n o f I nf or m at Io n Te ch no lo gy M od el - M oo re , 19 89 TO W A R D S SU B JE C TI V E N O R M V O LU N TA R IN ES S A TT IT U D E IN N O V A TI V EN ES S A D O PT IN G FI G U R E 2 D iff us io n o i I nn ov at io ns M od el - R og er s, 19 83 Pe rc ei ve d A ttr ib ut es C om m un ic at io n C ha nn el s o f I nn ov at io ns - M as s M ed ia - R el at iv e A dv an ta ge - In te rp er so na l - Co m pa tib ili ty - Tr ia la bi lit y - Co m pl ex ity - O bs er va bi lit y Ty pe o f I nn ov at io n- D ec is io n R 4T E O F A D O PT IO N E xt en t o f C ha ng e A ge nt s’ - O pt io na l rs Pr om ot io n E ff or ts - Co lle ct iv e O F IN N O V A TI O N S - A ut ho rit y I___ ____ ____ ____ ____ ____ ____ _ N at ur e o fS oc ia l S ys te m - N or m s - D eg re e o fI nt er co nn ec te dn es s FI G U R E 3 C O M M U N IC A TI O N S N ET W O R K In no va tio n D ec is io n M od el - Fi sh be in & A jze n, 19 75 (A da pt ed by M oo re , 19 89 ) O B JE C TI V E C H A R A C TE R IS TI C S O F IN N O V A TI O N TO W A R D S A D O PT IN G IN TE N TI O N IN N O V A TI O N D EC IS IO ) ) SU B JE C TI V E - 1 PE R SO N A L C H A R A C TE R IS TI C S O F A D O PT ER S N O R M B EH A V IO U R A L N. 1/ B EH A V IO U R (A DO PT IO N/ RE JE CT IO 1 A TT IT U D E O B JE C TI V E C H A R A C TE R IS TI C S O F PR EC U R SO R C FI G U R E 4 St ag es o ft he In no va tio n D ec isi on Pr oc es s M od el R og er s, 19 83 (A da pte d by M oo re , 19 89 ) K N O W LE D G E PE R SU A SI O N _ _ _ _ _ _ _ _ _ _ _ D E C IS IO N A D O PT IO N R E JE C TI O N ff iM A T IO N Jl FI G U R E 5 N O N -C A S U SE RS R ES U LT D EM O N ST R A B IL IT Y R s A c T A N L G E 01 00 01 01 01 03 01 04 01 05 01 06 01 07 01 68 02 03 02 04 02 06 02 07 02 08 R E SP O N D E N T 02 09 02 10 02 11 02 16 02 17 02 19 02 33 P1 24 P1 25 7 6 4 3 0 - A — I I I I I I I I I I I I I I I • U 15 U 23 A U33 1 FI G U R E 6 D IF FU ST IO N O F IN FO R M A TI O N TE C H N O LO G Y M O D EL PL S LO A D IN G S O N O R IG IN A L M O D EL C = LA TE N T V A RI A BL E (L V) I I= M A N IF ES T V A RI A BL E (M V) IN N ER M A TR IX = OU TE R M A TR IX -R EF LE CT IV E OU TE R M A TR IX -F O RM A Tr V E Su bo rd in at es 8Z 5. 3 . 91 41 70 92 . 90 94 6 q 9 “ 7 3 FI G U R E 7 D IF FU SI O N O F IN FO R M A TI O N TE C H N O LO G Y M O D EL PL S LO A D IN G S O N EX TE N D ED M O D EL . 29 35 = LA TE N T V A R IA B LE I I=M A N IF ES T V A R IA B LE . 91 36 CC 109 BIBLIOGRAPHY Ahituv, Niv, “Assessing The Value Of Information Problems and Approaches”, Proceedings of the Tenth International conference on Information Systems, December 4-6, 1989, pp. 3 15-3 25. Ajzen, Icek and Fishbein, Martin, Understanding Attitudes and Predicting Behavior, Prentice- Hall Inc., Englewood Cliffs, NJ, 1980. Alavi, Maryam and Weiss, Ira R., “Managing The Risks Associated With End-User Computing”, Journal of MIS, Vol. 2, No. 3, Winter 1985-86, pp. 5-20. Alavi, Maryarn; Nelson, R. Ryan; and Weiss, Ira R., “Strategies For End-User Computing: An Integrative Framework”, Journal of MIS, Vol. 4, No. 3, Winter 1987-88, pp. 28-49. Alexander, Mary B., The Adoption And Implementation Of Computer Technology In Organizations: The Example Of Database Machines, Unpublished PhD thesis, Indiana University, 1989. Allen, Brandt, “An Unmanaged Computer System Can Stop You Dead”, Harvard Business Review, November-December 1982, pp. 77-8 7. Arter, Terrence J., “Contract The Independent Computer Consultant”, Journal of Small Systems Management, January 1988, pp 36-37. Arthur Anderson & Co. “Entrepreneurs Speak Out”, BC Business, April, 1990, pp. 73-77. Attewell, Paul, “Survey & Other Methodologies Applied To IT. Impact Research”, Paper presented at the Workshop on Survey Research in MIS, Irvine, California, February 10- 11, 1989. Babcock, Charles, “Big 8 Firm Tied to 4GL Snafu”, Computer World, July 28, 1986, pp. 1,4. Bagozzi, Richard P. “Causal Modeling: A General Method For Developing And Testing Theories In Consumer Research”, American Marketing Association, 1982, pp. 195-20 1. Bailey, James E. and Pearson, Sammy W., “Development of a Tool For Measuring and Analyzing Computer User Satisfaction”, Management Science, Vol. 29, No. 5, May 1983, pp. 530-545. Baker, William H., “Status Of Information Management In Small Business”, Journal of Systems Management. April, 1987, pp. 10-15. Barki, H. and Huff, Sid L., “Measuring Decision Support System Success”, ASAC 1984 Conference. Barclay, D.W., Higgins, C.A., & Duxbury, L.E., An Introduction to the Partial Least Squares Approach to Causal Modeling, University of Western Ontario, WP-91-08, 1991. 110 Baronas, Ann-Marie K. and Louis, Meryl Reis, “Restoring a Sense of Control During Implementation; How User Involvement Leads to System Acceptance”, MIS Quarterly, March 1988, pp. 111-123. Baroudi, Jack J., Olson, Margrethe H., and Ives, Blake, “An Empirical Study Of The Impact Of User Involvement On System Usage And Information Satisfaction”, Communications of the ACM, Vol. 29, No. 3, March 1986, pp. 232-238. Batch, CR.; Fisher, J.; and McIntyre, R,, “Accountants Take To IT In A Big Way”, Ihc Bottom Line, November 1989, pp. 22-24. Benjamin, Robert I., “Information Technology in the 1990s: A Long Range Planning Scenario”, MIS Quartey, June 1982, pp. 11-31. Benjamin, Robert I. and Scott Morton, Michael S., “Information Technology, Integration, And Organizational Change”, Interfaces, Vol. 18, No. 3, May-June 1988, pp. 86-98. Benson, David H., “A Field Study of End User Computing: Findings and Issue&’, MIQuarterly, December 1983, pp. 35-45. Boon, J.A. and Pienaar, H., “Some Ideas On The Microcomputer And The Information I Knowledge Workstation”, Microcomputers for Information Management, Vol. 6, No. 2, pp. 113-122. Bouldin, Barbara M., Agents of Change, by Edward Press Computing Series, 1989. Boynton, Andrew C. and Zmud, Robert W., “Information Technology Planning in the 1990’s: Directions for Practice and Research”, MIS Quarterly, March 1987, pp. 59-71. Bracker, Jeffiey S. and Pearson, John N., “The Impact Of Consultants On Small Firm Strategic Planning”, Journal of Small Business Management, July 1985, pp.23-30. Bradbard, David A.; Norris, Dwight R.; and Kahai, Paramjit H., “Computer Security in Small Business: An Empirical Study”, Journal of Small Business Management, Vol. 28, No. 1, January 1990, pp. 9-19. Brancheau, James C., The Diffusion Of Information Technology: Tejg And Extending Innovation Diffusion Theory In The Context Of End-User Computing, Unpublished PhD thesis, University of Minnesota, 1987. Brancheau, James C., and Wetherbe, James C., “The Adoption Of Spreadsheet Software: Testing Innovation Diffusion Theory In The Context of End-User Computing”, Information Syst Research, Issue 1: Vol.2, pp. 115-141. Brown, Donald A., “So Far, So Good”, CA Magazine, April 1992, pp. 56-58. 111 Bryman, Alan, and Cramer, Duncan, Quantative Data Analysis For Social Scientists, Routledge, London UK, 1990. Buckler, Grant, “Boost Your Training Buck”, Small Business, March 1990, pp. 65-66,7 1. Buechert, Dennis “Skilled Labour Shortage Chronic, List Shows”, The Vancouver Sun, Friday, January 3, 1992. BYTE, McGraw-Hill Inc., Vol. 16, No. 8, August 1991, pp. 218-224. Chin, Wynne. University of Calgary. Based on telephone conversations with the researcher during early June, 1992. Churchill, Gilbert A., “A Paradigm for Developing Better Measures of Marketing Constructs”, Journal of Marketing Research, Vol. XVI, February 1979, pp. 64-73. Clemons, Eric K. and Row, Michael, C., “Information Technology And Economic Reorganization”, Proceedings of Tenth International Conference on Information Systems, December 1989, pp. 341-35 1. Cooper, Randolph B. and Zmud, Robert W., “Information Technology Implementation Research: A Technological Diffusion Approach”, Management Science, Vol. 36, No. 2, February 1990, pp. 123-139. Cox, Gary H., “Technology’s Rewards Without The Risks”, Datamation, February 1, 1990, pp. 69-73. Davidson, WA., “Straw Accountants”, CA Magazine, September 1991, pp. 43. Davidson, R.A. and Dalby, J.T., “Personality Profiles of Public Accountants”, Working Paper, University of Calgary, January 1991, from CA Magazine, September 1991, pp. 43, 45. Davis, Gordon B., “Caution: User-Developed Decision Support Systems Can Be Dangerous To Your Organization”, Working Paper No. MISRC-WP-82-04, Management Information Systems Research Center, School of Management, University of Minnesota, 1981. Delente, Gisele, Miller, Brian A., & Stovel, Gordon, “Y CAs R MVPs”, CA Magazine, February 1990, pp. 43-45. Delone, William H. “Determinants of Success for Computer Usage in Small Business”. Mi Quarterly, March 1988. Dillman, Don A., “Mail and Telephhone Surveys: The Total Design Method”, Wiley and Sons, New York, 1978. Dimnik, Tony “An Introduction to LISREL”, The University of Western Ontario, Working Paper Series No. 86-21, 1986. 112 Dragich, Martha J., “Information Malpractice: Some Thoughts On The Potential Liability Of Information Professionals’, Information Technology and Libraries, September 1989, pp.265-272. Ein-Dor, Phillip and Segev, Eli, “Organizational Context and MIS Structure: Some Empirical Evidence”, MIS Ouarterly, September 1982, pp. 55-68. FBDB, “Small Business Facts and Figures”, Federal Business Development Bank, 1987. Fournell C., Tellis G.J., and Zinkhan G.M., “Validity Assessment: A Structural Equations Approach Using Partial Least Squares”, Proceedings, American Marketing Association Educators’ Conference, 1982, pp. 1-5. Frame!, John E., “Managing Information Costs And Technologies As Assets”, Journal of Systems Management, February 1990, pp. 12-19. Gable, Guy G. “Consultant Engagement for First Time Computerization: A Pro-Active Client Role in Small Enterprises”, ASAC 1989 Conference, pp. 57-60. Gallupe, R. Brent, “End-User Computing: Individual Differences and Successful End Users, An Exploratory Case Study”, ASAC Conference Montreal, 1989, pp. 61-75. Geilman, Harvey, “Soaring With Eagles”, Office Systems and Technology, Vol. 37, No. 2, March 1991, pp. 28-42. Goodhue, Dale, “IS Attitudes: Toward Theoretical And Definition Clarity”, Proceedings of Seventh Interritional Conference on Information Systems, December 1986, pp. 181-194. Goodson, Leslie, “What’s Wrong With This Picture?”, CA Magazine, August 1990, pp. 20-28. Gotleib, Leo, “IS May Be An ‘Honest Broker’ Between Vendors, Users”, Globe & Mail, March 6, 1990, pp. C15. Gremillion, Lee L. and Pyburn, Philip, “Breaking the Systems Development Bottleneck”, Harvard Business Review, March-April, 1983, pp. 130-137. Gunning, Ken “Accounting For Canada”, CA Magazine, April 1992, pp. 3 9-42. Hamilton, Marilyn, Small Business Views Of’ The Accounting Profession in Canada - Survey 9, Dobson Marketing Services, 1989, Abbotsford, B.C. Portions of this survey were reproduced in Sial1 usinessMagazii.ç, January 1990. Hill, Thomas; Smith, Nancy D. and Mann, Millard F., “Role of Efficacy Expectations in Predicting the Decision to Use Advanced Technologies: The Case of Computers”, Journal of Applied Psychology, Vol. 72, No. 2, pp. 307-3 13. 113 Holmes, Scott and Nicholls, Des. “Modelling the Accounting Information Requirements of Small Business”, Accounting and Business Research, Vol. 19, No. 74, 1989, pp. 143- 150. Huber, 0. P., “A Theory of the Effects of Advanced Information Technologies on Organizational Design, Intelligence, and Decision Making”, Academy of Management Review, Vol. 15, No. 1, January 1990. Ives, Blake, Olson, Margrethe H. and Baroudi, Jack J., “The Measurement of User Information Satisfaction”, Communications of the ACM, Vol. 26, No. 10, October 1983, pp.785-93. Jenish, D’Arcy, “A ‘Terrorist’ Virus”, Maclean’s, March 16, 1992, pp. 48-51. Kauffman, Robert J. and Weill, Peter, “An Evaluative Framework For Research On The Performance Effects Of Information Technology Investment”, Proceedings of the Tenth International Conference on InformatiolLSystems, December 4-6, 1989, pp. 377-388. Kraut, Robert; Dumais, Susan; and Koch, Susan, “Computerization, Productivity, AndQuality Of Work-Life”, Communications of the ACM, Vol. 32, No. 2, February 1989, pp. 220-23 8. Kuntz, M.A., and Maingot, M., “Viruses Are Alive and Well”, CA Magazine, December 1989, pp. 60-6 I. Lees, John D., ‘Successful Development Of Small Business Information Systems”, Journal of Systems Management, September 1987, pp. 32-39. Lees, John D. and Lees, Donna D., “Realities Of Small Business Information System Implementation”, Journal of Systems Management, January 1987, pp. 6-13. Lefebvre, Elizabeth and Lefebvre, Louis A., Planning For Information Technology Acquisition: The Case Of Small Business, unpublished working paper, the University ofQuebec at Montreal, 1990. Lin, Engming and Ashcraft, Phillip, “A Case Of Systems Development In A Hostile Environment”, Journal Of Systems Management, April 1990, pp. 11-14. Lind, Mary R. arid Zmud, Robert W., “The Influence of a Convergence in Understanding Between Technology Providers and Users on Information Technology Irinovativeness”, Org. Science, Vol. 2, No. 4, 1990. Lohmoller, Jan-Bernd, “LVPLS 1 .6 Program Manual: Latent Variables Path Analysis With Partial Least-Squares Estimation”, Munich, 1984. Lucas, Henry C. Jr., “Empirical Evidence For a Descriptive Model of Implementation”, MISQuarterly, June 1978, pp. 27-41. 114 Luscombe, N. “Representation or Fragmentation?”, CA Magazine, May 1990, pg. 3. McFarlan, Warren F. and McKenny, James L., “Corporate Information Systems Management: The Issues Facing Senior Executives”, Richard D. Irwin Inc., USA, 1983. McGill, Penny, “Focused Approach Helps When Designing Firm’s Information Systems”, Globe & Mail, March 6, 1990, pp. C12. McKeen, James D., “Successful Development Strategies for Business Appliction Systems”, MIS Quarterly, September 1983, pp. 47-65. McLean, E. R., “End Users As Application Developers”, MIS Quarterly, December 1979, pp. 37-46. Melone, Nancy P., “A Theoretical Asessment Of The User-Satisfaction Construct In Information Systems Research”., Management Science, Vol. 36, No. 1, January 1990, pp. 76-91. Melone, Nancy P. and Bayer, Judy, “A Dynamic Model Of The Impact Of Communication- Channel Use On The Technology-Transfer Process”, GSIA WP# 1990-21, Carnegie Mellon University, July 1990. Miller, Howard W., “Developing Information Technology Strategies”, Journal of Systems Management, September 1988, pp. 28-3 5. Millman, Zeeva and Hartwick, Jon, “The Impact Of Automated Office Systems On Middle Managers And Their Work”, MIS Quartjy, December 1987, pp. 479-491. Moore, Gary C., The Examination of the Implementation of Information Teno1ogv For End Users: A Diffusion of Innovations Perspective, Unpublished PhD thesis, UBC 1989. Moore, Gary C. and Benbasat, Izak, “Development of an Instrument To Measure The Perceived Characteristics of Adopting an Information Technology Innovation”, Thc Institute of Management Sciences, Sept. 1991, pp. 192-222. Nelson, R.R. and Cheney, Paul H., ‘Training End Users: An Exploratory Study”, MISQuartiy, December 1 987. Newman, Michael, “Changing The Change Agent: An Admissions Sysem Case Study”, Journal of Systems Management, January 1990, pp. 6-12. Nilakanta, Sree and Scamell, Richard W., “The Effect Of Information Sources And Communication Channels On The Diffusion Of Innovation In A Data Base Development Environment”, Management Science, Vol. 36, No. 1, January 1990, pp. 24-40. Nunnally, Jum C., Psychometric Theory, McGraw Hill, New York, 1967. Nunnally, Jurn C., Psychometric Theory, McGraw Hill, New York, 1978. 115 Overbey, John T., Carland, Jo Ann, and Carland, James W., “Impact Of Microcomputers On Accounting Systems”, Journal of Systems Management”, June 1987, pp. 20-27. Pavitt, K.; Robson, M.; and Townsend, J., “Technological Accumulation, Diversification And Organisation In UK Companies, 1945-1983”, Management Science, Vol. 35, No. 1, January 1989, pp. 8 1-99. Peat, Marwick, Mitchell, and Co., “CPA’s Ranked First As Outside Consultants For Small, Private Companies”, Journal of Accountancy, Vol. 157, No. 1, 1984, pp. 22-24, found in Bracker & Pearson (see above reference). Pendegraft, Norman; Morris, Linda; and Savage, Kathryn, “Small Business Computer Security”, Journal of Small Business Management, October 1987, pp. 55-61. Pentland, Brian T., “Use And Productivity In Personal Computing: An Empirical Test”, Proceedings of the Tenth International Conference on Information Systems, December 1989, pp. 211-222. Pratkanis, Anthony R; Breckler, Steven J, and Greenwald, Anthony G; (editors), Attitude, Structure And Function, Lawrence Eribaum Associates, Hilisdale, NJ, 1989. Raymond, Louis, “Organizational Characteristics and MIS Success in the Context of Small Business’, MIS OuarteiJ, March 1985, pp. ??? Reich, Blaize H. and Benbasat, Izak, “The Use Of Information Technology For Competitive Advantage in Canada: An Examination Of Information Systems Linking Companies To Their Customers”, ASAC 1988 Conference, 1988, pp. 171-18 1. Richmond Business, “Small Business is Big”, Richmond Review (quoting from a report prepared by the provincial Ministry of Regional and Economic Development in 1990), October 3, 1990, pp.3. Rivard, Suzanne and Huff, Sid L., “Factors of Success for End-User Computing”, Communications of the ACM, Vol. 31, No. 5, May 1988, pp. 553-561. Rockart, John F. and Flannery, Lauren S., “The Management of End User Computing”, Communications of the ACM, Vol.26, No. 10, October 1983, pp. 776-784. Rockburn, Jefl “The Terrorism of Viruses May Be Prevalent But is Also Preventable”, Globe & Mail, March 6, 1990, pp. C6. Rogers, Everett M., Diffusion of Innovations, Third Edition, The Free Press, New York, 1983. Rosen, R.J; Baker, G.R. Healy; and RH., Rogers, D.W., cQp Control Guidelines, Second Edition, Canadian Institute of Chartered Accountants, 1986. 116 Sanders, G.L. and Courtney, JR. “A Field Study of Organizational Factors Influencing DSS Success”, MIS Quarterly, March 1985. Sein, Maung K., Bostrom, Robert P.; and Olfman, Lorne, “Training End Users To Compute: Cognitive, Motivational And Social Issues”, INFOR, Vol. 25, No. 3, August 1987, pp. 236-255. Small Business Magazine, October 1989, title & author unknown. Smith, Charlie. “The Dominant Solution”, Equity, April 1989, pp 25. Steiren, Carl, “Firms Find On-Site Instruction Inexpensive and Convenient”, Globe & Mail, March 6, 1990, pp. C4. Stone, Eugene, Research Methods in Organizational Behavior, Scott, Foresman and Company, 1978). Stulberg, Gregg, Focus On Office Technology, Financial Post, March 22, 1991, pp. 22 Thompson, Ron L., “An Empirical Investigation Of Factors Affecting The Use Of Personal Computers By Knowledge Workers”, ASAC 1989 Conference, 1989, pp. 141-152. Walker, Robert, “Let’s Go For IT”, CA Magazine, April 1991, pp. 33-35. Walton, Charles and Durham, Ashley, “Information Systems Liability”, Journal of Systems Mc.ment, October 1988, pp. 35-41. Willits, Stephen D., “Decisions, Decisions, Decisions”, CA Magazine, August 1990, pp. 51- 54. Wold, H., “Systems Analysis By Partial Least Squares”, in Measuring The Unmeasurable, NATA AS! Series, Martinus Nijhoff Publishers, Boston, 1985, pp. 22 1-252. Zmud, Robert W., “Individual Differences and MIS Success: A Review of the Empirical Literature”, Management Science, Vol. 25, No. 10, October 1979, pp. 966-979. 1/ ) 118 APPENDIX I-A Relative Advantage: the degree to which an innovation is perceived as being better than its precursor Compatibility: the degree to which an innovation is perceived as being consistent with the existing values, needs, and past experiences of potential adopters Ease of Use: the degree to which an innovation is perceived as being difficult (Complexity) to use Trialability: the degree to which an innovation may be experimented with before adoption Observability: the degree to which the results of an innovation are observable to others Image: the degree to which use of an innovation is perceived to enhance ones image or status in ones social system Voluntariness: the degree to which use of the innovation is perceived as being voluntary, or of free will Visibility: the degree to which the innovation is apparent to the sense of sight r — APPENDIX I-B INNOVATIVENESS: (Moore, 1989, pp. 133) Adoptive degree to which an individual is relatively early in adopting an innovation. Implementation degree to which an individual puts an innovation to use within a given use domain. Use degree to which an individual who has adopted the innovation uses it to solve novel problems, or in a new use domain. 119 (From Stone, 1978) VALIDITY ITEM DEFINITION Content Validity Measurement items are representative sample of domain of items associated with variable being measured. Construct Validity Appropriate operational definition cf an abstract variable (construct) Criterjon-related —— Use of scores obtained from one measure Validity (predictor) to infer individual’s probable standing on another variable (criterion) Face Validity Item appears to measure what it claims to measure. Incremental Validity Item provides an improvement in predictive power in conjunction with other measure(s) over the use of the other measure(s) alone. Convergent/Discriminant Scores on the measure correlate highly with Validity scores on other independent measures of the variable and correlate low on measures of other variables. APPENDIX I-C Note: There are several other Validity items, only the most commonly used items are discussed above. APPENDIX H-A 120 121 Appendix II- Al Questionnaire 122 WELCOME! You are about to participate in a study of opinions about the usage of microcomputers in the accounting function. In some sections you will be asked questions about and see reference to the term CAS, which stands for COMPUTERIZED ACCOUNTING SYSTEM. A CAS is defined for the purposes of this study as a set of computerized tools for an individual, and usually consists of a personal or microcomputer with one or more software packages, such as an accounting program, and/or other software such as spreadsheet, database, word processing, etc. in support of the accounting function. The key aspect of a CAS is that it is computer technology that you would use directly, as opposed to having someone else use for you. In completing the questionnaire, please remember: 1. All the information you give is kept confidential. 2. We need answers to all questions. Please don’t skip any. 3. Be honest - tell it like it is. 4. Please don’t talk to others about how you respond to the questions. We would like your opinion, not the opinion of your associates. 5. Even if you have never used a CAS, please answer all the relevant questions as best as you can. 6. Move rapidly through the questionnaire. We are interested in your first impressions, so please don’t spend an excessive amount of time on each question. 123 INSTRUCTIONS In the attached questionnaire, we ask questions which make use of rating scales with seven places; you are asked to place an ‘X’ in the place that best describes your opinion. For example, if you were asked to rate “Driving a car in winter is easy” on such a scale, it would appear as follows: Driving a car in winter is easy. likely I I I I I I unlikely extremely quite slightly neither slightly quite extremely If you think that it is extremely likely that driving a car in winter is easy, you would make your mark as follows: Driving a car in winter is easy. likely L x I I I I I unlikely extremely quite slightly neither slightly quite extremely If you think that it is neither likely nor unlikely that driving a car in winter is easy, you would make your mark as follows: Driving a car in winter is easy. likely I I I X f unlikely extremely quite slightly neither slightly quite extremely In addition to the “likely-unlikely” pairs, other pairs such as “disagree-agree” will also be used. They should be answered in the same fashion. In making your ratings, please remember the following points: 1. Place your marks in the middle of spaces, NOT ON TILE BOUNDARIES. likely I X I unlikely extremely quite slightly neither slightly quite extremely THIS NOT ThIS 2. Never put more than one ‘X’ on a single answer line. One other question format will be used. In this case, you will be asked to circle a number or letter corresponding to a particular answer for a question. Please be careful to see that your circle goes around only the letter or number which corresponds to your desired response. 124 TO BEGIN, WE WOULD LIKE TO ASK YOU ABOUT YOUR EXPERIENCE WITH. COMPUTERS AND OTHER HIGH-TECHNOLOGY PRODUCTS AND SERVICES. A-i Have you ever used a multi-function telephone (including such functions as call forward, speed dialing, call waiting, etc.). (Place an ‘X’ beside the appropriate answer): ___ NO YES If yes, which functions do you use? (Place an ‘X’ beside the appropriate functions): CALL TRANSFER (CONSULTATIONS) HOLD THREE-WAY CONFERENCE CALL FORWARDING CALL PARKING - CALL PICKUP CALL WAITING RING AGAIN/AUTOMATIC CALL BACK SPEED CALLING LAST NUMBER DIALLED SAVE NUMBER AND REPEAT 125 A-2 How often do you use the products listed below? (Place an ‘X’ under the appropriate column for each applicable area): About 1- More Less than 3 Times About 2- About than Once Not at Once per per Once per 4 Times Once per per Day All Month Month Week per Week Day a. Automated Teller Machine b. Programmable Calculators c. Home Computers d. Business_Computers c. Video Gaines f. Programmable Microwave Ovens A-3 How often do you carry out the computer-related activities listed below; on paper, via electronic mail, on floppy disk, etc...? (Place an ‘X’ under the appropriate column for each applicable area): About 1- Less than 3 Times About 2- About More Not at Once per per Once per 4 Times Once per than Once all Month Month Week per Week Day per Day Receive computer output (reports/documents) Submit documents, etc. to others for word processing Submit data to others for computer analysis 126 A-4 What is your current keyboarding (typing) ability? a. Mark with an ‘X’: ___ HUNT & PECK TOUCH TYPE b. Place an ‘X’ in the place that best reflects your speed I I I I I I I I Owpm 1-15 16-30 31-45 46-60 61-75 >75 wpm wpm wpm wpm wpm wpm A-5 How many educational courses (at any level) have you had about computers, but which did not include your personal hands-on use? (eg: “theory” courses) ________ COURSES A—6 How many educational courses have you had which required your personal hands-on use of computers? (eg: “applied” courses) COURSES A-7 My firm receives non-computer support for the following areas (place an ‘X’ under the appropriate column for each applicable area): none constant 1 2 3 some 5 6 7 Accounting Audit BUSIneSS Advice Financial Planning Gov’t Compliance Marketing Tax Other (please specify) 127 A-8 My firm receives non-computer support from the following sources external to the firm (place an ‘X’ under the appropriate column for each applicable source): none constant 1 2 3 sOme 5 6 7 Personal friend (non employee) Public accounting firm Non-Accountant computer consultant None Other (please specify) A-9 I am satisfied with the current level of support for non-computer areas I receive from the following sources external to the firm (place an ‘X’ under the appropriate column for each applicable source): satisfied unsatisfied extremely quite slightly neither slightly quite f extremely Personal friend (non employee) Public accounting firm Non-Accountant computer consultant None Other (please specify) A-b How much access to the use of a CAS do you feel you currently have? un.li.nüted I I I I I I extremely quite slightly neither slightly quite extremely 128 A-il How knowledgeable do you feel you are of the uses of the CAS? unlimited I I I I I extremely quite slightly neither slightly quite extremely A-12 Have you ever used a CAS? (Place an ‘X’ beside the appropriate column): ___ CURRENTLY USE A CAS Please go on to the next page HAVE NEVER USED A CAS Please go on to the next page USED TO USE A CAS BUT NO LONGER DO SO Please answer A-13 to A-iS only if you used to use a CAS but no longer do. A-13 Could you please indicate approximately when you first began to use a CAS, and when you stopped using it. STARTED ______ ______ MONTh YEAR STOPPED MONTH YEAR A-14 Please indicate which of the CAS functions below you used by indicating the number of months you used them. Accounting Graphics Information Report Spreadsheet Statistical Text/word Other Software Generation Retrieval Generation Analysis Processing (please spccit’) MONTHS A-i5 Could you please indicate very briefly why you no longer use the CAS. Please go on to the next page 129 FIRST WE WOULD LIKE TO GET YOUR IMPRESSIONS OF THE CAS. IN THE FOLLOWING, WE WILL PRESENT YOU WITH A NUMBER OF STATEMENTS EXPRESSING PARTICULAR VIEWPOINTS ABOUT THE CAS. WE WOULD LIKE YOU TO INDICATE HOW MUCH EACH STATEMENT REFLECTS YOUR PERSONAL VIEWPOINT BY PLACING AN ‘X’ IN THE APPROPRIATE PLACE ON THE DISAGREE-AGREE SCALES PROVIDED. ALTHOUGH THERE MAY APPEAR TO BE A NUMBER OF SIMILAR STATEMENTS, PLEASE PROVIDE A RESPONSE TO EACH ONE. U-i Using a CAS enables me to accomplish tasks more quickly. disagree I I I I I I _i agree strongly quite slightly neither slightly quite strongly U-2 Using a CAS is completely compatible with my current situation. disagree I I I I I I I agree strongly quite slightly neither slightly quite strongly U-3 Using a CAS is compatible with all aspects of my work. disagree I I I I I I strongly quite slightly neither slightly quite strongly U-4 My superiors expect me to use a CAS. disagrecl I I I I I I agree strongly quite slightly neither slightly quite strongly U-S I believe that a CAS is cumbersome to use. disagreel I I I I I I agree strongly quite slightly neither slightly quite strongly U-6 Using a CAS improves my image within the organization. disagreel I I I I I I agree strongly quite slightly neither slightly quite strongly 130U-7 Using a CAS improves the quality of work I do. disagree I I I I I I I agree strongly quite slightly neither slightly quite strongly U-8 Using a CAS makes it easier to do my job. dLsagreel I I I I I I agree strongly quite slightly neither slightly quite strongly U-9 I think that using a CAS fits well with the way I like to work. disagree I I I I I I iaglee strongly quite slightly neither slightly quite strongly U-lO My use of a CAS is voluntary (as opposed to required by my superiors or job description). disagrecl I I I I I I I agree strongly quite slightly neither slightly quite strongly U-li I have seen what others do using their CAS. disagree I I I I I I I agree strongly quite slightly neither slightly quite strongly U-12 I’ve had a great deal of opportunity to try various CAS applications. disagree I I I I I I I agree strongly quite slightly neither slightly quite strongly U-13 In my organization, one sees CAS on many desks. disagree I I I I I I agree strongly quite slightly neither slightly quite strongly U-14 My boss does not require me to use a CAS. disagreel I I I I I I agree strongly quite slightly neither slightly quite strongly 131 U-15 I would have no difficulty telling others about the results of using a CAS. disagreci I I I I I I I agree strongly quite slightly neither slightly quite strongly U-16 I know where I can go to satisfactorily try out various uses of a CAS. disagreci I I I I I I agree strongly quite slightly neither slightly quite strongly U-17 People in my organization who use a CAS have more prestige than those who do not. disagree I I I I I I rec strongly quite slightly neither slightly quite strongly U-18 Although it might be helpful, using a CAS is certainly not compulsory in my job. disagreel I I I I I jagree strongly quite slightly neither slightly quite strongly U19 My using a CAS requires a lot of mental effort. disagreel I I I I agree strongly quite slightly neither slightly quite strongly U-20 Using a CAS is often frustrating. disagree I I I I I I agree strongly quite slightly neither slightly quite strongly U-21 People in my organization who use a CAS have a high profile. disagree I I agree strongly quite slightly neither slightly quite strongly U-22 A CAS was available to me to adequately test run various applications. disagreej I I I I I I I strongly quite slightly neither slightly quite strongly 132 U-23 I believe I could communicate to others the consequences of using a CAS. disagree [ strongly quite slightly neither U-24 I believe that it is easy to get a CAS to do what I disagreel I I I strongly quite slightly U-25 Overall, I believe that a CAS is easy to use. disagree I I I strongly quite slightly neither U-26 Using a CAS improves my job performance. disagreeL I I I slightly quite strongly want it to do. I I I I I I I I I I I I U-29 Before deciding whether to use any CAS applications, I was able to properly try them out. strongly quite slightly U-30 Learning to operate a CAS is easy for me. I I I I I I neither slightly quite strongly slightly quite strongly agree agree agree agree agree agree strongly quite slightly neither slightly quite strongly U27 CAS are not very visible in my organization. disagreel I I I strongly quite slightly neither slightly quite strongly U-28 Overall, I find using a CAS to be advantageous to my job. disagreel j I I I strongly quite slightly neither slightly . I I quite strongly disagree I I I I I disagree I neither slightly quite strongly I I I I jagree j agree strongly quite slightly neither slightly quite strongly 133 U-3 1 Using a CAS enhances my effectiveness on the job. disagrec[ I I I I I I lagree strongly quits slightly neither slightly quite strongly U-32 Using a CAS fits into my work style. disagrcc I I I I I I I strongly quits slightly neither slightly quite strongly U-33 I would have difficulty explaining why a CAS may or may not be beneficial. disagrec I I I I I I agree strongly quits slightly neither slightly quits strongly U-34 I was permitted to use a CAS on a trial basis long enough to see what it could do. disagreel I I I I I I I agree strongly quite slightly neither slightly quits strongly U-35 Using a CAS gives me greater control over my work. disagrec I I I I 1 I 1 agree strongly quite slightly neither slightly quits strongly U-36 Using a CAS increases my productivity. disagreel I I I I I I agree strongly quite slightly neither slightly quite strongly U-37 Having a CAS is a status symbol in my organization. disagree I I I I I I agree strongly quite slightly neither slightly quits strongly U-38 It is easy for me to observe others using CAS in my firm. disagrecl I I I I I lagree strongly quits slightly neither slightly quits strongly U-39 I have had plenty of opportunity to see the CAS being used. disagreel I I I I I I strongly quite slightly neither slightly quite strongly 134 FINALLY, IN THIS SECTION WE WOULD LIKE TO ASK YOU A FEW QUESTIONS ABOUT YOUR USE OF THE CAS. B-i Overall, my using a CAS in my job is (place an X on all four scales): wme negative extremely quite slightly neither slightly quite extremely foolish po8itwe B-2 Assuming that any decision to use the CAS is totally up to you, how would you rate your potential use of the CAS in the next six months? likely I I I I I I I unlikely extremely quite slightly neither slightly quite extremely improbable probable goodi I I I I I I I hannfiil extremely quite slightly neither slightly quite extremely I I I I I I I extremely quite slightly neither slightly quite extremely bad beneficial I I I I I I I extremely quite slightly neither slightly quite extremely I I I I I I I I extremely quite slightly neither slightly quite extremely 135 B-3 Approximately when (month and year) did you first start using a CAS beyond any trial of it you may have carried out? MONTH YEAR B-4 How regularly do you now use a CAS? (Place an ‘X’ under the appropriate column): Less than About 1-3 About About More than once per times per once per 2-4 times once per once per Not at all month month week per week day day I I I B-S On average, how frequently do you currently use the following functions (place an ‘X’ under the appropriate column): Less About than 1-3 About 2-4 About More once times once times once than Not at per per per per per once all month month week week day per day Accounting Software Graphics Generation Information Retrieval Report Generation Spreadsheet Statistical Analysis Text/word Processing Other (please specify) 136 B-6 For each of the following questions, place an ‘X’ under the appropriate column for each applicable function: a. On average how many hours per week do you spend using the CAS on the following functions? b. Please indicate approximately how long (in months) you have been regularly using any of the following functions. Other Accounting Graphics Information Report Statisi.ical Texilword (pleasc Software Generation Retneval Generation Spreadsheet Analysis Processing pecify) HOURS MONTHS B-7 Overall, how do you expect your usage of the CAS will change in the six months? (Place an ‘X’ in the appropriate column): increasel I I I I I I I decrease signifi- some- marginally same marginally some- signifi cantly what what cantly B-8 Overall, how has your usage of CAS changed in the last six months? (Place an ‘X’ in the appropriate column): increasedi I I I I I I I decreased aigniii- some- marginally same marginally some- signifi cantly what what cantly B-9 I have been using a CAS for (place an ‘X’ under the appropriate column): Less than About 1-3 About About More than once per times per once per 2-4 times once per once per Not at all month month week per week day day 137 B-1O When I started using my CAS, I received continuing support (training or help) for my CAS from the following sources (place an ‘X’ under the appropriate column for each source): About 1- More Less than 3 Time. About 2- About than Once Not at Once per per Once per 4 Times Once per per Day All Month Month Week per Week Day Other personnel from my company Personal friend (non employee) Public accounting firm Non-Accountant computer consultant Other (please specify) B-il I currently receive continuing support (training or help) for my CAS from the following sources (place an ‘X’ under the appropriate column for each source): About 1- More Less than 3 Times About 2- About than Once Not at Once per per Once per 4 Times Once per per Day All Month Month Week per Week Day Other personnel from my company Personal friend (non employee) Public accounting firm Non-Accountant computer consultant Other (please specify) 138 B-12 Currently, if I need help with my CAS, I know I can get support from the following sources (place an ‘X’ under the appropriate column for each source): none ongoing 1 2 J 3 some 5 6 Other personnel from my company Personal friend (non-employee) Public accounting finn Non-Accountant computer consultant Other (please specify) B-13 I plan on getting my future CAS help from the following sources (place an ‘X’ under the appropriate column for each applicable source): none ongoing 1 2 3 some 5 6 7 Other personnel from my company Personal friend (non-employee) Public accounting finn Non-Accountant computer consultant Other (please specify) 139 B-14 I am satisfied with the current level of continuing support for my CAS that I receive from the following sources (place an ‘X’ under the appropriate column for each applicable source): satisfied unsatisfied N/A (Don’t receive any auppoit) extremely quite slightly neither slightly quite extremely Other personnel from my company Personal friend (non employee) Public accounting firm Non-Accountant computer consultant Other (please specify) B-15 How effective do you feel the following were in helping you to get started in your use of a CAS? (Place an ‘X’ under the appropriate column for each applicable source): effective ineffective extremely quite slightly neither slightly quite extremely Other personnel from my company Personal friend (non- employee) Public_accounting_firm Non-Accountant computer consultant Other (please specify) 140 B-16 How effective do you feel the following have been in helping you in your current use of a CAS? (Place an ‘X’ under the appropriate column for each applicable source): effective ineffective extremely quite slightly neither slightly quite extremely Other personnel from my_company Personal friend (non employee) Public accounting firm Non-Accountant computer consultant Other (please specify) THANK YOU FOR YOUR PERSEVERANCE AND COOPERATION SO FAR. NOW, PLEASE GO ON TO THE NEXT PAGE. 141 IN THE FOLLOWING, WE WILL PRESENT YOU WITH A NUMBER OF STATEMENTS EXPRESSING PARTICULAR VIEWPOINTS ABOUT THE CAS. WE WOULD LIKE YOU TO INDICATE HOW MUCH EACH STATEMENT REFLECTS YOUR PERSONAL VIEWPOINT PLACING AN ‘X’ IN THE APPROPRIATE PLACE ON ThE DISAGREE- AGREE SCALE. ALTHOUGH THERE MAY APPEAR TO BE A NUMBER OF SIMILAR STATEMENTS, PLEASE PROVIDE A RESPONSE TO EACH ONE. N-i Using a CAS would enable me to accomplish tasks more quickly. disagreci I I I I I strongly quits slightly neither slightly quite strongly N-2 Using a CAS would improve the quality of work I do. disagree I I I I I I I agree strongly quits slightly neither slightly quite strongly N-3 Using a CAS would be compatible with all aspects of my work. disagree I I I I I I agree strongly quits slightly neither slightly quite strongly N-4 My superiors expect me to use a CAS. disagrecl I I I I I I ] agree strongly quits slightly neither slightly quits strongly N-5 I believe that a CAS would be cumbersome to use. disagree 1 I I I I I agree strongly quits slightly neither slightly quits strongly N-6 Using a CAS would improve my image within the organization. disagrecf I I I I I I I agree strongly quite slightly neither slightly quite strongly N-7 Using a CAS would be completely compatible with my current situation. disagreei I I I I I I agree strongly quits slightly neither slightly quits strongly 142 N-8 Using a CAS would make it easier to do my job. disagrecj I I I I I I strongly quits slightly neither slightly quite strongly N-9 I think that using a CAS would fit well with the way I like to work. disagreci I I I I I I I agree strongly quits slightly neither slightly quite strongly N-IO My use of a CAS is voluntary (as opposed to required by my superiors or job description). disagrcc I I I I I I I agree strongly quite slightly neither slightly quite strongly N-Il I have seen what others do using their CAS. disagreel I I I I I I I agree strongly quits slightly neither slightly quite strongly N-12 I’ve had a great deal of opportunity to try various CAS applications. disagree I I I I I agree strongly quits slightly neither slightly quits strongly N-13 In my organization, one sees CAS on many desks. disagree I I I I I I agree strongly quite slightly neither slightly quits strongly N-14 My boss does not require me to use a CAS. disagree I I I I I I agree strongly quits slightly neither slightly quits strongly N-15 I would have difficulty telling others about the results of using a CAS. disagree I I I I I I agree strongly quite slightly neither slightly quite strongly 143N-16 I know where I can go to satisfactorily try out various uses of a CAS. disagrec I I I I I I agree strongly quite slightly neither slightly quite strongly N-17 People in my organization who use a CAS have more prestige than those who do not. disagreel I I I I I I I strongly quite slightly neither slightly quite strongly N-18 Although it might be helpful, using a CAS is certainly not compulsory in my job. disagrees I I I I I agree strongly quite slightly neither slightly quite strongly N-19 My using a CAS would require a lot of mental effort. disagrcc I I I I I I I agree strongly quite slightly neither slightly quite strongly N-20 Using a CAS would often be frustrating. disagreel I I I I I I I agree strongly quite slightly neither slightly quite strongly N-21 People in my organization who use a CAS have a high profile. disagree I I I I I I agree strongly quite slightly neither slightly quite strongly N-22 A CAS is available to me to adequately test run various applications. disagreel I I I I I I agree strongly quite slightly neither slightly quite strongly N-23 I believe I could communicate to others the consequence of using a CAS. disagree I I I I I agree strongly quite slightly neither slightly quite strongly 144 N-24 I believe that it would be easy to get a CAS to do what I want it to do. disagreci I I I I I I agree strongly quite slightly neither slightly quite strongly N-25 Overall, I believe that a CAS would be easy to use. diaagreei I I I I I I I re strongly quite slightly neither slightly quite strongly N-26 Using a CAS would improve my job performance. disagree I I I I I I I agree strongly quite slightly neither slightly quite strongly N-27 CAS are not very visible in my organization. disagree I I I I I I I agree strongly quite slightly neither slightly quite strongly N-28 Overall, I would find using a CAS to be advantageous in my job. disagreci I I I I I I strongly quite slightly neither slightly quite strongly N-29 Before deciding whether to use any CAS applications, I would be able to properly try them out. disagree I I I I I I agree strongly quite slightly neither slightly quite strongly N-30 Learning to operate a CAS would be easy for me. disagree I I I I I I agree strongly quite slightly neither slightly quite strongly N-31 Using a CAS would enhance my effectiveness on the job. disagree[ I I I I I I lagrec strongly quite slightly neither slightly quite strongly 145 N-32 Using a CAS would fit into my work style. disagreel I I I I strongly quits slightly neither slightly quite strongly N-33 If I were to use a CAS, I would have difficulty explaining why using a CAS may or may not be beneficial. disagree I I I I I agree strongly quits slightly neither slightly quite strongly N-34 I would be permitted to use a CAS on a trial basis long enough to see what it could do. disac I I I I I strongly quite slightly neither slightly quite strongly N-35 Using a CAS would give me greater control over my work. disagreel I I I strongly quite slightly neither slightly quite strongly N-36 Using a CAS would increase my productivity. disagreci I I I I strongly quite slightly neither slightly quite strongly N-37 Having a CAS is a status symbol in my organization. disagreel I I I I quite strongly I I I strongly quite slightly neither slightly N-38 It is easy for me to observe others using a CAS in my firm. disagreci I — I I I strongly quits slightly neither slightly N-39 I have had plenty of opportunity to see the CAS being used. disagree I I I I I I agree I jagree jagree jagree agree agree agree quite strongly strongly quite slightly neither slightly quite strongly 146 FINALLY, IN THIS SECTION WE WOULD LIKE TO ASK A FEW GENERALQUESTIONS. C-I Overall, my using a CAS in my job would be (place an X on all four scales): I I I I I I I ‘‘ extremely quite slightly neither slightly quite extremely hannfulf I I I I I I extremely quite slightly neither slightly quite extremely wme I I I I I foolish extremely quite slightly neither slightly quite extremely negative L__.._. I I I I I I positive extremely quite slightly neither slightly quite extremely C-2 Assuming that any decision to use the CAS is totally up to you, how would you rate yourpotential use of the CAS in the next six months? likely unlikely extremely quite slightly neither slightly quite extremely improbable probable extremely quite slightly neither slightly quite extremely 147 C-3 Approximately how often in the past have you gone to the following for help in using a CAS? (Place an ‘X’ under the appropriate column for each applicable source): Less About About than 1-3 2-4 More Once Times Once Times About than Did not Notat per per per per Once Once Use AU Month Month Week Week per Day per Day CAS Other personnel from my company Persona] friend (non employee) Public accounting finn Non-Accountant computer consultant Other (please specify) C-4 How effective do you feel the following were in helping you use a CAS? (Place an ‘X’ under the appropriate column for each applicable source): Less About About than 1-3 2-4 More Once Times Once Times About than Did not Notat per per per per Once Once Use A]] Month Month Week Week per Day per Day CAS Other personnel from my company Personal friend (non- employee) Public accounting firm Non-Accountant computer consultant Other (please specify) 148 C-.5 Identify your SUPPORT GROUP, whose official function it would be to support you in the CAS (if more than one choose the primary source of help); (place an ‘X’ under the appropriate column for each applicable source): Other personnel Non-accountant from my Professional computer company Personal friend accounting firm consultant None THANK YOU FOR YOUR PERSEVERANCE AND COOPERATION SO FAR. NOW, PLEASE GO ON TO THE NEXT PAGE. 149 In this last section, we would like to ask you some questions about yourself. Remember, all answers are confidential, and no respondent can be identified, so please give as candid a response as possible. FIRST, WE WOULD LIKE YOU TO ONCE AGAIN INDICATE AGREEMENT OR DISAGREEMENT WITH A NUMBER OF STATEMENTS; THIS TIME ABOUTYOURSELF. PLEASE PLACE AN ‘X’ IN THE APPROPRIATE SPACE. I-i I am generally cautious about accepting new ideas. disagrec I I I I I 1ag strongly quite slightly neither slightly quite strongly 1-2 I rarely trust new ideas until I can see whether the vast majority of people around me accept them. disagreei I I I I .1 I strongly quite slightly neither slightly quite strongly 1-3 1 am aware that I am usually one of the last people in my group to accept something new. disagree I I I I I agree strongly quite slightly neither slightly quite strongly 1-4 1 am reluctant about adopting new ways of doing things until I see them working for people around me. disagree I I I I I I agree strongly quite slightly neither slightly quite strongly 1-5 1 find it stimulating to be original in my thinking and behaviour. disagreel I I I I I I agree strongly quite slightly neither slightly quite strongly 1-6 I tend to feel that the old way of living and doing things is the best way. disagree I I I I I I J agree strongly quite slightly neither slightly quite strongly 150 1-7 I am challenged by ambiguities and unsolved problems. disagreef I I I I I I f agec strongly quite slightly neither slightly quite strongly 1-8 I must see other people using new innovations before I will consider them. aisagrecf I I I I I I (agree strongly quite slightly neither slightly quite strongly 1-9 1 am challenged by unanswered questions. disagree I I I I I I I agree strongly quite slightly neither slightly quite strongly 1-10 1 often find myself sceptical of new ideas. disagree I I I I I I I agree strongly quite slightly neither slightly quite strongly NEXT, WOULD YOU PLEASE INDICATE HOW LIKELY OR UNLIKELY EACH OF THE FOLLOWING STATEMENTS ARE BY ONCE AGAIN PLACING AN ‘X’ IN THE AIPROPRIATE SPACE. S-i Most people who are important to me think I should use the CAS in my job. likely unlikely extremely quite slightly neither slightly quite extremely S-2 My close friends think that I should use the CAS in my job. likely I I I I I I unlikely extremely quite slightly neither slightly quite extremely S-3 My co-workers (peers) think that I should use the CAS in my job. likely I I I I unlikely extremely quite slightly neither slightly quite extremely 151 S-4 My immediate superiors think that I should use the CAS in my job. likelyl I I I I I I unlikely extremely quite slightly neither slightly quite extremely S-5 Senior management thinks that I should use the CAS in my job. lik.eIyI I I I I I I I unlikely extremely quite slightly neither slightly quite extremely S-6 My subordinates think I should use the CAS in my job. likelyl I I I I I I I uldly extremely quite slightly neither slightly quite extremely S-7 Generally speaking, I want to do what most people who are important to me think I should do. likely unlikely extremely quite slightly neither slightly quite extremely S-8 Generally speaking, I want to do what my close friends think I should do. likely I I I I I I unlikely extremely quite slightly neither slightly quite extremely S-9 Generally speaking, I want to do what my co-workers think I should do. 1iicly I I I I 1 I I unlilcely extremely quite slightly neither slightly quite extremely S-lO Generally speaking, I want to do what my immediate supervisors think I should do. likely I I I I I I J unlikely extremely quite slightly neither slightly quite extremely S-li Generally speaking, I want to do what senior management thinks I should do. likely I I I I unlikely extremely quite slightly neither slightly quite extremely S-12 Generally speaking, I want to do what my subordinates think I should do. 152 likely I I I I I I I unlikely extremely quite 8lghtly neither 8lightly quite extremely FINALLY, WE WOULD LIKE TO ASK A FEW QUESTIONS ABOUT YOURSELF [OR STATISTICAL PURPOSES. COULD YOU PLEASE INDICATE: P-i Your sex (place an ‘X’ beside the appropriate column): MALE FEMALE P-2 Your present age: years P-3 Your department: P4 Your job title: ______________________________________________ P-5 Years you have worked in your current department: years P-6 Years you have worked in this company: years P-7 What is the highest level of education that you completed? Place an ‘X’ beside the appropriate column): GRADE SCHOOL SOME HIGH SCHOOL HIGH SCHOOL GRADUATE ___ SOME TECHNICAL COLLEGE TECHNICAL COLLEGE GRADUATE SOME COMMUNITY COLLEGE COMMUNITY COLLEGE GRADUATE SOME UNIVERSITY UNIVERSITY GRADUATE POSTGRADUATE 153 P-8 The job that best describes my organizational level is (place an ‘X’ beside the appropriate column): ___ EXECUTIVE/TOP MANAGEMENT MIDDLE MANAGEMENT SUPERVISORY PROFESSIONAL TECHNICAL __ CLERICAL OTHER (please specify) GENERAL BUSINESS INFORMATION F-i Number of: Employees Accounting Staff Full Time Part Time F-2 Annual sales last year: < $250,000 < $500,000 < $1,000,000 c $10,000,000 > $10,000,000 F-3 Type of organization (e.g., profit, non-profit, CO-OP. etc...) F-4 What industry does your firm operate in? INDUSTRY:_____________________ F-5 Does your firm plan to implement or expand a CAS in the next two years? YES NO DON’T KNOW If YES, approximately how much do you expect your firm to spend on the CAS in this time? $______________ 154 F-6 THANK YOU VERY MUCH FOR YOUR PARTICWATION! If you wish to add any comments or further observations, please use the space below or simply attach them to this page. 1 c I ._J Appendix II- A2 Pilot Study 156 WELCOME! You are about to participate in a study of opinions about the usage of microcomputers in the accounting function. In some sections you will be asked questions about and see reference to the term CAS, which stands for COMPUTERIZED ACCOUNTING SYSTEM. A CAS is defined for the purposes of this study as a set of computerized tools for an individual, and usually consists of a personal or microcomputer with one or more software packages, such as an accounting program, and/or other software such as spreadsheet, database, word processing, etc. in support of the accounting function. The key aspect of a CAS is that it is computer technology that you would use directly, as opposed to having someone else use for you. In completing the questionnaire, please remember: 1. All the information you give is kept confidential. 2. We need answers to all questions. Please don’t skip any. 3. Be honest - tell it like it is. 4. Please don’t talk to others about how you respond to the questions. We would like your opinion, not the opinion of your associates. 5. Even if you have never used a CAS, please answer all the relevant questions as best as you can. 6. Move rapidly through the questionnaire. We are interested in your first impressions, so please don’t spend an excessive amount of time on each question. 157 INSTRUCTIONS In the attached questionnaire, we ask questions which make use of rating scales with seven places; you are asked to place an ‘X’ in the place that best describes your opinion. For example, if you were asked to rate “Driving a car in winter is easy” on such a scale, it would appear as follows: Driving a car in winter is easy. likely I I I I I I I unlikely extremely quite slightly neither slightly quite extremely If you think that it is extremely likely that driving a car in winter is easy, you would make your mark as follows: Driving a car in winter is easy. likely X I I I I unlikely extremely quite slightly neither slightly quite extremely If you think that it is neither likely nor unlikely that driving a car in winter is easy, you would make your mark as follows: Driving a car in winter is easy. likely f I I I X I I I unlikely extremely quite slightly neither slightly quite extremely In addition to the “likely-unlikely” pairs, other pairs such as “disagree-agree” will also be used. They should be answered in the same fashion. In making your ratings, please remember the following points: 1. Place your marks in the middle of spaces, NOT ON TIlE BOUNDARIES. likely I X unlikely extremely quite slightly neither slightly quite extremely TillS NOT TillS 2. Never put more than one ‘X’ on a single answer line. One other question format will be used. In this case, you will be asked to circle a number or letter corresponding to a particular answer for a question. Please be careful to see that your circle goes around only the letter or number which corresponds to your desired response. 158SECTION A TO BEGIN, WE WOULD LIKE TO ASK YOU ABOUT YOUR EXPERIENCE WITH COMPUTERS AND OTHER HIGH-TECHNOLOGY PRODUCTS AND SERVICES. A-i Have you ever used a multi-function telephone (including such functions as call forward, speed dialing, call waiting, etc.). (Place an ‘X’ beside the appropriate answer): NO )L ys If you use a multi-function phone, which functions do you use? (Place an ‘X’ beside the appropriate functions): CALL TRANSFER (CONSULTATIONS) HOLD THREE-WAY CONFERENCE CALL FORWARDING CALL PARKING ___ CALL PICKUP CALL WAiTING RING AGAIN/AUTOMATIC CALL BACK % SPEED CALLING X LAST NUMBER DIALLED SAVE NUMBER AND REPEAT 159 A-2 How often do you use the products listed below? (Place an ‘X’ under the appropriate column for each applicable area): About 1- More Less than 3 Times About 2- About than Once Not at Once per per Once per 4 Times Once per per Day All Month Month Week per Week Day a. Automated Teller Machine b. Programmable Calculators c. Home Computers d. Business Computers > c. Video Games f. Programmable Microwave Ovens A-3 How often do you carry out the computer-related activities listed below; on paper, via electronic mail, on floppy disk, etc...? (Place an ‘X’ under the appropriate column for each applicable area): About 1- Less than 3 Times About 2- About More Not at Once per per Once per 4 Times Once per than Once all Month Month Week per Week Day per Day Receive computer output x(reports/documents) Submit documents, etc. to X others for word processing Submit data to others for computer analysis 160 A-4 What is your current keyboarding (typing) ability? a. Mark with an ‘X’: ___ HUNT & PECK < TOUCH TYPE b. Place an ‘X’ in the place that best reflects your speed I I I I>I I I I Owpm 1-15 16-30 31-45 46-60 61-75 >75 wpm wpm wpm wpm wpm wpm A-5 How many educational courses (at any level) have you had about computers, but which did not include your personal hands-on use? COURSES A-6 How many educational courses have you had which required your personal hands-on use of computers? COURSES A-7 My firm receives non-CAS support for the following areas (place an ‘X’ under the appropriate column for each applicable area): none constant 1 2 3 j some 5 6 7 Accounting Audit Business Advice Financial Planning Gov’t Compliance — Marketing Tax X Other (please specify) 161 A-8 My firm receives non-CAS support from the following sources external to the firm (place an ‘X’ under the appropriate column for each applicable source): none constant 1 2 3 1 some 5 6 Personal friend (non employee) Public accounting finn Non-Accountant computer consultant None Other (please specify) A-9 I am satisfied with the current level of support for non-CAS areas I receive from the following sources external to the firm (place an ‘X’ under the appropriate column for each applicable source): satisfied unsatisfied extremely quite slightly neither slightly quite extremely Personal friend (non cmploy) Public accounting finn Non-Accountant computer consultant None Other (please specify) A-1O How much access to the use of a CAS do you feel you currently have? unlimited I I I I I I hunted extremely quite slightly neither slightly quite extremely 162 A-i 1 How knowledgeable do you feel you are of the uses of the CAS? unlimited I < I I I I I limited extremely quite slightly neither slightly quite extremely A-12 Have you ever used a CAS? (Place an ‘X’ beside the appropriate column): )( CURRENTLY USE A CAS_______________ PLEASE SKIP TO SECTION B HAVE NEVER USED A CAS PLEASE SKIP TO SECTION C USED TO USE A CAS BUT NO LONGER DO SO Please answer A-13 to A-is only if you used to usea CAS but no longer do. A-13 Could you please indicate approximately when you first began to use a CAS, and when you stopped using it. STARTED ______ ______ MONTH YEAR STOPPED MONTH YEAR A-i4 Please indicate which of the CAS functions below you used by indicating the number of months you used them. Accounting Graphics Information Report Spreadsheet Statistical Textlword Other Software Generation Retrieval Generation Analysis Processing (please specify) MONTHS[ A-i5 Could you please indicate very briefly why you no longer use the CAS. PLEASE SKIP TO SECTION C SECTION B 163 Please answer questions in this section only if you currently use the CAS. FIRST WE WOULD LIKE TO GET YOUR IMPRESSIONS OF THE CAS. IN THE FOLLOWING, WE WILL PRESENT YOU WITH A NUMBER OF STATEMENTS EXPRESSING PARTICULAR VIEWPOINTS ABOUT THE CAS. WE WOULD LIKE YOU TO INDICATE HOW MUCH EACH STATEMENT REFLECTS YOUR PERSONAL VIEWPOINT BY PLACING AN ‘X’ IN THE APPROPRIATE PLACE ON THE DISAGREE-AGREE SCALES PROVIDED. ALTHOUGH THERE MAY APPEAR TO BE A NUMBER OF SIMILAR STATEMENTS, PLEASE PROVIDE A RESPONSE TO EACH ONE. U-i Using a CAS enables me to accomplish tasks more quickly. disagreci I I I I I I strongly quite slightly neither slightly quite strongly U-2 Using a CAS is completely compatible with my current situation. disagreej I I I I I ><‘ I iag strongly quite slightly neither slightly quite strongly U-3 Using a CAS is compatible with all aspects of my work. disagree I I I I I X agree strongly quitc slightly neither slightly quite strongly U-4 My superiors expect me to use a CAS. 1 disagrce I I I I I I agree strongly quite slightly neither slightly quite strongly U-5 I believe that a CAS is cumbersome to use. I disagree X I I I I I agree strongly quite slightly neither slightly quite strongly U-6 Using a CAS improves my image within the organization. disagree I I I agree strongly quite slightly neither slightly quite strongly 164 dagree I I I I strongly quite slightly neither slightly U-9 I think that using a CAS fits well with the way I like to work. disagrec I I I I I strongly quite slightly neither slightly quite U-lO My use of a CAS is voluntary (as opposed to required by my superiors disagreel I I <I I I I strongly quite slightly U-il I have seen what others do using their CAS. disagreci I U-12 U-7 Using a CAS improves the quality of work I do. disagrec I I I strongly quite slightly U-8 Using a CAS makes it easier to do my job. I - neither slightly quite j agree agree strongly x quite strongly >( agrec strongly or job description). agree neither slightly quite strongly U-13 . lxi strongly quite slightly neither slightly quite strongly I’ve had a great deal of opportunity to try various CAS applications. disagrec I I I I strongly quite slightly neither slightly quite strongly In my organization, one sees CAS on many desks. disagree I I agree agree agree agree strongly quite slightly neither slightly quite strongly U-14 My boss does not require me to use a CAS. strongly disagree)<} I I I I I I quite slightly neither slightly quite strongly II U-17 I would have no difficulty telling others about the results of using a CAS. disagrecf I I I I I I strongly quite slightly neither slightly quite strongly I know where I can go to satisfactorily try out various uses of a CAS. disagree strongly quite slightly neither slightly quite strongly People in my organization who use a CAS have more prestige than those who do not. disagrcc I I I I I I strongly quite slightly neither slightly quite strongly U-18 Although it might be helpful, using a CAS is certainly not compulsory in my job. disagree ‘< I I I I I I I agree strongly quite slightly neither slightly quite strongly U-19 My using a CAS requires a lot of mental effort. ‘xdisagree strongly quite slightly neither slightly quite strongly Using a CAS is often frustrating. I I disagree I I I I I strongly quite slightly neither slightly quite strongly People in my organization who use a CAS have a high profile. disagree strongly quite slightly neither slightly quite strongly A CAS was available to me to adequately test run various applications. disagrees I I I strongly quite slightly neither slightly quite strongly agree U-15 U-16 -1 165 agree agree agree U-20 U-21 U-22 agree agree agree 166 U-23 I believe I could communicate to others the consequences of using a CAS. xdisagree strongly quite slightly neither slightly quite strongly U-24 I believe that it is easy to get a CAS to do what I want it to do. disagree _____________________________________________________ strongly quite slightly neither slightly U-25 Overall, I believe that a CAS is easy to use. disagree [__L I I I I I>cI I agree strongly quite slightly neither slightly quite strongly U-26 Using a CAS improves my job performance. disagreci I I I strongly quite slightly neither U-27 CAS are not very visible in my organization. disagreei ‘<‘ I I I I strongly quite slightly neither slightly quite strongly U-28 Overall, I find using a CAS to be advantageous to my job. disagreef I I I I I U-29 I agree agree quite strongly slightly quite strongly Jagrcc jagre agree U-30 strongly quite slightly neither slightly quite strongly Before deciding whether to use any CAS applications, I was able to properly try them out. disagree f I > I I I I I agree strongly quite slightly neither slightly quite strongly Learning to operate a CAS is easy for me. disagree I I I I I I I agree strongly quite slightly neither slightly quite strongly 167 U-31 Using a CAS enhances my effectiveness on the job. disagree I L I strongly quite slightly neither slightly quite 8troflgly U-32 Using a CAS fits into my work style. disagree I I I I I I I agree strongly quite slightly neither slightly quite strongly U-33 I would have difficulty explaining why a CAS may or may not be beneficial. disagree I X I I I agree strongly quite slightly neither slightly quite strongly U-34 I was permitted to use a CAS on a trail basis long enough to see what it could do. dsagree _______________ strongly quite slightly neither slightly U-35 Using a CAS gives me greater control over my work. I I I I I I I I I quite strongly neither slightly quite I I strongly quite slightly neither slightly quite 3trongly U-37 Having a CAS is a status symbol in my organization. disagie strongly quite slightly neither slightly quite strongly U-38 It is easy for me to observe others using CAS in my firm. disagree I I I agree disagree strongly quite slightly U-36 Using a CAS increases my productivity. disagrcej I strongly >c agree agree agree agree agree -I I j strongly quite slightly neither slightly quite strongly 168 [NALLY, IN THIS SECTION WE WOULD LIKE TO ASK YOU A FEW QUESONSABOUT YOUR USE OF THE CAS. B-i Overall, my using a CAS in my job is (place an X on all four scales): good I I I I I I ‘‘ extremely quite slightly neither slightly quite extremely harmful f I I I I I X beneficial extremely quite slightly neither slightly quite extremely wiscj I I I I I Ifoolish extremely quite slightly neither slightly quite extremely negative I I I I I potive extremely quite slightly neither slightly quite extremely B-2 Assuming that any decision to use the CAS is totally up to you, how would you rate yourpotential use of the CAS in the next six months? likely f ‘ I I I I I I I unlikely extremely quite slightly neither slightly quite extremely 1 improbable probable extremely quite slightly neither slightly quite extremely B-3 Approximately when (month and year) did you first start using a CAS beyond any trial of it you may have carried out? 10 MOWTH YEAR 169 B-4 Overall, how many hours per week do you use a CAS? ‘0 HOURS B-5 How regularly do you now use a CAS? (Place an ‘X’ under the appropriate column): Less than About 1-3 About About More than once per times per once per 2-4 times once per once per Not at all month month week per week day day )< B-6 For each computer function listed below, please indicate whether you use it on a mainframe/mini computer, on a microcomputer, on both, or on neither. (Place an ‘X’ under the appropriate column): Mainframe? Mini Micro Both Neither Accounting Software Graphics Generation Information Retrieval Report Generation Spreadsheet Statistical Analysis >( Textlword Processing Other (please specify 170 B-7 On average, how frequently do you currently use the following functions (place an ‘X’ under the appropriate column): Less About than 1-3 About 2-4 About More once times once times once than Not at per per per per per once all month month week week day per day Accounting Software Graphics Generation Information Retrieval Report Generation Spreadsheet Statistical Analysis >c Text/word Processing Other (please specify) B-8 For each of the following questions, place an ‘X’ under the appropriate column for each applicable function: a. On average how many hours per week do you spend using the CAS on the following functions? b. Please indicate approximately how long (in months) you have been regularly using any of the following functions. Other Accounting Graphics Jnformation Report Statistical Textlword (please Software Generation Retrieval Generation Spreadsheet Analysis Processing specify) HOURS I I I MONTHS I 8- 8— 8— 171 B-9 Overall, how do you expect your usage of the CF\S will change in the p six months? (Place an ‘X’ in the appropriate column): increasedi I I I I I Id sigmfi- some- marginally same marginally some- signifi candy what what cantly B-lO Overall, how has your usage of CAS changed in the last six months? (Place an ‘X’ in the appropriate column): increasedf I I I I fdecreased signifi- some- marginally same marginally some- signifi candy what what candy B-li I have been using a CAS for (place an ‘X’ under the appropriate column): Less than About 1-3 About About More than once per times per once per 2-4 times once per once per Not at all month month week per week day day . >( 172 B-12 When I started using my CAS, I received continuing support (training or help) for my CAS from the following sources (place an ‘X’ under the appropriate column for each source): About 1- More Less than 3 Times About 2- About than Once Not at Once per per Once per 4 Times Once per per Day AU Month Month Week per Week Day Other personnel from my company Personal friend (non employee) Public accounting firm Non-Accountant computer consultant Other (please specify) B-13 I currently receive continuing support (training or help) for my CAS from the following sources (place an ‘X’ under the appropriate column for each source): About 1- More Less than 3 Times About 2- About than Once Not at Once per per Once per 4 Times Once per per Day All Month Month Week per Week Day Other personnel from my company Personal friend (non employee) Public accounting firm Non-Accountant computer consultant Other >{ (please specify) 173 B-14 The last 10 times I received continuing support from a source external to my firm using my CAS (Place an ‘X’ under the appropriate column for each applicable source, up to a maximum of 10 times in total. Total may be less than 10.): ‘‘1h123 41516 7.819110 Personal friend (non employee) Public Accounting finn )( Non-Accountant computer consultant Other (please specify) B-15 Currently, if I need help with my CAS, I know I can get support from the following sources (place an ‘X’ under the appropriate column for each source): none constant 1 2 3 some 5 6 7 Other personnel from my company Personal friend (non-employee) Public accounting finn Non-Accountant computer consultant Other (please specify) 174 B-16 I plan on getting my future CAS help from the following sources (place an ‘X’ under the appropriate column for each applicable source): none constant 1 2 3 some 5 6 7 Other personnel from my company Personal friend (non-employee) Public accounting firm Non-Accountant computer consultant Other (please specify) B17 I am satisfied with the current level of continuing support for my CAS that I receive from the following sources (place an ‘X’ under the appropriate column for each applicable source): satisfied unsatisfied extremely quite slightly neither j slightly quite extremely Other personiel from my company Personal friend (non- employee) Public accounting firm Non-Accountant computer consultant None Other (please specify) 175 B-18 How effective do you feel the following were in helping you to get started in your use of a CAS? (Place an ‘X’ under the appropriate column for each applicable source): effective ineffective extremely quite slightly neither slightly quite extremely Other personnel from my company Personal fricnd (non employee) —_______ Public accounting fri-rn Non-Accountant computer consultant Other (please specify) B-19 How effective do you feel the following have been in helping you in your current use of a CAS? (Place an ‘X’ under the appropriate column for each applicable source): effective ineffective extremely quite slightly neither slightly quitc extremely Other personnel from my company Personal fiicnd (non- employee) Public accounting finn X Non-Accountant computer consultant Other (please specify) THANK YOU FOR YOUR PERSEVERANCE AND COOPERATION SO FAR. NOW, PLEASE SKIP TO THE FINAL SECTION, SECTION D. SECTION D 176 In this last section, we would like to ask you some questions about yourself. Remember, all answers aie confidential, and no respondent can be identified, so please give as candid a response as possible. FIRST, WE WOULD LIKE YOU TO ONCE AGAIN INDICATE AGREEMENT OR DISAGREEMENT WITH A NUMBER OF STATEMENTS; THIS TIME ABOUT YOURSELF. PLEASE PLACE AN ‘X’ IN THE APPROPRIATE SPACE. 1-1 I am generally cautious about accepting new ideas. A disagree agree strongly quite slightly neither slightly quite strongly 1-2 I rarely trust new ideas until I can see whether the vast majority of people around me accept them. I I I I strongly quite slightly neither slightly quite strongly 1-3 1 am aware that I am usually one of the last people in my group to accept something new. disagree I I I I I agree strongly quite slightly neither slightly quite strongly 1-4 I am reluctant about adopting new ways of doing things until I see them working for people around me. disagrees I I xl I I I I strongly quite slightly neither slightly quite strongly 1-5 I find it stimulating to be original in my thinking and behaviour. disagree I I I I I Jagree strongly quite slightly neither slightly quite strongly 1-6 I tend to feel that the old way of living and doing things is the best way. disagree I I I I I I agree strongly quite slightly neither slightly quite strongly 177 1-7 I am challenged by ambiguities and unsolved problems. disagree ________ ________ ________ ________ _____ strongly quite sightly neither slightly 1-8 I must see other people using new innovations before I will disagreef I I I strongly quite slightly 1-9 I am challenged by unanswered questions. disagree I I I strongly quite slightly 1-10 I often find myself sceptical of new ideas. disagreci I I I strongly quite slightly neither NEXT, WOULD YOU PLEASE INDICATE HOW LIKELY OR UNLIKELY EACH OF THE FOLLOWING STATEMENTS ARE BY ONCE AGAIN PLACING AN ‘X’ IN THE APPROPRIATE SPACE. S-i Most people who are important to me think I should use the CAS in my job. likely I I I I I I I unlikely extremely quite slightly neither slightly quite extremely S-2 My close friends think that I should use the CAS in my job. likely I I ‘‘ I I I unlikely extremely quite slightly neither slightly quite extremely S-3 My co-workers (peers) think that I should use the CAS in my job. likely I I I I I I unlikely I I I agree quite strongly consider them. agree neither slightly quite strongly neither I I_i I slightly quite strongly agree agreeI I I I slightly quite strongly extremely quite slightly neither slightly quite extremely 178 S-4 My immediate superiors think that I should use the CAS in my job. likely unlikely extremely quite slightly neither slightly quite extremely S-S Senior management thinks that I should use the CAS in my job. likely unlikely extremely quite slightly neither slightly quite extremely S-ó My subordinates think I should use the CAS in my job. likely unlikely extremely quite slightly neither slightly quite extremely S-7 Generally speaking, I want to do what most people who are important to me think I should do. likely I I I < I I I J unlikely extremely quite slightly neither slightly quite extremely S-8 Generally speaking, I want to do what my close friends think I should do. ><likely unlikely extremely quite slightly neither slightly quite extremely S-9 Generally speaking, I want to do what my co-workers think I should do. likely I I I I I I unlikely extremely quite slightly neither slightly quits extremely S-1O Generally speaking, I want to do what my immediate supervisors think I should do. likely unlikely extremely quite slightly neither slightly quite extremely S-i 1 Generally speaking, I want to do what senior management thinks I should do. likely unlikely extremely quite slightly neither slightly quite extremely S-12 Generally speaking, I want to do what my subordinates think I should do. likely I I I ‘> I I I I unlikely extremely quite slightly neither slightly quite extremely 179 FINALLY, WE WOULD LIKE TO ASK A FEW QUESTIONS ABOUT YOURSELF FOR STATISTICAL PURPOSES. COULD YOU PLEASE INDICATE: —— P-i Your sex (place an ‘X’ beside the appropriate coLn) KALE ___ ___ ___ ___ ___ FEMALE P-2 Your present age: —________ years P-3 Your department: i -r -i P-4 Your job title: CL-Dc)r-- P-5 Years you have worked in your current department.1years P-6 Years you have worked in this coeçany. I “years P-7 What is the hi9hest LeveL of education that you coapLeted? (pLace an ‘X’ beside the appropriate colwn) GRADE SCHOOL SOME HIGH SCHOOl. HIGH SCHOOL GRADUATE SOME TECHNICAL COLLEGE TECHNICAL COLLEGE GRADUATE SOME COMMUNITY COLLEGE COMMUNITY COLLEGE GRADUATE SOME UNIVERSITY UNIVERSITY GRADUATE POSTGRADUATE P-8 The job that best describes my organizationaL Level is (place an ‘X’ beside the appropriate coluT) ___ __ EXECUTIVE/TOP MANAGEMENT _ MIDDLE MANAGEMENT SUPERVISORY PROFESSIONAL/EXEMPT Ic TECHNICAL/NON-PROFIT CLERICAL OTHER (pLease specify) GENERAL BUSINESS INFORMATION 180 F-i Nuiter of: EirçLoyees Accounting staff Full time ______ -3 Part time F-2 Annual Sales last year (in thousands of dollars. k=1,000). < $250k I $250k-$SOOk I $500k-$1,000k F-3 Type of organization (eg. profit, non-profit, CO-OP, etc...) \Jj t F-4 Does your firm plan to iirçilement or expand a CAS in the next two years? (Yes or No) Yes No Don’t Know ‘7 If Yes, approximately how nich do you expect to spend on the CAS in this time? $ F-5 THANK YOU VERY MUCH FOR YOUR PARTICIPATION! If you wish to add any coments or further observations, please use the space below or simpLy attach them tthis page. r $1,000k-$10,000k > $10,000kI I I APPENDIX 11-B 181 182 Appendix II- Bi Client Letter 183 YOUR FIRM’S LEtTERHEAD Dear Client: The University of British Columbia has contacted our firm about participating in a study on Information Technology (iT). They have also requested permission to contact our clients in order to ask you to participate in the study. Our firm has met with the researchers from UBC to find out more about the nature of the study. We believe that the results from this study would be important to both our finn and to our clients’ helping us to manage the new forms of IT that will be introduced into firms like yours over the next few years (and beyond). We would like to encourage you to participate in this study and fill out the enclosed questionnaire(s). You may find more than one questionnaire with the enclosed material. Please distribute a questionnaire to the owner/manager, the chief accountant, and to any other accounting staff members interested in participating. Also, please use the enclosed return envelope to mail the completed questionnaires. Confidentiality is assured and will be maintained in two ways: 1. Your responses cannot be traced back to your firm as the UBC researchers do not and will not have access to your name or addrçss (unless you specifically include this information on the questionnaire). All mailings are handled by our firm. 2. Since you will be mailing the completed questionnaire back to the UBC researchers, no personnel from our accounting firm will have access to your responses. If you have not been provided with enough questionnaires, please call our office or photocopy sufficient additional questionnaires. If you have any questions about this study please contact the UBC researchers at the phone number on the attached letter. Our firm is not sponsoring or otherwise associated with either the research study or the UBC researchers. 184 Appendix II- B2 Partner Letter Faculty of Commerce Albert S. Dexter 185 and Business Administration Associate Professor 2053 Main Mall Management Information Systems ___ ___ _ Vancouver, B.C. Canada V6T 1Z2 Telephone: (604) 822-8380 Fax: (604) 822-8489 September 13, 1991 Dear Sir/Madame: We are conducting a study at the University of British Columbia on Information Technology (iT). We would like to determine how IT is affecting Small Business firms. Many firms have installed computer systems, which are a type of IT. Some of these systems have been installed successfully while others have not been very successful. The purpose of our research is to determine what the difference is between firms that have successfully installed computer systems and those that were not so successfully installed. We will obtain this information from a questionnaire that asks respondents their opinions about using computers. We hope to use the results from this study to help owners and managers make sound business decisions about acquiring other iT in the future. It is undeniable that firms will be purchasing other IT in the future. Technology such as Teleconferencing, Networking, Image Processing, Desktop Publishing, Multimedia, etc., are currently becoming established as the newest forms of IT that many businesses are looking at to improve their competitive position. Over the next five to ten years there will be other if’s that we can scarcely conceive as yet (could you have imagined our current if ten years ago?). We would like to get your opinions about using computers by filling out a questionnaire. This will take approximately 20 to 25 minutes. Your opinion is important, whether or not you currently use a computer, and we would like to hear from you. Please note that your answers will be completely confidential, and that anonymity is assured. Once again, the results of this research should help us to better understand what people think about personally using computers. Other studies have shown that there is a link between what employees think and how an organization performs. Thus our results should enable organizations to better manage the spread of computers and other IT. As a token of our appreciation, once the stud is completed, we would be pleased to send you a copy of our findings, conclusions and recommendations if you send us a card indicating your name and address. We hope to receive your completed questionnaire by the end of the week. Please mail it in the envelope provided. If you have any questions about this questionnaire, please call Rick Laktin at (604)-270-8953. Thank you for your assistance. Sincerely, Rick Laktin Albert S. Dexter

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 4 2
Japan 2 0
China 2 2
Indonesia 1 0
City Views Downloads
Phoenix 2 0
Tokyo 2 0
Unknown 1 1
Guangzhou 1 0
Beijing 1 0
Sunnyvale 1 1
Ashburn 1 0

{[{ mDataHeader[type] }]} {[{ month[type] }]} {[{ tData[type] }]}

Share

Share to:

Comment

Related Items