Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Adoption of information technology in a small business setting Laktin, Richard S. 1992

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
831-ubc_1992_spring_laktin_richard_s.pdf [ 3.41MB ]
Metadata
JSON: 831-1.0086714.json
JSON-LD: 831-1.0086714-ld.json
RDF/XML (Pretty): 831-1.0086714-rdf.xml
RDF/JSON: 831-1.0086714-rdf.json
Turtle: 831-1.0086714-turtle.txt
N-Triples: 831-1.0086714-rdf-ntriples.txt
Original Record: 831-1.0086714-source.json
Full Text
831-1.0086714-fulltext.txt
Citation
831-1.0086714.ris

Full Text

ADOPTION OF iNFORMATION TECHNOLOGY IN A SMALL BUSINESS SETTINGbyRICHARD S. LAKTTNB. Comm., University of British Columbia, 1982CA, Canadian Institute of Chartered Accountants, 1991A THESIS SUBMITTED IN PARTIAL FULFILMENT OFTHE REQUIREMENTS FOR THE DEGREE OFMASTER OF SCIENCEinTHE FACULTY OF GRADUATE STUDIESDepartment of Commerce and Business AdministrationWe accept this thesis as conformingto the required standardTHE UNIVERSITY OF BRITISH COLUMBIAApril 1992© Richard S. Laktin, 1992In presenting this thesis in partial fulfilment of the requirements for an advanced degree at theUniversity of British Columbia, I agree that the Library shall make it freely available forreference and study. I further agree that permission for extensive copying of this thesis forscholarly purposes may be granted by the head of my department or by his or herrepresentatives. It is understood that copying or publication of this thesis for financial gainshall not be allowed without my written permission.Department of Commerce and Business AdministrationThe University of British Columbia1956 Main MallVancouver, British ColumbiaCanadaV6T 1Y3Date: 30 April, 199211AB STRACTMany small businesses are turning to Information Technology as a means ofcompetitive advantage and survival in today’s tougher business climate. The PublicAccounting profession portrays itself in the role of Information Consultant to small businesswhen it comes to information technology. The role that Public Accountants play in theinformation technology adoption process is poorly understood. The purpose of this researchwas to examine more closely the role that information consultants play in the adoptionprocess, with particular emphasis on the public accountant.The Dffusion of Information Technology model (Moore, 1989) was used as thetheoretical foundation for this study. The Diffusion of Information Technology model is llgrounded 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 theDffiision 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 ofpublic accounting firms. Profiles of information technology users and non-users weregenerated from questionnaire data. These profiles were subject to regression analysis andstructural equation modelling using PLS (Partial Least Squares). The analysis provided someanswers to the role accountants play in the information technology adoption process as well assupporting the Diffusion of Information Technology model in a small business domain.111TABLE OF CONTENTSABSTRACT iiTABLE OF CONTENTS iiiLIST OF TABLES ivLIST OFFIGURES vACKNOWLEDGEMENTS viiiCHAPTER 1: INTRODUCTION AND OVERVIEW OF RESEARCH 11.1 RESEARCH STUDY RATIONALE 11.2 RESEARCH DIRECTION 31.3 THE COMPUTERIZED ACCOUNTING SYSTEM 41.4 TOWARDS A SMALL BUSINESS ORIENTATION 51.5 THE ROLE OF INFORMATION TECHNOLOGY IN ORGANIZATIONS 7CHAPTER 2: LITERATURE REVIEW 10CHAPTER 3: ADOPTION OF INFORMATION TECHNOLOGY 133.1 DIFFUSION OF INNOVATIONS 143.2 THE THEORY OF REASONED ACTION 153.3 DIFFUSION OF INFORMATION TECHNOLOGY 153.4 SIGNIFICANCE OF THE DIFFUSION OF INFORMATION TECHNOLOGYMODEL 15CHAPTER 4: TECHNOLOGY TRANSFER- RESEARCH QUESTIONS 174.1 RESEARCH HYPOTHESES 18CHAPTER 5: INSTRUMENT DEVELOPMENT 22SECTION A- INTRODUCTION 225.1 GENERAL 225.1.1 RELIABILITY 225.1.2 VALIDITY 235.1.3 QUESTIONNAIRE SELECTION 24SECTION B: QUESTIONNAIRE DESIGN - PILOT STUDY 245.2 CLIENT/DIFFUSION OF INFORMATION TECHNOLOGY QUESTIONNAIRE 245.2.1 PERCEIVED CHARACTERISTICS OF INNOVATIONS 245.2.2 SYSTEM USAGE 265.2.3 CLIENT COMPUTERIZED ACCOUNTING SYSTEM SUPPORT 30SECTION C: FINAL SURVEYS - SCALE RELIABILITIES 305.4 GENERAL 305.5RESULTS 31SECTION D: QUESTIONNAIRE DESIGN 315.6 GENERAL 315.7 FORMAT 325.7.1 PAMPHLET 325.7.2 QUESTION LAYOUT 335.7.3 COVERING LETTER 33CHAPTER 6: DATA COLLECTION AND ANALYSIS 34SECTION A: DATA COLLECTION AND CONDITIONING 346.1 INTRODUCTION 346.2 SURVEY SAMPLE 346.2.1 TARGET POPULATION SELECTION 346.2.2 PROBLEMS ENCOUNTERED 366.2.3 RESPONSE RATES 376.3 CLIENT FIRM’S SURVEY 386.3.1 RESULT DEMONSTRABILITY 396.4 CONDITIONING THE DATA 406.4.1 GENERAL 406.4.2 ACCURACY OF INPUT DATA 406.4.3 MISSING DATA 40iv6.4.4 OUTLIERS AN]) SKEWNESS .416.4.5 NON-LINEARITY AND HOMOSCEDASTICITY 41SECTION B: DESCRIPTIVE STATISTICS 426.5 GENERAL 426.6 DEMOGRAPHICS 436.7 ATTITUDE TOWARDS INNOVATING 446.8 PERCEIVED CHARACTERISTICS OF INNOVATING 456.9 SUBJECTIVE NORMS 466.10 INNOVATIVENESS MEASURES 466.11 COMPUTERIZED ACCOUNTING SYSTEM SUPPORT 49SECTION C: REGRESSION ANALYSIS 496.12 GENERAL 496.13 THE EFFECT OF PERCEIVED CHARACTERISTICS OF INNOVATIVNESS ANDVOLUNTARINESS ON ATTITUDE 506.14 THE EFFECT OF ATTITUDE. SUBJECTIVE NORM. PERCEIVEDCHARACTERISTICS OF INNOVATIVENESS. VOLUNTARINESS AN])SUPPORT ON INNOVATIVENESS 536.14.1 GENERAL 536.14.2 ATTITUDE. SUBJECTIVE NORM AND VOLUNTARINESS ONINNOVATIVENESS 536.14.3 PERCEIVED CHARACTERISTICS OF INNOVATING. SUBJECTIVE NORM,AND VOLUNTARINESS ON INNOVATIVENESS 556.14.4 OTHER REGRESSIONS 57SECTION D: PATH MODELING 596.15 CHOICE OF PATH MODEL COMPUTER IMPLEMENTATION- LISRELvsPLS 606.15.1 DESIGN OF PLS PATH MODEL 616.15.2 ANALYSIS OF SAMPLE SIZE REQUIREMENTS 626.15.3 GOODNESS OF FIT DETERMINATION 646.15.4 ASSESSMENT OF HYPOTHESES TESTING 666.16 SUMMARY OF RESULTS: PATH ANALYSIS 69SECTION F: SUMMARY OF DATA ANALYSIS 706.17 GENERAL 706.18 SUMMARY OF DESCRIPTIVE STATISTICS 706.19 SUMMARY OF HYPOTHESES TESTING 70CHAPTER 7: CONTRIBUTIONS. IMPLICATIONS AND LIMITATIONS 737.1 INTRODUCTION 737.2 SUMMARY OF THE RESEARCH PROCESS 737.3 THE RESEARCH QUESTIONS ANSWERED 747.3.1 QUESTION TWO 747.3.2 QUESTION ONE 757.4 CONTRIBUTIONS 767.5 LIMITATIONS OF THE STUDY 777.6 CONCLUSION 78TABLES 80FIGURES 100BIBLIOGRAPHY 108APPENDICES 117VLIST OF TABLESTABLE 1 - RELIABILITY COEFFICIENT: PILOT TEST (SPSS) 81TABLE 2 - RELIABILITY COEFFICIENT: ACTUAL STUDY (SPSS) 81TABLE 3 - RELIABILITY COEFFICIENT: ACTUAL STUDY (SPSS) USERS VS NON-USERS 82TABLE 4- RELIABILITY COEFFICIENT: MOORE 82TABLE 5(a) - DEMOGRAPHIC BACKGROU1D OF SURVEY RESPONDENTS 83TABLE 5(b) - DEMOGRAPHIC BACKGROU1D OF SURVEY RESPONDENTS 84TABLE 6(a)- SURVEY VARIABLES- DESCRIPTIVE STATISTICS 85TABLE 6(b) - SURVEY VARIABLES- DESCRIPTIVE STATISTICS 86TABLE 7(a) - USERS VERSUS NON-USERS 87TABLE 7(b) - USERS VERSUS NON-USERS 88TABLE 8- REGRESSION RESULTS 89TABLE 9- REGRESSION RESULTS 90TABLE 10(a) - REGRESSION RESULTS 91TABLE 10(b) - REGRESSION RESULTS 92TABLE 11(a) - REGRESSION RESULTS 93TABLE 11(b)- REGRESSION RESULTS 94TABLE 12(a) - REGRESSION RESULTS 95TABLE 12(b) - REGRESSION RESULTS 96TABLE 13(a)-SUIVIMARY RESULTS OF HYPOTHESES TESTING 97TABLE 13(b) - SUMMARY RESULTS OF HYPOTHESES TESTING 98TABLE 14-GENERAL PLS STATISTICS FOR TESTED MODELS 99viLIST OF FIGURESFIGURE 1- DIFFUSION OF INFORMATION TECHNOLOGY MODEL 101FIGURE 2- DIFFUSION OF INNOVATIONS MODEL 102FIGURE 3- INNOVATION DECISION MODEL 103FIGURE 4- STAGES OF TNE INNOVATION DECISION PROCESS MODEL 104FIGURE 5- NON-CAS USERS: RESULT DEMONSTRABILITY 105FIGURE 6- DIFFUSION OF INFORMATION TECHNOLOGY MODEL: PLS LOADINGS ON 106ORIGINAL MODELFIGURE 7- DIFFUSION OF INFORMATION TECHNOLOGY MODEL: PLS LOADINGS ON 107EXTENDED MODELviiACKNOWLEDGEMENT SThe completion of this dissertation required the help, support and frequentencouragement by several individuals, to whom I owe a great deal of gratitude. Specialrecognition is directed at my thesis supervisor, Professor Al Dexter, who was always there todirect, encourage, and refocus my efforts at producing this paper. I would like to also thankthe other members of my committee, Chino Rao and Gary Moore, for their helpful criticismand comments. I would also like to thank my fellow graduate students at UBC and the facultymembers 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 Calgaryfor his help with the PLS analysis. Without his timely intervention, a major part of thestatistical 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 thispaper. 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 verificationand reviewing this paper for grammatical errors.Thank you Lindsay, for your encouragement and insightful suggestions on the finalversions of this paper.Finally, I would like to thank my parents, Cyril and Doris, for their financial andmaterial support provided at various times throughout the time spent completing the graduateprogram. Without their help this thesis would not have been completed.1CHAPTER 1: JNTRODUCTION AND OVERVIEW OF RESEARCH1.1 RESEARCH STuDY RATIONALEMost small firms have limited access to information sources on informationtechnology (IT). As a result, information is often sought from an external informationconsultant (Goodson, 1990). The role of the external information consultant as an informationsource to small firms is an important research area. In the role of information consultantsprofessional accountants have been involved with many computer systems that have beenconsidered successful by their users, and several that have been considered failures. To theaccountant as well as the small business client they serve, the success or failure of theintroduction of an information technology may seem to be not only a product of planning buta product of fortune as well. To an accountant, working in a profession that sells informationand methods of generating information as products, unsuccessful implementation ofcomputerized accounting systems is to be avoided. Maintaining good client relations is thebottom line to professional accounting organizations and failures (perceived or otherwise) areunacceptable, as small businesses cannot afford the emotional and monetary costs of anunsuccessfully implemented computerized accounting system. Research that can illuminate theinteraction between small firms and their public accountant may provide the accountingprofession with an understanding of how to better deliver the current information technologyservices it already provides to small business clients.Equally as important to small businesses are suggestions for coping with informationtechnology and finding ways to increase productivity given the scarcity of trained and skilledspecialists. There is a growing belief that information technology will be the most importanttechnology to change business and society in the 1990’s as Canada moves from an economythat is resource based to one that is service based (Gunning, 1992). Small businesses may endup in the unenviable position of relying on information technology much more than theycurrently are, and unable to find ready assistance (in the form of skilled labour) to implementand manage the information technology they require.2For the public accountant, this research should help reinforce the need to beadequately trained in areas that will be called upon increasingly more often by current andfuture clients, such as information technology. Public accountants are finding themselvesmore 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 isgetting 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 informationtechnology to become independent of further extensive professional aid in using theinformation 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 ininstalling computerized accounting services to ensure that the system will meet theaccountant’s requirements as well as the manager’s. This expectation arises from the publicperception of the accountant’s expertise with information technology. Public accountants nowfind that some 95% of their audit clients have information technology installed (Walker,1991). Often, however, accountants are not familiar enough with automated systems andtreat them as automated manual systems (which they are not), resulting in potential disserviceto the client (Overbey et al, 1987). To avoid the public and private humiliation that adverseheadlines tend to bring, as well as the subsequent lawsuits and loss of business, research isrequired that will aid the professional accountant in helping his client successfully adopt anynew information technology.Despite the good reputation of information consultants, failures still occur. Practicaladvice based on solid research, designed to minimize the risk of failure, would be verywelcome. 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.3with the introduction of the information technology need to be defined in an attempt to avoidany 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 publicaccounting profession (Batch et al, 1989) as well as in other professions (Newman, 1990). Ifthese information technology specialists are resistant to learning and adopting newerinformation technology, it should be no wonder that the information technology specialistsexperience user resistance to the introduction of even basic information technology. Thisresearch should provide motivation for the information consultant to continue on the arduoustask of bringing his clients into the 1990’s by introducing a theory backed approach on how tosuccessfully introduce new information technology into an organization.1.2 RESEARCH DIRECTIONA review of any major MIS publication will show that the majority of research in MISis carried out on large organizations (Attewell, 1989). The result is similar for studies on howinformation technology affects organizations as well. It can be easy to fall into the trap ofthinking that results from these studies apply equally well to small organizations. However, ithas been shown that small firms differ from large firms in many areas, including job creationand 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. Therehas been little empirical research that has looked at the role of information consultants in theadoption of information technology in a small business setting.The role of the information consultant in the diffusion of innovations process will beexamined. For small business managers this is an important issue as small firms usually lackthe resources to develop necessary expertise in-house. These businesses often look to theirprofessional accountant for advice on their information requirements. For professionalaccountants this is also an important issue as their associations are attempting to transform4their members into information specialists to meet the needs of their clients. For example, theCanadian Institute of Chartered Accountants (CICA) is currently considering recognition ofareas of specialization (if not accreditation) amongst CA’s, one such area being informationtechnology (Brown, 1992, Goodson, 1990; Luscombe, 1990).1.3 THE COMPUTERIZED ACCOUNHNG SYSTEMThe Computerized Accounting System is the specific information technology ofinterest to the accounting profession and small business in general. The ComputerizedAccounting 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 computerizedtools designed for an individual; is used on a microcomputer or terminal connected to aminicomputer or mainframe; is accompanied by appropriate software; and is used directly(hands on) (Moore & Benbasat, 1991). The Personal Work Station is general and not functiondependent. A Personal Work Station can be used in marketing, finance, production or anyother area of an organization. The choice of tools (hardware/software) comprising thePersonal Work Station is usually up to the individual.A Computerized Accounting System for the purposes of this research is defined as aset of computerized tools for an individual, and usually consists of a personal ormicrocomputer with one or more software packages, including an accounting program and/orother software such as a spreadsheet, database, word-processing, etc. in support of theaccounting function. A Computerized Accounting System is similar to the Personal WorkStation defined by Moore. The major differences between a Computerized Accounting Systemand Personal Work Station are that the use of a Computerized Accounting System(hardware/software) is usually an organizational decision and a Computerized AccountingSystem supports the accounting function primarily.1.4 TOWARDS A SMALL BUSiNESS ORIENTATIONResearch into information technology, now entering its third decade, has primarilyfocused on large organizations. Although there are several issues regarding whether or not itis necessary to study small businesses separately from other businesses, the main issue iswhether the organizational factors found in small firms are sufficiently similar to those oflarger firms. If the main factors of interest are common across firms then it is appropriate andeconomically prudent to limit research studies to large firms and extrapolate the results to allother firms, given the difficulty in obtaining results from small firms. If these factors aredissimilar, then we as researchers have been omitting a significant group of organizations fromour studies and we cannot claim with confidence that our results are generalizable across allfirms.This orientation towards big business is natural, as larger firms tend to operate incomplex conditions. Understanding the environmental and internal factors that influence howa firm will behave is important to the enterprise and to society. This understanding isnecessary because large firms have high public profiles, are large employers, and make largecontributions to local economies, research institutes, and governments in the form of taxes ordonations. 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 inCanada, commissioned by the Federal Business Development Bank (FBDB) in 1986 andreleased in 1987, found some unexpected results. Small businesses (defined as firms withsales under $2 million and typically with less than 20 employees) accounted for 25% of ourGNP, 96% of all business organizations (over 700,000), created the greatest employmentopportunities for women and young people (under 25 years old), had less of a wage gapbetween 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 allnew 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 smallbusinesses created over 98% of the new jobs in the period 1984-1987 (Small Business6Magazine, October 1989). The increasing importance of small businesses can be shown inB.C., where small businesses employed almost 60% of B.C. workers by the end of 1988compared to under 45% in 1986 (Smith, 1989), represented 92% of all businesses (RichmondBusiness, 1990) and created 96% of net new jobs (Richmond Business, 1990). Similargrowth has occurred all across Canada during this time. In the USA, small businesses in thelate 1970’s and early 1980’s accounted for 98% of all non-farm business organizations; 39% ofthe 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 period1970-1979 (Pavitt et al, 1989) and the portion of innovating small firms (under 200employees) 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. Evenin boom times many small firms experience a rocky road. The Canadian experience in theperiod 1978-82, for firms employing 5 or less full time employees, indicated that for every100 net new jobs created: 52 were in currently existing firms; 106 were for newly createdfirms which survived; and 58 were lost for new firms that didn’t survive (FBDB, 1987). Dueto this sensitivity to the economic environment, smaller firms are often perceived to be morerisky, subject to higher failure rates, have more problems collecting receivables, have moredifficulty keeping adequate records (DeLone, 1988).’It is also evident that small firms are very important to public accountants, and viceversa. 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 appearsthat many small businesses in Canada rely on their public accountants for more than theiraccounting and tax knowledge (Goodson, 1990; Delente et al, 1990; Hamilton, 1989), whilemost small firms in Australia still seek mainly year end accounting and tax services from theiraccountants (Holmes & Nicholls, 1989). A recent Canadian study on small firm’s relationshipwith their accountants found that one of the reasons small firms initially engaged their1The researcher has encountered several small firms that have experienced most, if not all, of the aboveproblems through his own involvement in accounting public practice.7accountant was to install a computer system (ranked 8th on the top 10 list), a responseprovided by 21% of the survey firms. However, when asked about ongoing work performedby their accountants, “advice on computers” did not make the top 10 list. A significantportion of firms requested that more services, including computer systems advice, be providedby their accountant (Hamilton, 1989).1.5 THE ROLE OF INFORMATION TECHNOLOGY 1N ORGANIZATIONSUnplanned and uncontrolled adoption of information technology are major problemsfor 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 theseproblems 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 haveinternal resources to help overcome these problems (in-house expertise, financial resources toacquire adequate information technology) most small firms remain at risk due to their lack ofresources.Factors contributing to the problem of unmanaged information technology includeignorance 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 moreconcerned about the information technology’s impact on himself (Baronas & Louis, 1988);management ignorance of the skills the organization has available for using the informationtechnology (Benson, 1983); and management reluctance or inability to provide adequate usertraining (Buckler, 1990 and others). For large and small firms the information technology useris often unsophisticated because the technology is new to the firm and personnel familiar withit would be relatively few (Lees & Lees, 1987). To learn to use the information technology theuser 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 sourcesdepends on the individuals own skills and the organizations resources.The ability of large firms to cope with the above problems of information technologyare generally better than for small firms. A problem faced by many small business managers isthat they attempt to manage information technology based on practices that they are familiarwith, strategies aimed at obtaining or maintaining stability. Such practices are not conduciveto 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. Mostlarge firms have experienced these major changes several years (or decades) ago and will bemore familiar in dealing with change than their smaller counterparts. In large firms users oftenhave skilled resources to fall back on such as an EDP department or personnel who hadrecently come from a firm with the information technology. With the increasing complexity ofcomputer technology even these traditional sources are finding it increasingly difficult to keepup (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 businesseson 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, orthemselves (Lefebvre & Lefebvre, 1990; Gable, 1989; Delone, 1988; Lees, 1987). In manycases 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 smallfirms (Bracker & Pearson, 1985). For the small business the specialist is often theirprofessional advisor - their public accountant (Delente et. al, 1990; Peat et al, 1984). Recentstudies show that in Canada there is a growing shortage of skilled information technologyspecialists (Buechert, 1992). While this shortage poses problems from businesses in general, itprovides 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 ininformation technology on a large scale (Brown, 1992).9It has been suggested that the reasons a small firm seeks outside help for managinginformation 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 thebusiness manager will turn as the accountant often has provided the business planning adviceinitially. However, success in providing a business plan doesnt ensure success regarding theadoption of information technology. The failures of information systems installed with thehelp of information consultants have been well documented in the media. This is particularlytrue for accountants (e.g. see Babcock, 1986) and the fear of lawsuits over malpractice forproviding 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 accountingservices help small firms to be successful (Bracker & Pearson, 1985), there are also researchresults that claim using external information consultants, including accountants, provide lessthan satisfactory results for a small business (Hamilton, 1989; Baker, 1987; Lees, 1987; Lees& Lees, 1987; Bracker & Pearson, 1985). Some of these studies indicated that highersatisfaction could be achieved if the consultant provided a full range of support and services.10CHAPTER 2: LITERATURE REVIEWACCOUNTANTS:THEN“Observe that much of the difficulty in the conception of profit, taxes, costs, and so on, canbe seen to come from the professionalization of the accountants as a group. They are theones who force upon the industrial situation the concern with numbers, with exchangeablemoney, with tangibles rather than intangibles, with exactness, with predictability, withcontrol, 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 theprofessional groups. I added that the psychiatrists think of them as being the mostobsessional of any group. From what I know of them, they also attract to the schools ofaccounting those who are number bound, those who are interested in small details, thosewho 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.... Forstarters, accountants tend to be more assertive, independent-minded, unconventional,cheerful, enthusiastic, rebellious, experimenting, liberal, self-sufficient, careless of socialrules and standards, nonconforming, anxious, independent and impulsive.” [Davidson &Dalby, 1991].From an organizational perspective, there is a growing realization that information canbe 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 andcompetitive business environment and the recent technological and software trends making itfeasible and less costly to acquire an information technology, allowing firms to better manageand protect their information (Huber, 1990; McGill, 1990). The concept of information as anasset is not new to large firms or to public accountants, but to many small businesses it is anovel 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 ofits readers (including large and small firms) found that 53% of respondents had suffered lossof critical data costing an average of $14,000 (BYTE, August 1991).The wide spread diffusion of information technology has left many firms open to theissue of security. Many small firms appear to be ignorant of the necessity of information11technology security (Pendegraft et a!, 1987). For the small business it has been suggested thatsecurity is even more important than for large firms due to the high degree of reliance oninformation technology (Pendegraft et a!, 1987) and to the high degree of {unrelated] thirdparty knowledge about the use of the information technology, particularly microcomputersoftware (Bradbard et al, 1990; Overbey et al, 1987). Large firms tend to experience lesssecurity problems as they tend to use large computers and more restricted software. Priorexperience with larger computer systems also provides large firms with an advantage insafeguarding their data and micro computer systems.There are often several reasons that firms acquire information technology. Initiativeto introduce information technology is either due to a PUSH (organizational) environment ora PULL (individual) environment.A PUSH environment exists when events external to the user (or the firm) forceinformation technology on the user. Firms acquire information technology due to competitivepressures such as improved business value indicators like return on investment (ROT) figuresor 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 informationtechnology due to his own work environment. Employees may acquire information technologyfor higher job satisfaction (Kraut et al, 1989; Pentland, 1989; Millman & Hartwick, 1987).Additionally, non-IS employees may acquire information technology due to frustration withthe 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 specialiststo 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 developedhimself (Gremillion & Pyburn, 1983); but he is also responsible for the implementation (Davis,1981).12Research of successful adoption of information technology has focused on measurableattributes associated with success. Over the past decade or so, the definition of success hasevolved 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 aperspective 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 theincreasing knowledge on information technology adoption processes, currently success isviewed as a relative term (Gallupe, 1989). In other words, success is dependent on how wellthere is a match between the user’s expectation of what the information technology issupposed to accomplish, and what the information technology actually does.User attitudes have increasingly been seen as an important indicator of the success ofinformation technology adoption (Lin & Ashcrafi, 1990; Melone, 1990; Thompson, 1989;Goodhue, 1986). The research focus on user attitudes and behaviour is due to the increasedemphasis on theory based constructs such as attitudes (from the social and cognitivepsychology 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 attributeof 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 ofinformation technology, also called technology transfer (Bouldin, 1989). The study ofinformation technology using an adoption of innovations approach has been persued in theMIS field (Cooper & Zmud 1990; Alexander 1989; Moore 1989; Brancheau 1987; andothers) and in the psychology domain (Hill et al., 1987). Much of this work has drawn fromthe literature on diffusion of innovations which was pioneered by Rogers (1983), and from thework on attitudes and beliefs by Ajzen and Fishbein (1980) as well as other psychologists.13CHAPTER 3: ADOPTION OF INFORMATION TECHNOLOGYIt must be considered that there is nothing more difficult to carry out, nor more doubtful ofsuccess, nor more dangerous to handle, than to initiate a new order of things. For thereformer has enemies in all those who profit by the older order, and only lukewarmdefenders in all those who could profit by the new order. This lukewarmness arises partlyfrom fear of their adversaries, who have the laws in their favor, and partly from theincredulity of mankind, who do not truly believe in anything new until they have had anactual experience of it. (Machiavielli, Niccolo [1500’s], The Prince, Translated by LuigiRice, Rev. E. R. P. Vincent, New York: New American Library, 1952; cited in Foundationsof Business Systems (Flaaten et al, 1989, pg. 37)).A general criticism about information systems research has been the lack of anadequate theory of IS (Goodhue, 1986). There is considerable confusion on the issue of whata successful ny’brmation system is (Goodhue, 1986). The recent research on informationsystem attitudes and adoption of innovations has begun to clear up this confusion. The currentview 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 helpknowledge workers to do their jobs effectively and efficiently. Knowing these knowledgeworkers and what they are doing, as well as the information technology, would result inappropriate and successful application of the technology’ (emphasis added).The issue of knowing what the knowledge workers are doing is addressed by Moorein his study. Moore (1989) has developed a general model, the Diffusion of InformationTechnology model (see Figure 1- Diffusion of Information Technology), that explains theadoption and use of information technologies by individuals.This model was integrated from concepts contained in the Diffusion of Innovationsmodel by Rogers (1983) (see Figure 2) and the Theory of Reasoned Action by Ajzen andFishbein (1980) (see Figure 3), to explain the adoption of information technology byindividuals. In developing this model, Moore has attempted to overcome the previously notedweaknesses (ie. lack of theory, measuring information system success) in research in thisarea. The Moore model is the most comprehensive and theory backed work to date oninformation technology diffusion and adoption.143.1 DIFFUSION OF INNOVATIONSThe 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 - seeFigure 2.Based on a synthesis of the literature and research on innovations Rogers (1983) hasdetermined that there were five attributes of innovation that are all conceptually distinct fromeach other (Relative Advantage- the degree to which an innovation is perceived as beingbetter than its precursor; Compatibility- the degree to which an innovation is perceived asbeing 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 isperceived to enhance one’s image or status in one’s social system; and Visibility- the degree towhich the innovation is apparent to the sense of sight) and called the resulting seven attributesPerceived 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 ofInnovativeness variables and Voluntariness definitions.153.2 THE THEORY OF REASONED ACTIONIn 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 byhis decision or intention (which is reasoned) to perform that behaviour. The attitude towardthe specific behaviour (an individual’s personal attitude towards the behaviour) and hisSubjective 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 afunction of the individual’s belieJ. The basic premises of the Theory of Reasoned Action areillustrated in Figure 3.3.3 DIFFUSION OF iNFORMATION TECHNOLOGYThe link between the Diffusion of Innovations model and the Theory of ReasonedAction can be seen in Figure 1. The synthesized Diffusion of In!hrmation Technology modeldeveloped 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 theirperceptions of using the innovation that are key to whether an innovation diffuses.”To test the Diffusion of Information Technology model, a questionnaire wasdeveloped and administered in a cross sectional study involving individuals in six Canadianorganizations. The questionnaire results supported all eight Perceived Characteristics ofInnovation 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 TECHNOLOGYMODELThe Diffusion of Information Technology model attempts to predict, explain andinfluence individual behaviour towards the adoption of information technology. TheDiffusion 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 a16specific information technology other than the Personal Work Station, such as ComputerizedAccounting Systems. The Diffusion of Information Technology model should also applyequally well to small businesses and large businesses.These observations about the Diffusion of Information Technology model arise froman inspection of the theory on which the model is based. Because the Diffusion of InformationTechnology model is based on theoretical models, it will contain the characteristics of theunderlying models. An important characteristic of the Theory of Reasoned Action is the abilityto predict, explain and influence individual behaviour (Ajzen & Fishbein, 1980). The Theoryof Reasoned Action is generalizable and is applicable to all people. The Diffusion ofInnovations model focuses on the adoption of innovations. The Diffision of Innovationsmodel should be generalizable across all innovations, including information technology.It is important to examine whether the Diffusion of Information Technology model issufficiently 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 couldprovide a means for explaining why a particular innovation, such as installation of acomputerized accounting system, succeeds in one firm and not another. It could also be usedfor predicting if the innovation is likely to succeed, before significant time and resources arecommitted to a project, by determining if the firm has an adequate mix of similar attitudes andbeliefs as those found for the successful adopters.17CHAPTER 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 andsystems analyst by a factor of ten or more; yet these technologies are not even being used inmany DP organizations and have achieved only modest results in many others.... It took themilitary 75 years to go from the technology of muskets to the technology of rifles, so weshould not be too discouraged to learn that it takes 14-15 years ... for new softwaretechnologies to be accepted” (Yourdon, found in Bouldin, 1989, pg. xiii).As discussed in the previous section, the D/jiision of Information Technology modelhas the ability to predict and explain individual behaviour towards the adoption ofinformation technology. An underlying reason for this current study was to verif’ therobustness of Moore’s results for the Diffusion of Information Technology model. Equallyimportant 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 businessmanager or information technology specialist, it is expected that by understanding the factorsthat lead to successful adoption of information technology a systematic approach can bedeveloped to influence individual behaviour to adopt new information technology in thefuture.Besides attempting to extend the Diffusion of Information Technology model to thesmall business domain, this study attempted to obtain new knowledge regarding theappropriateness of using the theory based work of Rogers and Fishbein & Ajzen in the MISdomain. The Communication Channels section of Roger’s model, coupled with the Extentof Change Agent’s Promotion Efforts section (Figure 2) and the Connnunications Networksection of Fishbein & Ajzen’s model (Figure 3) essentially represent the same concept -information gathering/exchange (for convenience the term Communication Channels will beused 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 beunderestimated, as it has been pointed out that18“before a business unit can adopt and use a technology, members of the business unit mustbecome knowledgeable of the technology and be able to propose ideas for its use. Thisawareness results from communication behaviors ... whereby a ‘technology provider’ familiarwith the technology interacts with a potential ‘technology user’ not familiar with thetechnology” (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 externalconsultants as information sources on information technology has not been well established inthe Diffusion of Information Technology literature (Gable, 1989). Unlike most externalconsultants, accountants are often considered to be an integral part of their client’smanagement 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 HYPOTHESESThe Diffusion of Information Technology model provides a means to determine justwhat the characteristics of a successful interaction between the user and a specific informationtechnology are. Interactions between users and an information technology are registered bymeans of a questionnaire that Moore has developed and validated. Moore’s questionnaire didnot focus on the role of external information consultants, probably as a result of his focus onlarge business adoption of information technology where the necessary expertise would beavailable in-house through the Information Centre or similar department. As small firms donot have a similar body of in-house information expertise, the role of external informationconsultant becomes more important. This specific item may provide an important researcharea for small firms.As validating Moore’s results regarding the Diffusion of Information Technologymodel is one goal of this study, a summary of the Moore hypotheses (modified to reflect theComputerized Accounting System) is provided below.Hj: One attitude towards using a Computerized Accounting System will influence one’sinnovativeness with respect to Computerized Accounting System usage.19H2: Relative Advantage will have a contribution more than any other PerceivedCharacteristics of Innovation on one’s attitude towards adopting ComputerizedAccounting Systems.H3: Computer Avoidance will have a contribution less than any other PerceivedCharacteristics of Innovation on one’s attitude towards adopting ConiputerizedAccounting Systems.114: The Subjective Norm will influence one’s innovativeness with respect toComputerized Accounting System usage.H5. The Subjective Norni will influence one’s attitude toward adopting the ComputerizedAccounting System.H6: Voluntariness is negatively related to one’s innovativeness with respect toComputerized Accounting System usage.H7: Voluntariness i’ill be negatively related to one’s attitude towards usingComputerized Accounting System.An important research question for small business managers arises concerning the rolethat Support groups, especially external information consultants such as accountingprofessionals, play in the process of information technology diffusion. The researchhypotheses related to this question are developed in the following paragraphs.It has been shown, in Chapter 1, that small and medium firms rely on externalconsultants more than large firms. Because small firms have little in-house expertise ininformation technology, especially for an important information technology such as aComputerized Accounting System, the involvement of an external source of information andguidance should contribute to the success of the introduction and adoption of theComputerized Accounting System.H8. The involvement of a Support Group i’ili contribute to a successful adoption ofComputerized Accounting Systems.20As a Support Group is made up of different components, it follows that each of thesecomponents should contribute to a successful Computerized Accounting System. For thepurposes of this study, the Support Group is comprised of Friends, other Employees, externalAccountant, and external Consultant. This group generate the following hypotheses:H9. The involvement ofa Friend will contribute to a successful Computerized AccountingSystem.H10. The involvement of other Employees will contribute to a successful ComputerizedAccounting System.H1]• The involvement of an external Accountant will contribute to a successfulComputerized Accounting System.Hp. The involvement qf an external Consultant will contribute to a successfulComputerized Accounting System.An investigation of the Moore, Fishbein & Ajzen, and Rogers models indicate that thepresence of a communications channel will influence other areas of the Diffusion ofInformation Technology model as well as Innovativeness. In this study, communicationschannels is represented by the Support Group. The Fishbein & Azjen model (Figure 3) showsdirect links from Communications Network to Subjective Norm and Attitude. These linkssuggest the following two hypotheses:H13. The involvement of a Support Group will have a positive influence on SubjectiveNorm.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 PerceivedCharacteristics of Innovation variables may also be influenced by the communicationschannels. This link is suggested from a review of the adoption process indicated in Roger’sStages of the Innovation Decision Process model (Figure 4), where theKnot ‘ledge/Persuasion cycle (incorporating the communications channels) impacts the21Decision cycle (which incorporates the behavioural intention, which are shaped by PerceivedCharacteristics of Innovation variables). Finally, the perceptions of several PerceivedCharacteristics of Innovation variables in Figure 1 (ie. Trialability, Visibility, RelativeAdvantage and Image) can be influenced by how other people (eg. Support Group) perceiveor present information technology. From these observations an additional hypothesis can begenerated.H15. The involvement of a Support Group will have a positive influence on Perceivedcharacteristics ofInnovation variables.22CHAPTER 5: 1NSTRUMENT DEVELOPMENTSECTION A - TNTRODUCTION5.1 GENERALIn this section the development of the two questionnaires used in the study will bediscussed. 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 andcontroversial. It is common because it is convenient and often the only feasible way aresearcher can obtain sufficient volume of data in an economical manner. It is controversial asthe method is susceptible to a number of sources of error that could render any resultssuspect. A good questionnaire must therefore strike a balance between its length andcomplexity, presenting to respondents a form that isn’t intimidating, while obtaining data thatis reliable and valid.Moore spent a considerable amount of time establishing the reliability and validity ofhis questionnaire. The changes to the Diffusion of Information Technology questionnairediscussed in the next section were of a type to potentially call into question its reliability butnot its validity. The changes made were generally cosmetic, substituting ComputerizedAccounting System for Personal Work Station and cleaning up terminology to be moreconsistent with a small business environment. These changes were not expected to affect thefocus of the questionnaire from the underlying theoretical foundations, therefore the validityof the questionnaire should not have been affected. Changing the wording of individualquestions may have affected how they were interpreted, which is a reliability issue. As a resultreliability issues will be dealt with in more detail than validity issues.5.1.1 RELIABILITYReliability is defined as “the degree to which the results of measurement are free oferror” (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 a23low 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 forthe item of interest provide similar results (Churchill, 1979).The appropriate level of reliability is a factor of the goals of the researcher andpublished criteria for the type of research being done. Reliability numbers range from 0 to 1and are usually presented as decimal fractions, where the higher the fraction the better thereliability. 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 ofreliabilty 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 shouldbe stressed that .70 is the lower bound for an acceptable reliability level.5.1.2 VALIDITYValidity is defined as ‘the degree to which a measure actually measures what itpurports to” (Nunnally, 1967, pp. 75). In other words, the differences observed are truedifferences for the characteristics being investigated and not a result of some other source(Churchill, 1979). There are several items comprising validity which are summarized inAppendix I-C. It should be noted that not all of these factors may be an important issue withany 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 usedwas previously validated by Moore. Changes made to the questionnaire for this study did notfundamentaly alter what the questions were meant to measure. For example, questions meantto measure Image still measured Image, only the Image being measured was for aComputerized Accounting System (Modified question U-6: Using a CAS improves my imagewithin the organization) and not a Personal Work Station (Original question U-6: Using aPWS improves my image within the organization ). This substitution of CAS for PWSoccured for all 39 questions.245.1.3 QUESTIONNAIRE SELECTIONThe research issues being investigated indicated that two separate questionnaires wererequired. One questionnaire to test the Diffusion of Information Technology model andsimultaneously gather data on users (clients) information technology information sources, thesecond questionnaire to elicit data from accountants.The development of each of these questionnaires is discussed in the following sections.SECTION B: QUESTIONNAIRE DESIGN - PILOT STUDYAll references to questions in this section refer to the Pilot Study questionnaire.5.2 CLIENT/DIFFUSION OF INFORMATION TECHNOLOGY QUESTIONNAIRE5.2.1 PERCEIVED CHARACTERISTICS OF INNOVATIONSOne goal of this research study is to replicate the results Gary Moore obtainedvalidating his Diffusion ofInformation Technology model. Moore spent considerable time andeffort in developing a questionnaire that met suitable reliability and validity criteria (see Moore& Benbasat, 1991). It was determined that redeveloping an alternate questionnaire would beredundant, fitile, and not contribute to a cumulative discipline. Therefore, Moore’squestionnaire was adopted with some minor modifications which are discussed below.In Moore’s study, the measurement of Perceived Characteristics of Innovations wasobtained 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 Innovationvariables that Moore considered integral to the Diffusion of Information Technology model.Based on subsequent analysis of the Diffusion of Information Technology model usingLISREL (Linear Structural Relations Model), Moore was able to determine that only 8Perceived Characteristics of Innovation variables were significant factors. Moore also wasable to determine that the Perceived Characteristics of Innovation questions could betrimmed 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 instrument25unless otherwise noted.In this current study, Moore’s questionnaire was modified by changing all PersonalWork Station references to Computerized Accounting System to focus the study on theinformation technology Computerized Accounting System. Altering the questionnaireintroduced 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 andcompared to Moore’s results to establish that the modifications did not fundamentally alter thereliability of the questionnaire in relation to Moore’s Diffusion of Information Technologymodel. The major risk inherent in this approach is if the pilot study does not producestatistically similar results, it will be difficult to determine if the results are from the changes tothe instrument or from difference between large and small firms. Due to this potentialproblem, an additional pilot study was contemplated to be carried out on a relativelyunmodified version of Moore’s questionnaire. The only modification to this questionnairewould 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’sresults to ensure that the overall integrity of the questionnaire was not damaged. Anydifferences 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 todifferences between large and small firms.As it turned out the results for the pilot study were statistically similar enough toMoore’s findings to dispense with the second pilot study. The pilot study results werecompared using the reported reliability figures (Cronbach’s alpha) for each PerceivedCharacteristics of Innovation variable to Moore’s results. Pilot scores of .60 and higher wereconsidered as acceptable as reliability scores tend to increase with larger sample sizes(Nunnally, 1978). This pilot study had all Perceived Characteristics of Innovation variablesexcept Visibilty (.28) and image (.59) reporting scores above .60 (see Table 1). ThePerceived Characteristics of Innovation variable Visibility had a reliability score much lowerthan the minimum acceptable and was examined more closely. Upon reviewing Moore’s26rationale for using a subset of his original questionnaire it was decided that Visibility could beimproved by adding an additional question to the questionnaire, bringing the Diffusion ofInformation Technology subset of the questionnaire up to 39 questions. This additionalquestion had originally been dropped, by Moore, from the 50 item questionnaire in developingthe 38 item questionnaire. On the whole, the reliability results were encouraging. It wasdecided that the modified pilot study questionnaire would be used in the actual study. ThePerceived Characteristics of Innovation questions were labeled U-i to U-39 for ComputerizedAccounting 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 USAGEThe adoption of a Computerized Accounting System is the dependent variable ofinterest in this study. Like other success measures, measurement of adoption is difficult andsurrogate items are often used, such as system usage. After reviewing the literature it becameevident that usage was commonly measured by using one or two items. This is disturbing asreliability 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 arguesin his thesis, with dependent variables this is not as major a drawback as it is for independentvariables. As validating Moore’s model is an important part of this study, it was determined touse similar usage measures as those used by Moore.Adoption is measured by determining the usage of the information technology (thePersonal Work Station). The usage measures are called Innovativeness. There are threeaspects of innovative behaviour that were measured in his study, these are AdoptiveInnovativeness- 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 givenuse domain, and Implementation Innovativeness - degree to which an individual who hasadopted 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.27Adoptive InnovativenessAdoptive Innovativeness was considered to be the time of first use of theComputerized Accounting System. Two questions were included in the pilot questionnaire tomeasure this item. These questions were day and month the CAS was first used (B—3) and thenumber of months the CAS was regularly used (B-8b). A reliability scale was developed byconverting both questions to an interval scale from 1 to 7 (1=less than one month; 2=between1 and 3 months; 3=between 3 and 6 months; 4between 6 and 12 months 5=between 12 and18 months; 6=between 18 and 24 months; 7=more than 24 months). A reliability score of .97was calculated (see Table 1). The results were encouraging enough to leave these questionsunmodified. While it is preferable to use more than two items for reliablity testing, theresulting reliability score was high enough to indicate that a third question would not berequired.Implementation InnovativenessImplementation Innovativeness was measured by asking questions on hours of use andfrequency of use. The idea behind these questions was to determine the degree of use theComputerized Accounting System was currently receiving.There were two questions for hours of use in the Pilot study, overall weekly use of aCAS in hours (B-4) and weekly use, in hours, broken down by Jthiction (B-8a). Before areliability score could be determined for these two scales, the question on hourly use brokendown by function (B-8a) was converted to a single number by summing the hours of eachfunction 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 thequestionnaires indicated that there were problems that some respondents had in answeringthese questions consistently. The basic problem was that the process of summing hours ofComputerized Accounting System usage for functions in B-8a resulted in a single total thatseldom equaled the hours reported in the overall weekly usage scale (B-4). Often th totalsresulting from adding hours reported in question B-8a were considerably higher than the28overall number reported in question B-4. It was reasoned that individuals are probably morelikely to accurately remember how much they use individual Computerized AccountingSystem functions than to quickly provide an overall estimate of their time using allComputerized Accounting System functions, therefore it was decided to drop the overall CASusage measure (B-4: Overall, how many hours per week do you use a CAS?) from thequestionnaire and to rely on the question measuring CAS usage byfunction (B-8a: On averagehow many hours per week do you spend using the CAS on the following functions?...). It wasalso decided not to develop a replacement question for the item dropped as the bestalternative would have been to obtain actual usage figures. This alternative was not feasible asthe researcher had no access to the respondents’ place of work to measure usage due toconfidentiality. 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 thescales asked the same question, using slightly different wording. Both scales (B-5: Howregularly do you now use a CAS?, B-Il: I have been using a CAS for...) measuredComputerized Accounting System usage in an overall manner. The third question measuredfrequency of use of individual CAS functions (B-7: On average, how frequently do youcurrently 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 perweek, 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 questionmeasuring use by individual Computerized Accounting System functions (B-7) in determininga reliability score due to problems in interpreting these responses. For example, a person coulduse several functions about once a week (indicated by a “4” on the scale for B-7) yet reportusing a CAS more than once per day (indicated by a “7” on the scale for B-5 or B-i 1). Thesedifferent reponses could arise due to the timing of use of each function. This same problemwas noted by Moore.29The two scales (B-5, B-Il) were thought to ask the same question, nleasuringComputerized Accounting Systeni usage in an overall manner, and were included todetermine 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 thesetwo questions. The responses were reviewed as were these two questions. A possibleexplanation is that respondents interpreted B-5 (How regularly do you now use a CAS?) inthe present tense and B-i 1 (I have been using a CAS for...) in the past tense. The inclusionof the same 7 point ranking scale (discussed above) should have caused respondents to answerthe questions similarily. Dropping one of these questions (B-il) was considered; however itwas decided to retain this question to see if similar results would occur in the full study.Use InnovativenessThe Pilot study included four questions designed to measure system usage, called UseInnovativeness 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 weekeachfunction 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 averagenumber of functions used for each question. The reliability score was found to be low, .45 (seeTable 1). Further reliability calculations were performed on a reduced subset of questions andit was found that by dropping the question did the firm use a mainframe or micro (B-6) thescore improved to .77. The Pilot study indicated that all respondents only usedmicrocomputers, which made sense for a small business environment. It was decided to dropB-6 from the final questionnaire for the above reasons.305.2.3 CLIENT COMPUTERIZED ACCOUNTING SYSTEM SUPPORTGeneralAn important part of this study was to examine the role of the support group inComputerized Accounting System adoption. A series of questions were asked regarding themakeup of the support group and the role they play in helping the client with the use of theclient’s Computerized Accounting System.Current SupportFor Computerized Accounting System users, there were five questions designed tomeasure the composition of the support group. These questions were currently receivecontinuing support (B-13), ...iast JO source(s,) of CAS support (B-14), ...where to go fneedComputerized 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 questionmeasured different aspects of support, the results were transformed to a binary measure foreach 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 wasconfusing. A reliability measure of .93 was obtained on the other four questions. As therewere several comments about the length of the questionnaire, it was decided to drop B-i4from the final questionnaire, resulting in a shorter questionnaire and only a minor reduction onreliability.SECTION C: FINAL SURVEYS- SCALE RELIABILITIES5.4 GENERALAlthough full details of the full study are provided in the next chapter, the reliabilityscores 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. Forthe scales measuring Innovativeness and client Computerized Accounting System support,31only the 53 Computerized Accounting System user questionnaires were included, as the 22non-Computerized Accounting System user questionnaires did not capture any of thisinformation.5.5 RESULTSAs shown in Table 2, all of the results are above the minimum 70 except for ResultDemonstrability (.43), and Voluntariness (.69). The reliability scores generally indicate thatthe modifications made in the Pilot study achieved their intended purpose, to produce aquestionnaire with acceptable reliability scores. The implications of these results, includingResult Demonstrability, will be looked at in more detail in the next chapter. It can beconcluded that the scales can be used with confidence across different domains (firm size) anddifferent 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 inreliability appears to be a result of the respondents in the final sample interpreting the twoquestions 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 tobe, while the pilot study group generally interpreted the questions differently. The finalreliability results include respondents from both the pilot study and final study.SECTION D: QUESTIONNAIRE DESIGN5.6 GENERALThe original intention was to follow closely the design and layout used by Moore. Thisapproach was considered the most appropriate as Moores questionnaire design was based onthe Total Design Method which had been designed and tested by Diliman (1978). Thismethod was reported to have resulted in very high response rates. There were some variationsfrom Moore’s approach that were adopted due to a variety of reasons. These are discussedlater in this section.325.7 FORMAT5.7.1 PAMPHLETThe questionnaire was set up in booklet format, with coloured pages separating themajor sections of the questionnaire. A covering letter from UBC was also attached to thefront of the questionnaire. Moore had chosen this format in order to improve the overallappearance of the questionnaire in an attempt to make it appear more professional and worthyof a good response (Moore, 1989).After presenting a copy of the questionnaire to the Partners in one of the accountingfirms participating in the study, and discussing the possible distribution of a similarquestionnaire to their client base, it was determined that some changes would have to bemade. The Partners considered the questionnaire too long in appearance and that many oftheir clients would simply not fill it out, even though the covering letter stated that not all ofthe 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-ComputerizedAccounting System users. This approach was used for a number of reasons. First, it wasexpected that there would be differences between Computerized Accounting System users andnon-Computerized Accounting System users. Separating the questionnaire based on thisconsideration was consistent with the objectives of the research. Secondly, the researcherwould not have direct access to the client base of participating Accounting firms. Because thePartners or someone knowledgeable in each Accounting firm were to do the distribution totheir clients, they would know if the intended recipient was a user or non-user and distributethe appropriate questionnaire. Finally, the questionnaire each potential respondent was toreceive would be approximately half the size as originally designed which should enhancewillingness to participate. These factors made the splitting of the questionnaire practical anddesirable.335.7.2 QUESTION LAYOUTThe questionnaire layout was organized in a manner that emphasized reduction in thenumber of pages. This was done by rearranging the appearance of several of the questions sothat they were horizontally oriented and not vertically oriented. This approach was takenbecause of early feedback received on the apparent length of the questionnaire, even aftersplitting it into two parts. The Partners used to review the questionnaire were very cognizantof 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 specificquestion. At the end of each section encouragement was provided to complete the remainingpart(s) of the questionnaire.5.7.3 COVERING LETTERTwo covering letters were prepared for distribution with each questionnaire. One wasprinted on IJBC letterhead and explained the purpose of the research as well as theconfidentiality of the replies received. The second letter was prepared on the letterhead of theparticipating accounting firm and explained that the firm was not sponsoring the study butbelieved the results would be useful. Encouragement to participate and confidentiality werestressed in this letter also.Both of these covering letters (see Appendix IT-B) were designed after extensiveconsultation with Partners from different accounting firms and with the thesis supervisor. Itwas emphasized to the Partners that the wording of the second covering letter (the accountingfirm letter) was a suggestion only and that they were free to make changes as they chose. Therationale behind this approach was to win Partner support for helping out in the survey byallowing 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.34CHAPTER 6: DATA COLLECTION AND ANALYSISSECTION A: DATA COLLECTION AND CONDITIOMNG6.1 INTRODUCTIONThis chapter will present the data collection and analysis on the final versions of thequestionnaires used in this study. The reasoning behind the sample selection, the data integritychecks performed, the statistical analysis and results will be discussed in some detail. Beforeproceeding 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 inthe introduction and adoption of information technology in small businesses. This type ofinformation is vital, as several research studies have shown that public accountants are notgetting the message out, to their members and to the small business community, thataccountants are skilled information technology specialists (see Hamilton, 1989; Batch, 1989).As part of this analysis, the Diffusion of Information Technology model developed by Moorewill be examined in a Small Business setting, using Computerized Accounting System as theinformation technology of interest. This will be dine in order to evaluate whether (i) theDffusion of Information Technology model is generalizable across firm size and (ii) differentinformation technologies than those examined by Moore when developing this model. Recallthat the major differences between a Computerized Accounting System and Personal WorkStation are that the use of a Computerized Accounting System is usually an organizationaldecision, 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 WorkStation may encompass any functional area in an organization.6.2 SURVEY SAMPLE6.2.1 TARGET POPULATION SELECTIONThe target population is the client base of public practice accounting firms. Most smallbusinesses use a public accountant for tax purposes or for preparation of financial statements.35However, not all firms decide to use a public accountant. There may be differences betweenfirms who use public accountants and those who don’t.2The sample is drawn from the client lists of public accounting firms (CA and CGA). Aconvenience sample of small to medium size accounting firms in southwestern B.C. werecontacted to elicit interest in the study as these accounting firms were the most likely to havelarge numbers of small business clients. Due to the method of selecting the sample certainbiases 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 ofthe province or outside of B.C. A regional bias may exist due to possible differences inindividuals’ attitudes towards adoption of information technology, in southwestern B.C.,relative to the rest of Canada. Because the Diffusion of Information Technology modelmeasures individuals’ attitudes, any bias would affect the results. Other regional biases mayexist at the firm level as southwestern B.C. may have a larger than average number of smallbusinesses concentrated in specific industries. These industries could have their own peculiarrate of adoption, independent of an individual’s propensity to adopt. Also, there could be abias between small businesses in large cities and small cities. Additional regional bias could beintroduced at the public accounting firm level. B.C. public accounting firms could havedifferent levels of knowledge or initiative towards introducing information technology to theirclients.These potential biases inherent in this study should not greatly affect the objectives ofthis study (i.e. generalizability). One objective of this study is to provide a predictiveinstrument that can be used to help small firms successfully introduce an informationtechnology such as a Computerized Accounting System. This objective would be met bysuccessfully replicating Moore’s results. The Diffusion of Information Technology modelshould work as successfully in B.C. as any other province; therefore regional biases should not2Although there is no reliable information on the number of firms that don’t use public accounting firms, thisnumber is generally accepted to be small. Firms that fall into this category include inactive or nearly inactivecompanies. The inclusion of these firms in the study would cause misleading results as IT is not likely to be apriority with low activity firms. Public accounting firms are not likely to be interested in inactive businesseseither, as these firms are not likely to become clients nor pay their accounting fees.36be an issue. Also, members of public accounting firms (CA and CGA) must all take Canadawide exams as well as continuing Professional Development courses. All of theseprofessionals will have a similar educational exposure to information technology which shouldhelp reduce regional differences amongst public accounting firms level of knowledge aboutComputerized Accounting System.Data collection involved the use of survey instruments, with data analysis performedon self-reported data. Directed interviews were considered as a multi-method approach isconsidered appropriate for generating more assurance on the validity of the findings.However, the multi-method approach proved to not be feasible and the directed interviewapproach was abandoned.6.2.2 PROBLEMS ENCOUNTEREDNo different than any other research project, this one had its share of problems fromthe onset. Due to the volume and variety of problems encountered it was considered justifiableto 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 tothe promise of confidentiality made to all participants, especially clients. At one point,arrangements were made with selected accounting firm personnel for follow up interviews, butconditions in the working world interfered with the follow up process to a point where thewhole process was abandoned. Initially, a couple of key people went on two to three weekholidays shortly after agreeing to be interviewed, When they returned it was considered thattoo long a time period had elapsed to put confidence in their responses. Additionally, someparticipating accounting firms (along with participating personnel) decided to back out of theircommitments. It was too late to recruit new participants as the remaining accounting firmshad 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 manyquestionnaires that were delivered to clients during this time were either not filled out or37mailed in. As no facility to follow up on non respondents was available, these lost respondentscould not be recovered. Also, by the time the strike was over, the participating accountingfirms had entered the start of their busy season and distribution of questionnaires was givenlow priority. Regaining the initial enthusiasm exhibited by participating accounting firmsproved to be difficult. Data collection became a tedious task as researcher phone calls wouldoften not be returned and promised actions would not be delivered.6.2.3 RESPONSE RATESA total of 283 questionnaires were distributed to accounting firms and other contactpeople for distribution. Of these, 120 were returned by contacts who had decided to endparticipation in the survey, resulting in a total of 163 questionnaires being distributed tovarious clients. A total of 56 usable questionnaires were returned (no breakdown is availableon how many client firms responded) resulting in a response rate of 34%. This response ratewas lower than expected. A higher response rate was expected as the contact people hadagreed to solicit agreement to participate from their clients before distributing thequestionnaires. Based on follow up discussions with some of these contacts it appears thatsome firms sent the questionnaires out without consulting with the clients, while otherscontacted the clients first and then sent out the questionnaires. It also seems that some clientsdid 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 tothem. This lower than expected response rate resulted in a change in approach to analysing theDiffusion of Information Technology model by using FLS instead of LISREL. It was decidedthat the 19 responses from the pilot study would be included in the data analysis in order tohave enough questionnaires to use PLS. All results reported for the final survey, includingreliability results, included the pilot questionnaires. The pilot questionnaires were included asthere 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 tovarious constraints. Face to face contact with individuals of the participating public38accounting firms (and with selected clients) was required for purposes of cultivating interest inthe 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-ComputerizedAccounting System users as it has been pointed out that one should not leave out the “zerovalue” or control group when exploring the effects of an intervention (Attewell, 1989). Theintervention being controlled for in this case, the Computerized Accounting System, isconsistent with Moore’s approach. However, it is difficult to control for the intervention ofthe public accountant by acquiring data from firms who do not use accountants for anyreason. As discussed earlier, these firms may not exist or would be extremely difficult tolocate. Due to these limitations, any results obtained for the validation of Moores’ model canonly be generalized to firms that use Public Accountants. This limitation to the scope ofgeneralizability is not severe, as it has been previously mentioned that most firms use PublicAccountants.An initial sample size goal of 200-3 00 responses was set in order to accommodate theobjective of testing Moore’s Diffusion of Information Technology model using LISREL.However, as stated earlier, several unexpected problems arose that dramatically reduced thenumber of questionnaires that could be expected to be returned. As a result of these datacollection problems it was decided to use PLS instead of LISREL as PLS is widely consideredan acceptable alternative to LISREL (Barclay et al, 1991).6.3 CLIENT FIRM’S SURVEYThe results from the full study indicated general reliability support for the scales usedto describe the variables in Moore’s Diffusion of Information Technology model. All of thePerceived Characteristics of Innovating variables had reliability scores at the .70 level andhigher except for Result Demonstrability, which dropped from a reliability score of .62 in thepilot study (Table 1) to a reliability score of .43 in the actual survey (Table 2), andVoluntariness which scored .69 (Table 2) dropping from .74 (Table 1). Except for Result39Demonstrability, 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 DEMONSTRABILITYMoore 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 lowerthan Moore’s result of .79 (Table 4). This result could reflect a difference between sampledomains (firm size) or result from use of a subset (3 questions) of the 4 questions used tooriginally define this Perceived Characteristics of Innovating variable. A closer look at theresponses of non-Computerized Accounting System users indicates that the majority of theseindividuals perform non-accounting tasks. This visual analysis is substantiated by Mann-Whitney tests, which confirm that there is a statistical difference between ComputerizedAccounting System users and non-users for Result Demonstrability (discussed later in thischapter, also see Table 7(b)). Additional reliability figures were obtained by obtaining abreakdown between Computerized Accounting System users and non-users. ResultDemonstrability reliability improves to . 71 (Table 3) when Computerized Accounting Systemuser data only is used. A graph of Result Demonstrability non-Computerized AccountingSystem users was generated to determine why no reliability figure could be calculated for thisvariable. Inspection of this graph (Figure 5) shows that the three scales used to measureResult 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 thesame direction for each response. The graph in Figure 5 shows that the scores for each scalemove in opposite directions for each response, in most cases.Further inspection of Table 3 indicates that no other variable showed an obvioussimilar variability in responses by non-Computerized Accounting System users, althoughVoluntariness (alpha=.43) indicated that non-users did appear to have some difficulty with40this measure also.It is not clear why non-users would record responses that were as inconsistent as thoseobserved for Result Demonstrability (and possibly Voluntariness).6.4 CONDITIOMNG THE DATA6.4.1 GENERALBefore the data could be analysed, several steps were required to ensure that theresults would be meaningful. These include checking the data for accuracy, dealing withmissing data, and dealing with outliers. These are discussed below.6.4.2 ACCURACY OF INPUT DATAThe data was originally input into a spreadsheet program by the researcher, who thenrechecked large sections of each questionnaire. A printed copy of the input was thencompared 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. Theresearcher then compared the items identified as being incorrect to the correspondingquestionnaire and made appropriate corrections to the spreadsheet. Very few errors weredetected by the data checkers. With a relatively small sample it is unlikely that there would bemany undetected errors. After these error checking steps the accuracy of the data wasconsidered to be very high.6.4.3 MISSING DATADue to the variety in the types of questions, it was not possible to adopt one globalapproach in treating the data for missing values. Questionnaires that were missing data forlarge sections of the questionnaire were not used at all (there were 2 of these). Multi-itemscales, such as those used to define a Perceived Characteristics of Innovating variable, wouldhave the scale mean inserted if only one item was missing, otherwise the item was coded asmissing.416.4.4 OUTLIERS AND SKEWNESSTypical regression analysis assumes normal distribution of the data. Outliers (data withextreme values) can unduly influence regression results due to their effect on the regressionequation. The regression equations of interest in this study are those relating to the Diffusionof Information Technology model. Data relating to Perceived Characteristics of Innovatingvariables and Subjective Norm variables were reviewed for obvious nonsensical responses.One questionnaire was rejected as all Perceived Characteristics of Innovating questions weremarked neutral (4 on a 7 point interval scale) indicating the respondent had not taken time tounderstand or read the questions. Descriptive statistics were also reviewed to determine ifthere were any other cases of outliers. Except for the non-usei responses to ResultDemonstrability (discussed in a previous section), no others were found.A search for skewness is usually done to determine if the data distribution is normal aswell as whether there may be more outliers. Moore found that his data was generally skewedbut 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 anyquestionnaires could have, whether they were outliers or not, it was determined that therewould be little benefit in performing skewness tests.It should be noted that normal distribution is an underlying assumption of regressionanalysis and for LISREL. However, PLS does not assume data is multivariate normal (Barclayetal, 1991).6.4.5 NON-LINEARITY ANI HOMOSCEDASTICITYAn examination of scattergrams is used to reveal if the relationship between twovariables show linearity (straight line) and homoscedasticity (variability in scores areapproximately equivalent for all values of the two variables). Both of these, revealed by thepresence of an oval shaped scattergram, are required assumptions for multivariate regression.42Scattergrams were produced for the variables of interest and no significant violations of thesetwo assumptions were detected. Thus the data appeared to be of good quality for furtheranalysis. The accuracy was found to be high and missing data was minimal.SECTION B: DESCRIPTIVE STATISTICS6.5 GENERALAs well as the demographic data generated (Table 5), various descriptive statisticswere generated for the research variables including the mean, standard deviation, andmaximum and minimum reported values. These are summarized in Table 6. A comparison ofComputerized Accounting System users vs. non-Computerized Accounting System users isprovided in Table 7(a) and 7(b), including Mann-Whitney U test results. These results will bediscussed in detail later in this chapter. The Mann-Whitney U test is used to determine if thereare differences between Computerized Accounting System users and non-ComputerizedAccounting System users. The Mann-Whitney (M-W) test is used in order to avoid relyingupon 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 onthe variables of interest are provided in Table 8 through Table 11. Regression results arediscussed in the following sections.General comparisons will be made to Moore’s study, based on whether the resultssupport the hypothesis that is being tested. Specific numerical comparisons will be made toMoore’s study where the results from this study differ from Moore’s. A summary of resultsfrom hypothesis testing for this study can be found in Table 13(a), and for Moore’s study inTable 13(b).References to question numbers will refer to the final questionnaire (Appendix Il-Al)unless otherwise noted.436.6 DEMOGRAPHICSDemographic data is summarized in Table 5 with Adjusted Frequency figures used(these are corrected for missing data). The general categories reported on include Departmentof Employment; Organization Level; Education; Age; and Sex. Where relevant, comparisonsare made to Moore’s survey.The main focus of the data gathering effort was the accounting/finance function. Atotal of 51.5% of respondents were engaged full time in the accounting area. The remaining48.5% were distributed throughout other departments, including Administration (19%) andOther (29%) - “Other” consisted of areas not falling into Accounting or Administration. Inmany 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% ofrespondents were at the clerical level. This response rate seems to indicate that the targetedindividuals in the client firms were reached.There is a surprisingly high level of respondents that did not obtain education beyondhigh school (18%) while another 10% received some training from a trade school. Theremaining 72% received some College/University education, including 8% with Postgraduatedegrees. [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 isnot 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 themethods of gathering the above information differ and may cause the perceived differencesnoted (ie. Moore had respondents gathered into a room to fill out the questionnaire, somepotential respondents may have had to stay behind to “run the shop” and these may have beenthe younger employees). However, there appear to be definite differences in the age groups of44employees working in smaller firms.The SEX profile is also in sharp contrast to Moores study. This study had 33% malerespondents 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 tosome of this difference.The overall demographic profile of this study indicates sharp differences from Moor&ssurvey. Respondents are generally younger, more likely to be female, and have less formaleducation than in larger firms. These findings generally support earlier studies (discussed inChapter 2) on demographic characteristics of people employed in small to medium sized firms.6.7 ATTITUDE TOWARDS INNOVATTNGThe dependent variable Attitude was generated from a four item semantic differentialscale (good-bad; harmful-beneficial; wise-foolish; and negative-positive) in response to thequestion Overall, my using a C’AS in my job is (B-i). Various descriptive statistics weregathered on Attitude. These statistics are based on all 75 questionnaires. On a seven pointscale (lmost positive, 7=most pessimistic) an overall average of 2.2 (Table 6(a)) indicatesthat 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 asignificant 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 ofdetermining to what extent the overall mean of 2.2 is influenced by users and non-users. Thedescriptive statistics results in general, and M-W results for users specifically, indicate supportfor Hj [One attitude towards i/sing Computerized Accounting Systems will influence one’sinnovativeness with respect to Computerized Accounting System usagej. The claim forsupport of Hj is based on the assertion that the more positive the attitude the more aComputerized 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 issupported. This is similar to Moore’s findings.456.8 PERCEIVED CHARACTERISTICS OF INNOVAHNGPerceived Characteristics of Innovating scales were recorded so that higher numbersreflected a higher degree of agreement with the perception associated with that variable. All ofthe Perceived Characteristics of Innovating variables except for Voluntariness (3.1) had amean score of 4 (neutral) or higher (Table 6(a)). The most positive Perceived Characteristicsof Innovating variables are Relative Advantage (5.50), Compatibility (5.43), and ResultDenionstrability (5.19).Based on the M-W test, all of the Perceived Characteristics of Innovating variables aresignificantly different between Computerized Accounting System users and non-users at thep<.O5 level (Table 7(b)) except Ease of Use (.14). All of Moore’s Perceived Characteristics ofInnovating variables were significantly different at p<.O5. The uniformity of scores for thevariable Ease of Use amongst all small/medium firm respondents may be a result of the closerworking relationship amongst users and non-users contributing to common opinions aboutComputerized Accounting Systems. There is support for H2 LfRelative Advantage will have acontribution more than any other Perceived Characteristics of Innovating on one’s attitudetowards 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 tohave 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 themethod used for Perceived Characteristics of Innovating, with higher scores indicating amore positive response. As discussed previously in this section, Voluntariness had a meanscore of 3. 13 (Table 6(a)), which indicates a more unfavorable (negative) response than theseven Perceived characteristics of Innovating variables. The M-W test (Table 7(b)) showsthat Computerized Accounting System users means (2.72) are significantly lower than non-Computerized Accounting System users means (4.13), indicating support for H646[Voluntariness is negatively related to one’s innovativeness with respect to ComputerizedAccounting System usagej. This finding is the same as Moore’s.6.9 SUBJECTIVE NORMSValues for Subjective Norm scores were calculated by multiplying the NormativeBelief (ranging from 1 to 7) by the Motivation to Comply (ranging from -3 to +3). The rangeof scores could vary from -21 to +21. The mean scores reported in this study (Table 6(a)) aremixed 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 betweenComputerized Accounting System users and non-Computerized Accounting System users, atp.O5, arise from Senior Management (.0i9) and Subordinates (.003). In general, H4 [TheSubjective Norm it//I injinence oiies innovativeness with respect to Coniputerized AccountingSystem usage] is not supported. This differs from Moore’s study where H4 was supported (allof 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 indicatedifferences between large and small firms. In smaller offices, employees are more likely tointeract with people in other functional areas (cross-pollination of ideas) and the influence ofreference groups would be more uniform. Large firms would likely have less uniform opinionsabout reference groups due to the lack of interaction with people in other functional areas.6.10 INNOVATIVENESS MEASURESInnoi.’ativeness was previously discussed in chapter 5. The item usage, the surrogatefor adoption, was measured in four different ways: months since first use of ComputerizedAccounting System, hours of use, frequency of use, and number of functions used. Descriptivestatistics for each of these measures can be found in Table 6(b). Because Innovativenessinformation was only gathered for Computerized Accounting System users, M-W tests couldnot be run on Innovativeness variables and N’A appears for the boxes where statistics are notapplicable in Table 7(b). As a result of this data gathering approach, mean scores reported in47Table 6(b) and Table 7(b) for all Innovativeness variables are identical.The Innovative measure Months elapsed since Computerized Accounting Systemadoption was calculated by taking the average of the two measures time offirst CAS use (B3) and (‘AS use by function, in months (B-6(b)). An average of 56 months (Table 7(b)) wascalculated. 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 beenheld that larger firms adopt information technology before smaller firms. Perhaps theparticular information technology of interest, Computerized Accounting Systems, diffuseearlier than the other Personal Work Station items that Moore examined. It should be notedthat no statistical tests were done to determine if the values for both studies were significantlydifferent. If such tests were run it is possible that they could show no statistical difference inadoption periods between the two studies.Hours of use of Computerized Accounting System per week is calculated by using asingle 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 userand an overall average was calculated from the total hours calculated, for all ComputerizedAccounting System users. An average of 21.6 hours per week (Table 7(b)) is more than the15.9 hours reported by Moore. This average indicates that accounting/finance employeesspend a good deal of their time with Computerized Accounting System. No statistical testswere performed to determine if the values for both studies were significantly different.Frequency qf Computerized Accounting System use is calculated in two ways. In thefirst method, a general frequency of use is calculated by taking an average of the results forthe two questions which ask how long the GAS user has been using the CAS (B-4 and B-9) asthese two items ask the same question. Both items consist of a seven point scale, and anaverage 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), wereobtained by summing the coded values from a seven point scale (1=not at all, 4about onceper week, 7=more than once per day), for each of the eight applications. Ranges of values for48an individual Computerized Accounting System user could vary from 53 (didn’t use anyComputerized Accounting System applications) to 56 (used all eight applications more thanonce 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 12functions to the 8 Computerized Accounting System items identified in this study. As noted inthe footnote, the average value reported in this study may be understated as well. Nostatistical tests were performed to determine if the values for both studies were significantlydifferent.The number qf functions used is calculated by averaging the responses to thequestions asking the frequency (?f use byjiinction (B-5), how many hours per week each CASfunction 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 shouldhave had an answer. Each question was coded a zero (0) for no response or a one (1) for aresponse. By averaging the responses to each function for the three questions, effects frommissing data was likely to be minimized. An average of 4.5 Computerized Accounting Systemfunctions (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) thanPersonal Work Station functions used (5.9/12). It is not clear if this difference is due to theselection of functions. The Computerized Accounting System functions are basically a subsetof 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 informationtechnology. No statistical tests were performed to determine if the values for both studieswere 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. Themethod chosen to record the responses resulted in a “Minimum Score” of 4 instead of the theoretical 8discussed. This approach may result in understated Frequency of use results.496.11 COMPUTERIZED ACCOUNTING SYSTEM SUPPORTThis 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 isavailable from non-Computerized Accounting System users, therefore N/A appears in theother columns. Mean scores tabulated in Table 6(b) are identical to those in Table 7(b), as nodata from non-Computerized Accounting System users was gathered for ComputerizedAccounting Sysieni Support. Scores are tabulated on a seven point scale (1not at all, 4=onceper week, 7=more than once per day). All four sources of Computerized Accounting SystemSupport 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, noone group appears to be more dominant than any other. These results do not support thefollowing hypothesis: H8 [The involvement ofa Support Group will contribute to a successfuladoption of a Computerized Accounting SystemJ, H9 [The involvement of a Friend willcontribute to a successful Computerized Accounting SystemJ, H10 [The involvement of otherEmployees will contribute to a .s’uccessfui Computerized Accounting SystemJ, Hjj [Theinvolvement of an external Accountant will contribute to a successfiui ComputerizedAccounting SystemJ; and Hp [The involvement ofan external Consultant will contribute to asuccessful Computerized Accounting SysteniJ.SECTION C: REGRESSION ANALYSIS6.12 GENERALMoore’s research hypotheses and Diffusion of Information Technology model, as wellas Computerized Accounting System Support, were tested using multiple regression and PLS.This was done by examining the effects of the different independent variables (PerceivedCharacteristics of Innovating, Voluntariness, Attitude, Subjective Norm, and ComputerizedAccounting System Support) on each of the Innovativeness measures. The results arediscussed in this section. PLS results are discussed in the next section.506.13 THE EFFECT OF PERCEIVED CHARACTERISTICS OF INNOVATIVNESS ANDVOLUNTARINESS ON ATTITUIEThe initial regression model analysed was the seven Perceived Characteristics ofInnovating variables and Voluntariness on Attitude. The procedure followed paralleled that ofMoore (1989). A STEPWISE regression was run, with the F-value probability set at p<.O5 forentry and p>. 10 for removal of a variable once in the equation. Following this regression, asecond regression was run where all variables were forced into the equation in the same orderas the STEPWISE regression. The end result of the forced entry procedure is to produce aregression with all variables in the equation, but the stepped entry allows the directcontribution of each variable to R2 to be examined.The regression results on the full equation provide an R2.776 and an adjustedR2=.749 (Table 8), indicating that the Perceived Characteristics of Innovating variables aresignificant in the formation on Attitude towards using computerized accounting systems.. Theregression results indicate that the various Perceived Characteristics of Innovating variableshave 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 ofthe other variables contribute to R2 in any meaningful way. These results are summarized inTable 8 where part I lists the results for the forced step regression and part II lists the resultsfor the regression on the full equation. [Moore’s study had an adjustedR2.677, and more ofthe Perceived Characteristics of Innovating variables were significant].The regression results support H2 [Re/alive Advantage will have a contribution morethan any other Perceived characteristics of Innovation on one’s attitude trniards adoptingComputerized Accounting Systems], as Relative Advantage’s contribution to R2=.73 while thefull equation had anR2=.78. This result was similar to Moore’s.There was no support for H7 [Voluntariness will be negatively related to one’sattitude towards using Computerized Accounting Systems], as Voluntariness was notsignificant. Moore’s results showed a negative Beta for Voluntariness (which was significant),51and 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 wasto examine the influence of Subjective Norm on Attitude. A composite score for SubjectiveNorm was computed by summing the scores of the individual Subjective Norms for eachmeasure.The regression results are provided in Table 9, PCI and SN column. There is very littlechange from the results of the regression without Subjective Norm as Subjective Norm is notstatistically significant in this regression. Thus H5 [The Subjective Norm i4ll influence one’sattitude toward adopting the Computerized Accounting Systenzj is not supported. In Moore’sstudy H5 was supported as Subjective Norm was significant.The difference in results between this study and Moore’s may be due to the differencesin sample size as well as sample selection. The individual reference groups comprising thecomposite Subjective Norm are Friends, Peers, Superiors, Subordinates and SeniorManagement. Individual scores for each reference group could range from -21 to 21 (this wasdiscussed in Chapter 5). The composite measure calculated for Subjective Norm will thereforebe a neutral value (near zero) if the reference groups are not that important to the respondentor if the scores are extreme on either side of zero. A small sample size may not be able todifferentiate between these two possible explanations. A larger sample size would indicatesignificance, 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 scoreswere near neutral. A larger sample size would not help if two reference groups with oppositescores were dominant.Sample selection, and possibly firm size may be a factor as larger firms tend to haveestablished cultures and prevailing opinions on information technology use. Moore’s sample52consisted of individuals from six large firms, whereas this study contained responses frommany more (smaller) firms, possibly 20 or more. Moore’s respondents would likely show amore dominant culture affecting Attitude as, at most, there are six different cultures andpossibly less. A dominant culture could emphasize one reference group or combination overanother. It would be difficult to determine if Moore’s significant results for Subjective Normare due to the small sample size of large firms or the presence of a corporate culture that iscommon to large firms. The current study could have up to 20 or more cultures, which mayresult in no clearly dominant reference group. As no significant results for Subjective Normwere found in this study, there is a likelihood that there is no dominant corporate culturecommon to small businesses in general.The current regression involving Subjective Norm does not provide enoughinformation to determine if the hypothesis being tested is being correctly rejected (oraccepted).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 ComputerizedAccounting System Support. The purpose of this regression was to determine the influence ofComputerized Accounting System Support on Attitude in order to examine the extensionsmade to the Diffusion of Information Technology model. This regression was run withComputerized Accounting System Support included as a composite score (individual supportgroups consisted of Friend, Employee, Accountant, and Consultant). The regression resultsindicated that the composite Computerized Accounting System Support score was notsignificant in the regression equation (p=.3 1). These results are summarized on Table 9, PCLSN & SUPPORT column. H14 [The involvement of a Support Group wi/i have a positiveinfluence on Attitude], was not supported.The regression results on Support clearly indicate that this variable has no effect onAttitude. This result is unexpected and may be an artifact of the variable Attitude (discussedbelow).53The results of the regressions on Attitude, while not as supportive of the Diffusion ofInformation Technology model as Moore achieved, raise the same issue - theoperationalization of Attitude. Only Relative Advantage (p.OO), Visibility (p=.O4) and Easeof Use (p. 10) at p.lO were statistically significant (Table 8). Relative Advantage by itselfprovides an R2 of .73 while the full equation has an R2 of .78. The finding that Attitudecaptures Relative Advantage and not the Perceived Characteristics of Innovating variables ingeneral, 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 ofthe poor operationalization of Attitude.6.14 THE EFFECT OF ATTITUDE, SUBJECTIVE NORM, PERCEIVEDCHARACTERISTICS OF INNOVATIVENESS, VOLUNTARINESS ANDSUPPORT ON INNOVATIVENESS6.14.1 GENERALRegression analysis was also run on the independent variables and Innovativeness. Aninitial regression used Attitude, Subjective Norm, and Voluntariness as independent variables.An additional independent variable, Support, was added in a subsequent regression run toexamine the influence of this variable on Innovativeness. A different regression on individualPerceived characteristics of Innovating variables and Subjective Norm variables was alsorun. This second regression omitted the intervening variables AttItude and the overallSubjective Nonii measure. Once again Support was added in a subsequent run to measure itsimpact on Innovativeness.6.14.2 ATTITUDE, SUBJECTIVE NORM AND VOLUNTARINESS ONINNOVATIVENES SFour regressions were run, one for each of the dependent Innovativeness measures(Number of Functions Used, Frequency of Use, Months Since Adopted, and Hours of UsePer Week). The dependent variables for each regression run were Attitude, Subjective Norm54and Voluntariness. The results of the regressions, summarized in Table 10(a), will bediscussed in the following paragraphs. The regressions were done with all independentvariables and each dependent variable entering the equation at once.Subjective Norm is not significant in any of the regressions. H4 [The Subjective Normwill 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 ComputerizedAccounting System]. Moore’s study supported H7 for all Innovative measures.Attitude is significant for all Innovativeness variables, although the betas are allnegative, rejecting Hj [One’s attitude towards using Computerized Accounting Systems willinfluence one’s innovativeness with respect to Computerized Accounting System usage].Moore found that Attitude was significant for all Innovativeness variables and all had positivebetas, 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 ComputerizedAccounting 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 ahigh of .223. The variations in R2 indicate that the independent variables capture differentdegrees of the variance in the different forms of innovativeness. This low range of adjusted R2values 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 AccountingSystem Support as an independent variable in the aboye regression equation. The results fromthese regressions are summarized in Table 10(b). The composite Computerized AccountingSystem Support measure was used in this regression. Computerized Accounting SystemSupport was significant for all Innovativeness variables. This result supports H8 [Theinvolvement of a Support Group will contribute to a successful adoption of a ComputerizedAccounting Systenij.With the addition of Computerized Accounting System Support, the influence of theother independent variables on the various Innovativeness variables changed. With theaddition of the Support Group variable (compared to the regressions without the SupportGroup variable), Voluntariness became significant for Hours (p=.O5 vs p=. 14), while Attitudebecame 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, theinclusion of Computerized Accounting System Support has weakened the importance ofAttitude on Innovativeness (ie. weaker support for Hj). This effect indicates thatComputerized Accounting System Support has an influence in the Diffusion of InformationTechnology model.6.14.3 PERCEIVED CHARACTERISTICS OF INNOVATING, SUBJECTIVE NORM,ANT) VOLUNTARINESS ON INNOVATIVENESSIndividual Perceived Characteristics of Innovating variables and individual SubjectiveNorm variables were regressed on the dependent Innovativeness variables in order to measurethe magnitude of their direct effects on adoptive behaviour. As each individual PerceivedCharacteristics of Innovating and Subjective Norm measure were expected to have differentinfluences on the dependent variables, STEPWISE regression was used with the probability toenter 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 tosubsequent regression runs. In this case, the individual Support Group measures were usedinstead of the composite scale in order to examine the influence each scale has onInnovativeness. Regression statistics without Support Group are included in Table 11 (a) andwith Support Group in Table 11(b).For the regression without Support Group, the Subjective Norm variables weresignificant 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). The56overall Subjective Norm results weakly support Hq [The Subjective Norm wi/i influence one’sinnovativeness i’ith respect to Computerized Accounting System usagej. This result is similarto Moore’s where Subjective Norm variables found to be significant were Functions Used andFrequency.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 notsupporting H6 [Voluntariness is negatively related to one’s innovativeness with respect toComputerized Accounting System usagej. Moore’s results supported H6 for all Innovativenessmeasures.Relative Advantage was significant for Functions (p=.O5), Frequency (p’.OO) andweakly 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 onInnovativeness. 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 resultsupports H12 [The involvement of an external Consultant will contribute to a successfulComputerized Accounting Systemj.Personnel was significant for Functions (p=.OO) and Frequency (p.OO), moderatelysupporting H10 [I he i ivoivement of other Employees till contribute to a successfulComputerized Accounting SystemJ.Accountant was not significant, rejecting Hjj [The involvement of an externalAccountant will contribute to a successful Computerized Accounting Systenij.Friend was not significant, rejecting H9 [The invoii.’ement of a Friend will contributeto a successful Computerized Accounting SystemJ.57The overall regression results for the equations including Support Group variablesprovide support for H8 [The involvement of a Support Group vili contribute to a successfuladoption of Computerized Accounting SystemJ.Voluntariness became more significant for Hours (p=.Ol) than without the presence ofSupport Group while Visibility became not significant for Hours. There were less WeaklySignficant variables once Support Group was added to the regression equation.6.14.4 OTHER REGRESSIONSA series of regressions of Support Group, the independent variable on SubjectiveNorm, Attitude and Perceived Characteristics of Innovating, as dependent variables, wererun. None of these regressions were done by Moore, therefore no comparative statistics areavailable. 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 (SupportGroup in this study) and Subjective Norm (see Figure 3). The individual variables of SupportGroup were regressed against the composite Subjective Norm. Regression results indicatedthat the individual Support Group variables were not significant and H13 [The involvement ofa Support Group will have a positive influence on Subjective Norinj was not supported.The regression Support Group on the individual Subjective Norm variables was thenrun to see if the composite Subjective Norm was masking its individual components. WhileSupport Group Personnel was the only significant variable, the low adjusted R2’s and F’sindicate that the relationship between Support Group and Subjective Norm is weak at best.The regression Support Group on Perceived Characteristics of Innovating wassuggested by both the Fishbein & Ajzen model and Moore’s model. The Fishbein & Azjenmodel indicate a link between Communications Network and Attitude Towards Adopting58(Figure 3). Moore’s model indicates that Attitude Towards Adopting is determined by thePerceived Characteristics of Innovating variables (Figure 1 ).4 It seems logical to regressCommunications Network (Support Group) on the individual Perceived Characteristics ofInnovating variables if they have become a surrogate for Attitude.Regression of the individual Support Group variables were run against the PerceivedCharacteristics 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 Groupi’ili have a positive influence on Perceived Characteristics ofInnovation variablesJ.A regression of Support Group on the individual Innovativeness variables was runwithout the presence of other independent variables in order to determine how large an effectthe individual Support Group variables have on Innovativeness. The results are summarized inTable 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 asignificant effect on Innovativeness. The regression results support H8 [The involvement of aSupport Group will contribute to a success!;’,! adoption of computerized accounting systemJ,H10 [The involvement of other Employees will contribute to a successfiui computerizedaccounting systemJ and Hp [The im.’olvement of an external Consultant u ‘ill contribute to asuccessful computerized accounting systemj. These results were consistent to the earlierregression results reported (Table 11(b)).4As reported in a previous section of this paper. regression results indicated that some of the PerceivedCharacteristics of Innovating (PCI) variables better represented the concept “Attitude” than did the compositescales designed to represent “Attitude”. When Moore used LISREL to test his model, he used these PCIvariables to represent the concept “Attitude” instead of the original Attitude scales.59A regression of Support Group on Voluntariness was run to determine if there wasany 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 werefairly low.The rationale for running this regression was to see if the perception of Voluntarinesswas influenced by the Support Group. The results do not support this.SECTION D: PATH MODELINGSince there were some problems with simple regression analysis that caused difficultyinterpreting results for the Di/flision of Information Technology model for both studies theAttitude variables generated may not have captured the construct Attitude. It also appearedthat the construct Subjective Norm (Subjective Norm) may not have been appropriatelyspecified, because the regression analysis indicated that individual Subjective Norm variablesaccounted for more variance on the dependent variables than did Subjective Norm itselfAdditionally, the construct Innovativeness could not be generated using normal regressiontechniques due to the differences in the scales of the Innovativeness variables (eg. months vshours vs functions used). These factors indicated that an alternate method to regressionanalysis would assist in constructing Subjective Norm, Attitude and Innovativeness from theirindividual components. One such method is known alternatively as causal modeling (Barclayet a!, 1991; Bagozzi, 1982), structural equation modeling (Fournell et a!, 1982) or pathmodeling (Wold, 1985). For convenience the term path modeling will be used throughout thispaper.60Path modeling utilizes second generation5multivariate analysis techniques in order toobtain statistical information that cannot generally be obtained by first generation statisticaltechniques (Barclay et al, 1991; Dimnik, 1986). Path modeling is a method of research, andcan be used to determine internal consistency, reliability, construct validity, and for hypothesistesting (Bagozzi, 1982). Borrowed from econometrics (path models and manifest variables)and psychometrics (latent variables) (Wold, 1985), all path models have in common the traitsof latent variables linked to manifest variables by paths.6.15 CHOICE OF PATH MODEL COMPUTER IMPLEMENTATION - LISRELvsPLSThe two most common computer implementations of path modeling are LIS]?EL andPartial Least Squares (PLSI. Both of these programs have their strengths and weaknesses foranalysing models. The choice of which program to use depends on the stage of theorydevelopment being tested and the goals of the researcher. PLS is generally used in the earlystages of theory development while LISI?EL is better suited to models based on welldeveloped theory. LISREL is based on assumptions of multivariate normality in data whereasPLS requires no such assumptions. LISREL requires large sample sizes while PLS can be usedwith much smaller sample sizes (Barclay et al, 1991).Afier analysing the various characteristics of this study it was determined that the useof PLS would be most appropriate. This study is examining the Di/,/iision of InformationTechnology model first developed by Moore. While the theories underlying this model havebecome established in other fields, the synthesis of Rogers and Fishbein & Ajzens theories hasnot been tried before. The applicability of this approach has yet to be firmly established.Additionally, preliminary results from regression analysis indicated that multivariate normality5The term “second generation” is used to denote the use of more sophisticated mathematical models and statisticalcomputer programs. A second generation multivariate technique must meet four requirements (Fournell, C., ASecond 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 errorsin 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 componentsanalysis are all special cases of canonical correlation ... which itself is a special case of PLS ...“ (Barclay et. al.1991, pg. 4).61assumptions may not apply to the data in this sample. Finally, a relatively small sample sizewas obtained which indicated that the use of LISREL would not be feasible.6.15.1 DESIGN OF PLS PATH MODELPLS 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: oneequation describing the path (links) of each MV to each LV (outer design matrix); the secondequation describing the path connecting LVs to each other (inner design matrix). TheDiffusion of Information Technology model, with iWVs’ and LVs identified, can be seen inFigure 6. This model design is comparable with the LISREL model used by Moore (see Figure3).The outer design matrix for the Diffusion of Information Technology model isillustrated in Figure 6. Latent Variables are linked to Manifest Variables by paths. Paths canflow in either direction (indicated by arrows), depending on the underlying theory supportingthe model.(1) LV Subjective Norm has as its indicators MVs Supervisors, Peers, Senior Ivianagement,Subordinates, Friends and Perceived Voluntariness. The MV, Friends, which wasomitted from the Moore’s LISREL analysis, was included in the PLS analysis in order tofully analyse the Df/iision (?f Information Technology model The MV indicatorPerceived Vohintariness has been included because the regression results indicated astrong interaction with Subjective Norm. The inclusion of this All7 is consistent with theapproach 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 ofUse, Trialability, Visibility, and Result Demonstrability. These MVs are actuallyPerceived Characteristics ofInnovating indicators and not the original indicators derived62for Attitude. As discussed in the regression analysis section, the PerceivedCharacteristics of innovating indicators had a greater direct effect on Innovativenessthan 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 couldbe derived from the Perceived Characteristics ofInnovating indicators.(4) LV Innovativeness has as its indicators MVs Hours, Months, and Functions. Oneindicator, Frequency of Use, was not included in the model due to the difficulty ininterpreting the composite scale. Although Moore’s analysis indicated that MV Functionsand Frequency “may be tapping the same dimension of Innovativeness” (Moore, 1989,pg. 186), results from this study indicate that they may actually be tapping differentdimensions (see Table 11(a) and Table 11(b), noting the differences between thesevariables regression results).The inner design matrix has the path structure indicated in Figure 6. Path links in theinner 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 underlyingpremise of the model as set out by the model builder (Lohmoller, 1984). In the current modelthe scales are standardized and path loadings between LVc and MVs’ represent the relativeimportance of the composite scale score to the LV. The path loadings between LVs can alsorange 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 REQUIREMENTSWith PLS, a more modest sample size than with LISREL is used, because the lessrigorous statistical assumptions require a minimum sample size often times: (1) the number ofindicators from the most complex /brmative construct; or, (2) the largest number of predictors63leading to an endogenous construct (Barclay et al, 1991). A forniative construct (or LatentVariable 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 withlviVs 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. Theconstruct 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 versusparameters to estimate. PLS can deal with this situation because ... the iterative algorithmbehind 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 samplerequired 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 theirassociated LV depends on the researchers prior experience with the model and theunderstanding of the real world situation being studied. If the constructs are not welldeveloped then the IvIVs for that construct are considered formative. For the purposes of FLSanalysis of the Dffusion of Information Technology model (Figure 6) only the SubjectiveNorm MVs will be treated as formative indicators, while Subjective Norm andCommunication Network will be treated as formative in the extended model (Figure 7). TheMV for the other LVs will be treated as reflective indicators.In Figure 6, Subjective Norm and Voluntariness are exogenous L Vs, while Attitudeand 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 isInnovativeness with three (Subjective Norm, Voluntariness, and Attitude). This wouldindicate a minimum sample size of 50 (10 times the 5 Subjective Norm MVs).In Figure 7, Voluntariness and Communications hannei are exogenous L Vs, whileSubjective Norm, Attitude and innovativeness are endogenous L Vs. The largest number offormative indicators is five (Subjective Norm) while the endogenous LV with the largestnumber of predictor LVs is Innovativeness with four (Communications Channel, SubjectiveNorm, Voluntariness, and Attitude). This would indicate a minimum sample size of 50 (1064times the 5 Subjective Norm MVs).A sample size of 50 is much less than the sample size of 500-600 Moore required forhis initial development of the Diffusion of Information Technology model using LISREL. Thetotal number of usable questionnaires available for analysis is 75, which exceeds the minimalrequired sample size.6.15.3 GOODNESS OF FIT DETERMTNATIONThe Diffusion of Inforniation Technology model illustrated in Figure 6 was assessedby comparing PLS statistical results to various reduced versions of this model. This approachwas used as PLS does not have any single goodness of fit” measure. Three commondiagnostics used for PLS analysis are based on root mean square (RIVIS) covariances. Theseare 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 principalcomponent. 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; andn is the sample size.F2 measures the average squared multiple correlation between each endogenousconstruct and all exogenous constructs. The redundancy is the proportion of the variance thatcan 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 whereasCommunality applies to MVc (Lohmoller, 1984):F27, = 1 - SSEJ, / SSO/165Generally, the fit of the inner model is satisfactory if R2 is high enough; the fit of theouter model is satisfactory if H2 is high enough; and the fit of the total model is satisfactory ifF2 is high enough (Lohmoller, 1984). Determinig if there is a fit or not is clearly a judgementcall.The models which were compared to the full Diffusion of Information Technologymodel in Figure 6 were determined by eliminating the exogenous LV Voluntariness, then bothexogenous LVs (Voluntariness and Subjective Norm). Finally, a model that eliminated onlylow scoring Perceived Characteristics of Innovating variables was generated. The samemodels were run again, this time using only data points from Computerized AccountingSystem users. One final model was run which extended the model in Figure 6 with the addedLV Communications Channel (see Figure 7). The results from these comparisons arecontained in Table 14.The Diffusion of Information Technology model was run using PLS, with a number ofdifferent 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); minusboth Voluntariness and Subjective Norm (R2=. 14, H2=.28, F2=.04); minus four PerceivedCharacteristics 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 Systemuser data only (53 subjects). This set of PLS runs was generated because previous SPSSanalysis had indicated that there were significant differences between ComputerizedAccounting System users and non-users for several of the MV (refer to Table 7(b)). Whilethe sample size of 53 was somewhat less than the rule-of-thumb requirement for a minimumsample of 70, it was expected that this would not greatly affect results. Results were not thatdifferent from the full data set for the full model (R2=. 10, H2=.42, F2.05); minus LVVoluntariness (R2=. 10, H2=.36, F2=.04); minus both Voluntariness and Subjective Norm66(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 individualscores 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 regressionanalysis 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 onindividual 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 originalmodel (Attitude .28; and Innovativeness = .28; not calculated for Subjective Norm). Thedifferences in the “fit” indicators for the various models are not that large and would probablynot be statistically different from the original model. The extended model may have moreexplanatory power than the original model due to the presence of indirect affects thatCommunications Channel has on the other LV. The overall small numbers for the fitindicators suggest that the increased explanatory power may also not be statisticallysignificant.Based on these analysis, no alternative model was shown to be superior (on aqualitative basis) to the original Diffusion of Information Technology model represented inFigure 6. A model with four Perceived Characteristics of Innovating variables removedindicated higher scores on the indices examined, however there were no theoretical groundsfor removing these MVs. Compared to the original model, an extended model had slightlyimproved direct explanatory power and additional indirect explanatory power.6.15.4 ASSESSMENT OF HYPOTHESES TESTIIJGAs no tests for statistical significance between the various models have been done it isnot possible to quantitatively evaluate each hypothesis. Qualitative interpretations are possible67however, based on analysis of path loadings and R2 results. The first seven hypothesesdescribed in Chapter 4 are based on the full model (Figure 6) while the remaining eighthypotheses are based on the extended model (Figure 7). The various hypotheses andconclusions are presented below:Hi to H7 are analyzed based on the original Diffusion of Information Technology modeldescribed in Figure 6 and the extended model described in Figure 7. Based onqualitative analysis the conclusions for some of these hypotheses change dependingon which model is used.Hj. One’s attitude towards using Computerized Accounting System i’ill influence one’sinnovativeness t’ith re5pect to Computerized Accounting Systeni usage. Thehypothesis indicates that a positive coefficient is required to increase innovativeness.For the original model (.624) and the extended model (.440) the path coefficient ispositive thus Hi is supportedH2: Relative Advantage wi/i have a contribution more than any other PerceivedCharacteristics oJ Innovating on one’s attitude towards adopting ‘omputerizedAccounting Systems. Path loading for MV Relative Advantage on LV Attitude is thelargest for the original model (.9141) and the extended model (.9180), supportingH2.H3: C’omputer Avoidance i’iii have a contribution less than any other PerceivedCharacteristics of Innovating on one’s attitude towards adopting ComputerizedAccounting Systems. This hypothesis is not explored in this study.114: The Subjective Norm will influence one innovativeness with respect toComputerized Accounting System usage. For the original model (-.06 1) and for theextended model (.056) the path coefficient have a very small loading value indicatingno support for H4,H5: The Subjective Norm i’iii influence one’s attitude toward adopting the ComputerizedAccounting System. The original model (.122) and extended model (.053) have verysmall path coefficients, which indicate no support for H5.68“6• Voluntariness is negatively related to one’s innovativeness iith respect toComputerized Accounting System usage. The path coefficient of for the originalmodel (.324) and for the extended model (.3 13) indicates rejection ofH6.H7: Voluntariness will be negatively related to one attitude towards usingComputerized 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 Figure7).‘‘8 The involvenient of a Support Group will contribute to a successful ComputerizedAccounting System. A positive path coefficient of .321 (Innovativeness) indicatessupport for Hg.Hypothesis H9-H12 are indirectly tested as they are A.1V and contribute to the overallSupport Group (Communications Channel) LV path loading.H9. The involvement ofa Friend i ill contribute to a success/lu Computerized AccountingSstem. As the path coefficient is small (.0694), this suggests that the hypothesis isnot supported.Hj.The involvement of other Employees will contribute to a successful ComputerizedAccounting System. As the path coefficient of Personnel is positive (.3837), thissuggests 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 successfulComputerized 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 successfulComputerized Accounting System. As the path coefficient is large (.7121), thissuggests that the hypothesis is supported as the indirect effect (.7121*321=229) ismoderate.69H13. The involvement of a Support Group wi/i have a positive influence on SubjectiveNorm. 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. Asthe path coefficient is .381, this suggests that the hypothesis is supported.H15. The involvement of a Support Group will have a positive influence on Perceivedcharacteristics 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 ANALYSISUsing PLS statistical results, it was shown that a reduced alternative model with fourPerceived Chaiacteiistics of Innovating variables removed, provided a marginally (possiblynot statistically different) better fit indicators than the original Diffusion of InformationTechnology model. However, the better indicator scores did not appear to be different enoughto justify adopting the reduced model. An extended Diffusion of Information Technologymodel, including the LV Communications Network, did not provide any better fit indicatorsthan the original model either. Qualitative analysis of individual LV and MV indicators suggestthat improved predictive power may result when using the extended model. The path loadingsin both the inner and outer model change to varying degrees when LV CommunicationChannel is added to the original model (compare Figure 6 to Figure 7). The introduction ofthis LV into the model may have removed some of the “noise” from the model which mayhave previously skewed the loading values. As no statistical analysis has been performed onthe changes in loading values, no significance is claimed for the observed minor changes inloading values.70SECTION E: SUMMARY OF DATA ANALYSIS6.17 GENERALThree different techniques were used to analyse the data. The initial analysis included acomparison of the descriptive statistics between users and non-users of computerizedaccounting system adopters. Next, regression analysis was performed on the data to examinethe effect of various independent variables on the dependent variables. Finally, path analysiswas used to examine the theoretically derived Diffusion of Information Technology modeldeveloped by Moore and compare this model to other versions of this model to determinewhich model had the best fit to the data.6.18 SUMIVIARY OF DESCRIPTIVE STATISTICSWhile there were significant differences between Computerized Accounting Systemusers and non-users on several of the variables, there were fairly uniform Subjective Norms byall respondents. Overall, 71% of the sample were identified as Computerized AccountingSystem users, which indicates that non-users have a large impact on the overall results. Theaverage time elapsed since initial adoption is just under five years. Computerized AccountingSystems are used fairly often, with over 4 computerized accounting system functions beingused for 22 hours per week.6.19 SUMMARY OF HYPOTHESES TESTTNGHj. One attitude towards using a Computerized Accounting 5steni will influence oneinnovatii’eness i i/h respect to Computerized Accounting System usage.This hypothesis was supported by descriptive statistics, not supported by regressionanalysis, and partially supported by PLS. Regression analysis indicated that while Attitude wassignificant in the adoption process for all Innovativeness variables, all of the betas were alsonegative. PLS results indicated that the loadings were negative in value for the standard modeland positive for the extended model. The confusing Attitude results (Table 10(b)) may havebeen an artifact of the scales used to measure Attitude. Substituting the PerceivedCharacteristics ofInnovating variables for Altitude resulted in significant (positive) Perceived71Characteristics ofInnovating variables for all Innovativeness variables (Table 11(a)).112: Relative Advantage will have a contribution more than any other PerceivedCharacteristics of Innovating on one ‘c attitude towards adopting computerizedaccounting system.This hypothesis was generally supported using all methods. The regression resultsindicated that Relative Advantage was generally the most significant PerceivedCharacteristics of Innovating variable. PLS analysis was mixed with the original modelsupporting the hypothesis and the extended model rejecting the hypothesis.H3: Computer Avoidance will have a contribution less than any other PerceivedCharacteristics of Innovating on one’s attitude towards adopting ComputerizedAccounting Systems. This hypothesis is not explored in this study.H4: The Subjective Norm i’ill influence one innovativeness with respect toComputerized Accounting System usage.This hypothesis is only supported by PLS.H5: The Subjective Norm will influence one c attitude toward adopting the ComputerizedAccounting System.This hypothesis is supported only at the PLS stage.116: Voluntariness is negatively related to one’s innovativeness with respect toComputerized Accounting System usage.This hypothesis was not supported by any method.H7: Voluntariness will be negatively related to one attitude towards usingComputerized Accounting Systems.This hypothesis was generally not supported, except for the extended model usingPLS.118. The involvement of a Support Group ui/i contribute to a successful ComputerizedAccounting System.This hypothesis was supported by regression and PLS analysis.72H9. The involvement ofa Friend wi/i contribute to a successful Computerized AccountingSystem.This hypothesis was supported by regression and PLS analysis.Hj.The involvement of other Employees ii’iii contribute to a successfid ComputerizedAccounting System.This hypothesis was supported using PLS analysis.Hjj. The involvement of an external Accountant will contribute to a successfulComputerized Accounting System.This hypothesis was supported using PLS analysis.H12. The involvement of an external Consultant w’ii/ contribute to a successfulComputerized Accounting System.This hypothesis was supported using regression analysis and PLS.Hj.The involvement of a Support Group will have a positive influence on SubjectiveNorm.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 PerceivedCharacteristics ofInnovating variables.This hypothesis was not tested using descriptive statistics or PLS. It was supportedusing regression analysis.These results are summarized in Table 13(b).73CHAPTER 7: CONTRIBUTIONS, IMPLICATIONS ANI LIMITATIONS7.1 INTRODUCTIONThere has been much written on the topic of diffusion of innovations and the impact ofInformation 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 largecorporations is usually concerned with big business problems and big business solutions. Therelevance of these solutions to smaller firms is questionable, as small firms usually havedifferent problems than large firms, or experience large firm problems in ways that are uniqueto the small business domain. More research addressing real world problems from a smallbusinesss perspective is required which was the goal of this study.7.2 SUMMARY OF THE RESEARCH PROCESSThe motivation for this research came, in part from the lack of useful informationavailable to Public Accountants (and other information consultants) on how to prepare theirclients for the successful introduction (diffusion) of new Information Technologies. Theparticular information technology of interest was the Computerized Accounting System. Afierresearching the information system literature it was determined that the most effective tool forobtaining the type of information that information consultants required was from a model ondiffusion of information technology developed by Moore.This study has looked at the diffusion of information technology model first developedby Moore, in order to evaluate its robustness and generalizability to a small business,accounting domain. Quantitative and qualitative analysis were done using general descriptivedata, regression analysis, and path analysis. As part of this analysis, the role of the informationconsultant in the diffusion process was examined.74The major research questions answered are:1. What role do independent information consultants, such as accounting firms, play in theDffusion 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 InformationTechnology was a general model, three different statistical approaches were applied. Theseincluded analysing the differences between computerized accounting system users and non-users by general descriptive tests, performing regression analysis of the independent variableson the dependent variables, and finally, applying path analysis using PLS.7.3 THE RESEARCH QUESTIONS ANSWERED7.3.1 QUESTION TWOIs the Dffiis/on ofInformation Technology model a general model?Results showed overall support for the general model. The role of informationconsultants was not very significant when applied to the general model but did show someeffect 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 (H1received some support using all three methods). However, each hypothesis received eitherpartial support to definite support from at least one of the tests.The regression results indicated that a larger sample size may have obtained moresignificant results for some of the variables. What is clear from the results is that there aresome statistically significant differences between large firms and small firms. In the large firmstudy where the Diffusion of Information Technology model was first developed, basically allof the variables were significant and provided support for all of the hypotheses. In this studymost of the variables were not significant and at best, moderate support was provided to the75hypotheses. While a larger sample size would make more of the variables significant, itappears that, based on regression results, several variables would probably not becomesignificant.7.3.2 QUESTION ONEWhat role do independent information consultants such as accounting firms play in theDffusion qfInformation Technology process?Hypotheses H8 to H15 directly addressed this question. Statistical analysis indicatethat there is a relationship between the presence of outside support and adoption ofcomputerized accounting systems. Regression analysis indicate that the ComputerizedAccounting System Support composite variable is significant for all four Innovativenessvariables (number of Functions used, Frequency of use, Months since adopted, and Hours ofuse per week). The individual CAS Support variables had different levels of significance foreach of the Innovativeness variables.The CAS Support group Consultant was significant when regressed on number offunctions 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 PerceivedCharacteristics 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 PerceivedCharacteristics ofInnovating variables Image (p=, 004) and Ti/a/ability (p=.Ol 7).The CAS Support group Personnel was significant when regressed on theInnovativeness 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), SeniorManagement (p.O44) and Subordinate (p. 042).76The results indicate that CAS Support variables Information Consultants andPersonnel have both a direct and an indirect affect on the Computerized Accounting Systemadoption process. The direct affect can be seen from regression analysis and the indirect affectcomes from both regression and path analysis, where the Perceived (Jharacteristics ofInnovating 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 throughits influence on the construct Attitude. The Accountant was shown to have significantinfluence on the Perceived Characteristics qf Innovating variables Compatibility, Ease ofUse, Relative Advantage and Result Demonstrability. PLS analysis indicated the direction andmagnitude of this influence on Innovativeness through the intervening variable Attitude.7.4 CONTRIBUTIONSThis study has shown that the Diffusion ofInformation Technology model can be usedacross different information technology domains and for large or small organizations. Thestrength of this model is that once the attitudinal and societal characteristics of informationtechnology adoption are understood, information consultants will have the ability to predict ifan information technology will be adopted for a given organization. They will also be able torecommend to clients a methodology to maximize the success of the introduction andadoption of an information technology.The reduced Dffrsion of Information Technology questionnaire (39 Questions) hassufficient reliability to be used in similar research. This questionnaire captures PerceivedCharacteristics of innovating variables adequately, but does not capture the constructAttitude. Although the development of suitable Attitude scales would normally be arecommendation, the use of path analysis programs like PLS to indirectly synthesize thisvariable suggests that further scale development for Altitude may not be warranted.The inclusion of Communications Channel (ie. Support Group) as an extension to theDffusion of Information Technology model is an attempt to improve the robustness of this77model. Statistical analysis show that this extension does reveal some interactions in theDiffusion of Information Technology model not previously evident. However, theexplanatory power of the model has only been modestly improved.This study will add to the small but growing body of research literature specificallyoriented towards smaller organizations. While the Dffusion ofInformation Technology modelis generalizable and robust in both large and small business domains, it seems that a subset (inPLS 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 ofquantitative analysis using PLS.7.5 LIMITATIONS OF THE STUDYThis study, while providing some interesting results about the information technologyadoption process, has some underlying limitations which must be kept in mind.Sample size is always of some concern, as researchers are almost never satisfied withtheir sample. While the sample size of 75 was adequate for purposes of PLS analysis, therewere several regression results that would probably have become significant with a largersample. It was difficult to directly compare results with those obtained by Moore as he had amuch larger sample size (600) and nearly everything was significant for his study. It becomesdifficult 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 torespondents, resulted in a loss of control of sample selection. The sample should be called aconvenience sample because there was very little randomness in the selection process. Withthe problems encountered in collecting completed questionnaires it is most likely thatparticipating accounting firms selected potential respondents on the basis of individuals whowould be most likely to fill out the questionnaire,The demographic’s data for the sample indicate that approximately one third of therespondents were non-users. There were significant differences between users and non-users78for several categories. The relatively large proportion of non-users may have swamped ormasked some of the results on computerized accounting system adoption. Also, a largeportion (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 atimproving reliablity were made (asking the same question more than once), the generalproblems with questionnaire data still remain.While the Dfji’sion of Information Technology model is purported to be ageneralizable model, and statements are made about the generalizability of results from thisstudy, it should be kept in mind that the results are applicable to this study only. While it ishuman nature to make inferences and extrapolate results it should be noted that suchinferences and extrapolations are made at the risk of over-interpreting the results from anindividual study.7.6 CONCLUSIONThe role that information consultants currently play in the adoption process for smalland medium size firms is understood a little better. While there is a strong association betweenthe presence of information consultants and the successful adoption of a computerizedaccounting systems, many small businesses do not rely on this support group to help themwith new information technology. It appears that computerized accounting system users whohave used a computerized accounting system for a long period of time, and/or use manycomputerized accounting system functions, are more likely to rely on a support group. It isnot clear if the presence of the support group leads to long and versatile computerizedaccounting system use, or if the experience gained due to the passage of time and/or heavyuse has convinced users to seek outside help. If the latter case is true, then PublicAccountants have a lot of work to do to get the message out that experience doesn’t have tocome the hard way.79The extension to the Di/jitsion / Inftrniation Technology model, by includingSupport Group, has provided a modest improvement to the explanatory and predictive powerof the model. Clearly, there is much that is still not known about the factors that can lead tothe successful adoption of information technology. However, the extended Diffusion ofInformation Technology model does provide some insight into this process.TABLES80ocH C H H tn (1)C) mIt•ij-IjH HC HCr11— r11 H Hrn > H > > Z H C)C)—CcH—tTlcC—z0 CD 0 0 CD 0—H<tncmCo-rn1rn>cHr-(IDt>c,)>ci>C)r’iHH<tn<Cl)— m‘Hci)z(I)(/) HZH >>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)LILiLi004..-.—(ID— H C H O H (ID H2 H C C4-Li.LI.L.)004-C)4-11)2 CZ000000LC(IDII00LM—C)C)LtJC)C)CC)LiCL.)(J)Li.—L.J\C)L—C)C((ID-——--—-——-—-Cô’>-.(IDC)0OLC0CJIIL0C4-—Ck)CJLiC)—C4-CC)(IDJLi.L.JLM—JtC)(IDt.——-—CCtJC)4-C 2 (ID(ID (ID (IDLI,4-•00 C)0000LI,CLI.4-t)4-cc LI,00 C)(ID(ID (ID00TABLE 5(a)DEMOGRAPHIC BACKGROUND OF SURVEY RESPONDENTSRelative Ad justedNumber Frequency FrequencyDEPARTMENT OF EMPLOYMENTAdministration 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 LEVELExecutive 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%EDUCATIONSome 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%8384TABLE 5(b)DEMOGRAPHIC BACKGROUND OF SURVEY RESPONDENTSRelative AdjustedNumber Frequency FrequencyAGELess 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%SEXMale 26 34.7% 34.7%Female 49 65.3% 65. 3%Missing 0 0.0YTotal 75 100.0% 100.0%OTHER Minimum MaximumAverage Firm Size (Sales) $500k- <$250k >$ 10,000k$L000kAverage Firm Size (Full Time Employees) 26 92Avg. Accounting Staff (Full Time Employees) 3 1285TABLE 6(a)SURVEY VARIABLES - DESCRTPTWE STATISTICSMAXIMUM MINIMUM# SCALE MEAN STANDARD REPORTED REPORTEDITEMS SCORE DEVIATION SCORE SCOREPERCEIVED CHARACTERISTICS(Scale Range: 1 to 7)Compatibility 4 5.427 1.577 7.000 1.000EaseofUse 6 5.118 1.003 7.000 2.167Image 4 3.977 1.366 7.000 1.000Relative Advantage 8 5.503 1.560 7.000 1.000Result Demonstrability 3 5.187 1.179 7.000 2.667Trialability 5 4.128 1.426 7.000 1.000Visibility 5 4.856 1.484 7.000 1.000Voluntariness 4 3.130 1.480 5.750 1.000ATTITUDE 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 -18.000Peers 1 1.467 5.512 — 21.000 -8.000Supervisors 1 .547 5.194 21.000 -8.000Senior Management 1 -2.320 5.403 21.000 -15.000Subordnatcs 1 -2.773 5.562 21.000 -21.00086TABLE 6(b)SURVEY VARIABLES- DESCRIPTIVE STATISTICSMAXIMUM MINIMUM# SCALE MEAN STANDARD REPORTED REPORTEDITEMS SCORE DEVIATION SCORE SCOREINNOVATIVENFSS MEASURESFrcquency 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 1.000Friend 3 2.157 .993 5.000 1.000Accountant 3 2.308 1.025 4.667 1.000Consultant S 2.277 1.031 4.333 1.000-—C)C-CDC)CD—raCI)—)— C C — C z C rfj C.’) CI) Cd) C CI) CI) — Cl)CI)Cl)Cd) C Cl) C.’):-——--—--——-—CI)—J—t.)a—C-——L1t’J—t)L.JC——--——--—--—z—aCL.Jcj,Cl)k)çj,—CD’J—CC\0088TABLE 7(b)USERS VERSUS NON-USERSIL VARIABLE MEANS AND TESTS FOR DIFFERENCES (M-W TESTS)NON- U-TESTUSERS USERS Z-SCORE SIGNIFPERCEIVED CHARACTERISTICS (M-W)Compatibility 6.10 3.80 -4.86 .0000Ease of Use 5.25 4.81 -1.47 .1420Image 4.35 3.08 -3,45 .0006Relative Advantage 6.18 3.87 -4.81 .0000Result Demonstrability 5.67 4.02 -5.59 .0000Trialability 4.33 3.65 -1.9() .0575Visibility 5.31 3.76 -3.25 .0011Voluntariness 2.72 4.13 -3.74 .0002ATTITUDE 1.76 3.25 -3.72 .0002SUBJECTIVE NORMSFriends-.68 -2.00 -1.02 .3063Peers.98 2.64 -.72 .4719Supervisors.06 1.73 -.11 .9146SeniorManagement-1.38 -4.59 -2.35_ .0189Subordinates-1.74 -5.27 -2.95 .0032INNOVATWENESS MEASURESMonths elapsed since adoption 55.96 N/A N/A N/AHours of use per week 21.58 N/A N/A N/AFrequency of use - general once/day N/A N/A N/AFrequency of use - detail 27.79 N/A N/A N/ANumber of functions used 4.50 N/A N/A N/ACAS SUPPORTPersonnel from Firm 2.63 N/A N/A N/AFriend 2.16 N/A N/A. N/AAccounting Firm 2.31 N/A N/A N/AConsultant 2.28 N/A N/A N/A89TABLE 8REGRESSION RESULTSPERCEIVED CHARACTERISTICS AND VOLUNTARINESS ON ATTITUDEI. SUMMARY OF STEPPED FORCED ENTRY OF VARIABLESII. STA ‘ISTICS FOR VARIABLES IN TI-IF FINAL EOI ATIONSTEP VARIABLE IN BETA IN R2 F (EQN) SIG F(EQN)1 Relative Advantage -.852 .726 193 .0002 Visibility -.240 .755 111 .0003 Voluntariness .113 .763 76 .0004 EaseofUse-.095 .769 58 .0005 Image -.073 .773 47 .0006 Trialability .057 .775 39 .0007 Compatiblity .076 .776 33 .0008 Result Demonstrability -.019 .776 29 .000IVARIABLE BETA STD ERR F (B) STG FBETARelative Advantage -.632 .123 16.061 .000Visibility-.2 10 .083 4.260 .043Voluntariness .113 .067 1.924 .170Ease of Usc -.123 .090 2.720 .104Image -.079 .061 1.360 .248Trialability .050 .069 .377 .541Compatiblity .083. 123 .269 .605Result Demonstrability -.019 .078 .062 .805R2=.776Variance Explained Adjusted R2 .74990TABLE 9REGRESSION RESULTSPCI’S, VOLUNTARINESS, SN AND SUPPORT ON ATTITUDE(IUatiOII PCI Only PCI and SN PCI,SN & SUPPORTBeta Weights Beta Sig. F Beta Sig. F Beta 1 Sig. FRelative Advantage -.63 2 .000 -.653 .000 -.681 .000Visibility-210 .043 -.199 .054 -.200 .053Voluntariness .113 .170 .092 .277 .083 .326Ease of Use— -.123 .104 -.109 .155 — -.101 .188Image -.079 .248 -.085 .216 -.087 .206Trialability .050 .541 .057 .492 .060 .464Compaliblity .083 .605 -.083 .603 .083 .603Result Demonstrability -.019 .805 -.020 .793 -.048 .552Subjective Norm-.069 .282 -.077 .235CAS Support.074 .314Variance Explained R2 =776 R2 =780 R2 =.783Adj R2 =749 Adj R2 =750 Adj R2 =75091TABLE 10 (a)REGRESSION RESULTSATTITUDE, SN, AND VOLUNTARINESS ON INNOVATIVENESSVARIABLESDEPENDENT INDEPENDENT Beta Sig. F Adj. F F SigR2NUMBER OF VoLuntariness -.023 .861FUNCTIONS Attitude -.471 .000USED Subjective Norm -.074 .503 .193 6.903 .0004FREQUENCY OF Voluntariness -.092 .472USE Attitude -.454 .00()Subjective Norm -.035 .746 .223 8.069 .0001MONTHS SINCE Voluntariness .-. 107 .405ADOPTED Attitude -.433 .001Subjective Norm -.067 .538 .207 7.445 .0002HOURS OF USE Voluntariness -. 191 .138PER WEEK Attitude -.382 .003Subjective Norm -.073 .500 .218 7.881 .000192TABLE 10 (h)REGRESSION RESULTSATTITUDE, SN, VOLUNTARINESS & SUPPORT ON INNOVATIVENESSVARIABLESDEPENDENT INDEPENDENT Beta Sig Adj. F F SigR2NUMBER OF Voluntariness -.052 .605FUNCTIONS Attitude -.22 1 .035USED Subjective Norm -.087 .302Support .617 . .000 .524 21.338 .0000FREQUENCV Voluntariness-. 122 .185OF USE Attitude -.189 .050Subjective Norm -.050 .524Support .654 .000 .597 28.390 .0000rvIONTHS Voluntariness-. 124 .304SINCE Attitude -.289 .022ADOPTED Subjective Norm -.075 .462Support .354 .001 .309 9.260 .0000HOURS OF USE Voluntariness -.2 14 .054PER WEEK Attitude -.182 .110Subjective Norm -.084 .366Support .492 .000 .425 14.669 .000093TABLE 11 (a)REGRESSION RESULTSPCI AND SUBJECTIVE NORMS ON INNOVATIVENESSEQUATION 1: DEPENDENT VARIABLE - NUMBER OF FUNCTIONS USEDEntry Independent Variable Final Beta Sig. F Ad. R2 FStep1 Result Demonstrability .447 .0002 SN Peers-.297 .0013 SN Subordinates .268 .0044 Relative Advantage .217 .049 .504 20Weakly Significant:Ease of Use -.186 .061EQUATION 2: DEPENDENT VARIABLE - FREQUENCY OF USEEntry Independent Variable Final Beta Sig. F Ad. R2 FStepI Relative Advantage 403 .0002 Result Demonstrability .365 .001 461 33Weakly Significant:SN Subordinate .158 072Ease of Use -.172 .091EQUATION 3: DEPENDENT VARIABLE - MONTHS SINCE CAS FIRST ADOPTEDEntry Independent Variable Final Bela Sig. F Adj. R2 FStepI Result Demonstrability .597 .000 .347 40Weakly Significant:Relative Advantage .22-I .058Voluntariness-. 179 .065Coinpatibility .210 .078EQUAT ON 4: DEPENDENT VARIABLE - HOURS OF USE PER WEEKEntry Independent Variable Final Beta Sig. F Adj. R2 FStepI Result Demonstrability 445 .0252 Visibility .268 .013 .362 2294TABLE 11(h)REGRESSION RESULTSPCI, SUBJECTIVE NORMS, & SUPPORT ON INNOVATIVENESSEQUATION 1: DEPENDENT VARIABLE- NUMBER OF FUNCTIONS USEDEntry Independent Variable Final Beta Sig. F Mj. R2 FStepI Result Demonstrability .349 .0002 SN Peers -.281 .00()3 SUPP Personnel .290 .0014 SUPP Consultant .282 .0015 SN Subordinate .199 .009 .660 30EQUATION 2: DEPENDENT VARIABLE- FREQUENCY OF USEEntry Independent Variable Final Bela Sig. F Mj. R2 FStep1 SIJPP Personnel .400 .0002 SUPP Consultant .225 .0013 Relative Advantage -.464 .0064 Result Demonstrability .201 .0145 SN Peers-. 135 .030 .727 41Weakly Significant:SN Subordinate. 127 .062EQUATION 3: DEPENDENT VARIABLE- MONTHS SINCE CAS FIRST ADOPTEDEntry Independent Variable Final Beta Sig. F Adj. R2 FStep1 Result Demonstrability .497 .0002 Consultant .221 .035 .378 23Weakly Significant:Voluntariness-. 169 .076EQUATION 4: DEPENDENT VARIABLE- HOURS OF USE PER WEEKEntry Independent Variable Final Beta Sig. F Adj. R2 FStep1 SUPP Consultant .436 .0002 Result Demonstrability .3 12 .0013 Voluntariness—.225 .010 .510 2795TABLE 12(a)REGRESSION RESULTSCAS SUPPORT ON OTHER DEPENDENT VARIABLESEQUATION 1: DEPENDENT VARIABLE- SUBJECTIVE NORM (COMPOSITE)Dep. Variable Independent Variable Final Beta Sig. F Adj. R2 FSNc No Significant Variables---EQUATION 2: DEPENDENT VARIABLE- SUBJECTiVE NORM (COMPONENTS) +VOLUNTARINESSDep. Variable Independent Variable Fiiial Beta Sig. F Adj. R2 FFriend SUPP Personnel .200 .085 .027 3Peer No Significant Variables---Supervisor No Significant Variables---Senior Mgiut SUPP Personnel .234 .044 .042 4Subordinate SUPP Personnel .236 .042 .043 4EQUATION 3: DEPENDENT VARIABLE- PERCEIVED CHARACTERISTICS OF INNOVATINGDep. Variable Independent Variable Final Beta Sig. F Adj. R2 — FCompatibilty SLJPP Accountant .302 .012“ SUPP Personnel .272 .023 .225 12*********Ease of Use SUPP Accountant .257 .026 .053 5Image SUPP Friend .325 .004 .093 9Rd. Advant. SUPP Accountant .487 .000 227 23Res. Demon. SUPP Accountant .532 .00() .273 29Trialability SIJPP Friend .275 .017 .063 6Visibility SUPP Consultant .377 .001. IS I 1296TABLE 12(b)REGRESSION RESULTSCAS SUPPORT ON OTHER DEPENDENT VARIABLESEQUATION 3: DEPENDENT VARIABLE- INNOVATIVE (USE) VARIABLESDep. Variable Independent Variable Final Beta Sig. F Ad. R2 FFrequency SUPP Consultant .424 .000SUPP Personnel .481 .000 .606 58*********Functions SUPP Consultant .445 .00011 SUPP Personnel .383 .000 .503 38* ** * *** * *Hours SUPP Consultant .522 .000SUPP Personnel .181 .092 .384 24*********Months SUPP Consultant .327 .008SUPP Personnel .234 .053 .218 11EQUATION 5: DEPENDENT VARIABLE- VOLUNTARINESSDep. Variable Independent Variable Final Beta Sig. F Mi. R2 FVoluntariness SUPP Personnel-. 267 .020 .059 697TABLE 13 (a)—SUMMARY RESULTS OF HYPOTHESIS TESTINGHYPOTHESES ADOPTERS VS. REGRESSION PLSNON-ADOPTERS ANALYSIS ANALYSISHi: ATTITUDE -> INNOVATIVENESS SUPPORTED NOT SUPPORTEDSUPPORTEDH2: RELATIVE ADV> OTHER PCI SUPPORTED SUPPORTED SUPPORTEDH3: AVOIDANCE < OTHER PCI N/A N/A N/AH4: SN -> INNOVATIVENESS NOT NOT NOTSUPPORTED SUPPORTED SUPPORTEDH5: SN -> ATTITUDE N/A NOT NOTSUPPORTED SUPPORTEDH6: VOLUNTARY-> INNOVATIVENESS SUPPORTED NOT NOTSUPPORTED SUPPORTEDH7: VOLUNTARY -> ATTITUDE N/A NOT SUPPORTEDSUPPORTEDH8: SUPPORT -> INNOVATIVENESS NOT SUPPORTED SUPPORTEDSUPPORTEDH9: FRIEND -> INNOVATIVENESS NOT NOT NOTSUPPORTED SUPPORTED SUPPORTEDH1O: EMPLOYEE -> INNOVATIVENESS NOT SUPPORTED SUPPORTEDSUPPORTEDHi 1: ACCOUNTANT -> INNOVATIVE NOT NOT NOTSUPPORTED SUPPORTED SUPPORTEDH12: CONSULTANT-> INNOVATIVE NOT SUPPORTED SUPPORTEDSUPPORTEDHi 3: SUPPORT -> SN N/A NOT SUPPORTEDSUPPORTEDH14: SUPPORT -> ATTITUDE N/A NOT SUPPORTEDSUPPORTEDHI 5: SUPPORT -> PCVS N/A SUPPORTED SUPPORTED98TABLE 13(b)SUMMARY RESULTS OF HYPOTHESIS TESTING(MOORE)HYPOTHESES ADOPTERS VS. REGRESSION PLSNON-ADOPTERS ANALYSIS ANALYSISHI: ATTITUDE -> INNOVATIVENESS SUPPORTED SUPPORTED SUPPORTEDH2: RELATIVE ADV> OTHER PCI NOT SUPPORTED NOTSUPPORTED SUPPORTEDH3: AVOIDANCE < OTHER PCI SUPPORTED NOT SUPPORTEDSUPPORTEDH4: SN -> INNOVATIVENESS SUPPORTED NOT SUPPORTEDSUPPORTEDH5: SN -> ATTITUDE N/A SUPPORTED SUPPORTEDH6: VOLUNTARY -> INNOVATIVE SUPPORTED SUPPORTED SUPPORTEDH7: VOLUNTARY -> ATTITUDE N/A SUPPORTED SUPPORTED99TABLE 14GENERAL PLS STATISTICS FOR TESTED MODELSMODEL MULTIPLE R AVG. COMMUN. AVG. REDUND.. (R2) (H2) (F2)All data points:Full Model .1837 .5275 .1097Full- Voluntariness .1281 .4636 .0584Full- Voluntariness - SN 1357 .2835 .0424Full-4PCI’s .1992 .5558 .1252Full + Communications Channel. 1960 .5389 .1238CAS User data points: .0957 .4 173 .0470FullFull- Voluntariness. 1012 .3593 .0379Full - Voluntariness - SN .1501 .3745 .0470Full - 4 PCIs. 1209 .4575 .0624Full + Communications Channel. 1242 .4062 .0559—FIGURE1Diffusionof InformatIonTechnologyModel-Moore,1989TOWARDSSUBJECTIVENORMVOLUNTARINESSATTITUDEINNOVATIVENESSADOPTINGFIGURE2Diffusionoi InnovationsModel-Rogers,1983PerceivedAttributesCommunicationChannelsof Innovations-MassMedia-RelativeAdvantage-Interpersonal-Compatibility-Trialability-Complexity-ObservabilityTypeof Innovation-DecisionR4TEOFADOPTIONExtentof ChangeAgents’-OptionalrsPromotionEfforts-CollectiveOFINNOVATIONS-AuthorityI____________________________NatureofSocial System-Norms-DegreeofInterconnectednessFIGURE3COMMUNICATIONSNETWORKInnovationDecisionModel-Fishbein&Ajzen,1975(AdaptedbyMoore,1989)OBJECTIVECHARACTERISTICSOFINNOVATIONTOWARDSADOPTINGINTENTIONINNOVATIONDECISIO))SUBJECTIVE-1PERSONALCHARACTERISTICSOFADOPTERSNORMBEHAVIOURALN. 1/BEHAVIOUR(ADOPTION/REJECTIO1ATTITUDEOBJECTIVECHARACTERISTICSOFPRECURSORCFIGURE4StagesoftheInnovationDecisionProcessModelRogers,1983(AdaptedbyMoore,1989)KNOWLEDGEPERSUASION___________DECISIONADOPTIONREJECTIONffiMATIONJlFIGURE5NON-CASUSERSRESULTDEMONSTRABILITYRsAcTANLGE0100010101030104010501060107016802030204020602070208RESPONDENT0209021002110216021702190233P124P1257 6 4 3 0-A—IIIIIIIIIIIIIII•U15U23AU331FIGURE6DIFFUSTIONOFINFORMATIONTECHNOLOGYMODELPLSLOADINGSONORIGINALMODELC=LATENTVARIABLE(LV)II=MANIFESTVARIABLE(MV)INNERMATRIX=OUTERMATRIX-REFLECTIVEOUTERMATRIX-FORMATrVESubordinates8Z5.3.91417092.90946q9“73FIGURE7DIFFUSIONOFINFORMATIONTECHNOLOGYMODELPLSLOADINGSONEXTENDEDMODEL.2935=LATENTVARIABLEII=MANIFESTVARIABLE.9136CC109BIBLIOGRAPHYAhituv, Niv, “Assessing The Value Of Information Problems and Approaches”, Proceedingsof 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-UserComputing”, 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: AnIntegrative Framework”, Journal of MIS, Vol. 4, No. 3, Winter 1987-88, pp. 28-49.Alexander, Mary B., The Adoption And Implementation Of Computer Technology InOrganizations: The Example Of Database Machines, Unpublished PhD thesis, IndianaUniversity, 1989.Allen, Brandt, “An Unmanaged Computer System Can Stop You Dead”, Harvard BusinessReview, November-December 1982, pp. 77-8 7.Arter, Terrence J., “Contract The Independent Computer Consultant”, Journal of SmallSystems 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”, Paperpresented 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 TestingTheories In Consumer Research”, American Marketing Association, 1982, pp. 195-20 1.Bailey, James E. and Pearson, Sammy W., “Development of a Tool For Measuring andAnalyzing Computer User Satisfaction”, Management Science, Vol. 29, No. 5, May1983, pp. 530-545.Baker, William H., “Status Of Information Management In Small Business”, Journal ofSystems Management. April, 1987, pp. 10-15.Barki, H. and Huff, Sid L., “Measuring Decision Support System Success”, ASAC 1984Conference.Barclay, D.W., Higgins, C.A., & Duxbury, L.E., An Introduction to the Partial Least SquaresApproach to Causal Modeling, University of Western Ontario, WP-91-08, 1991.110Baronas, Ann-Marie K. and Louis, Meryl Reis, “Restoring a Sense of Control DuringImplementation; 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 ImpactOf User Involvement On System Usage And Information Satisfaction”, Communicationsof 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”, IhcBottom Line, November 1989, pp. 22-24.Benjamin, Robert I., “Information Technology in the 1990s: A Long Range PlanningScenario”, 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 IKnowledge 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 FirmStrategic Planning”, Journal of Small Business Management, July 1985, pp.23-30.Bradbard, David A.; Norris, Dwight R.; and Kahai, Paramjit H., “Computer Security in SmallBusiness: 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 ExtendingInnovation Diffusion Theory In The Context Of End-User Computing, Unpublished PhDthesis, 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. 56-58.111Bryman, 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 researcherduring 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 EconomicReorganization”, Proceedings of Tenth International Conference on InformationSystems, December 1989, pp. 341-35 1.Cooper, Randolph B. and Zmud, Robert W., “Information Technology ImplementationResearch: 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 DangerousTo Your Organization”, Working Paper No. MISRC-WP-82-04, ManagementInformation 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”. MiQuarterly, March 1988.Dillman, Don A., “Mail and Telephhone Surveys: The Total Design Method”, Wiley andSons, New York, 1978.Dimnik, Tony “An Introduction to LISREL”, The University of Western Ontario, WorkingPaper Series No. 86-21, 1986.112Dragich, Martha J., “Information Malpractice: Some Thoughts On The Potential Liability OfInformation Professionals’, Information Technology and Libraries, September 1989, pp.265-272.Ein-Dor, Phillip and Segev, Eli, “Organizational Context and MIS Structure: Some EmpiricalEvidence”, 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 EquationsApproach Using Partial Least Squares”, Proceedings, American Marketing AssociationEducators’ Conference, 1982, pp. 1-5.Frame!, John E., “Managing Information Costs And Technologies As Assets”, Journal ofSystems Management, February 1990, pp. 12-19.Gable, Guy G. “Consultant Engagement for First Time Computerization: A Pro-Active ClientRole 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 ofSeventh 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 - Survey9, Dobson Marketing Services, 1989, Abbotsford, B.C. Portions of this survey werereproduced in Sial1 usinessMagazii.ç, January 1990.Hill, Thomas; Smith, Nancy D. and Mann, Millard F., “Role of Efficacy Expectations inPredicting the Decision to Use Advanced Technologies: The Case of Computers”,Journal of Applied Psychology, Vol. 72, No. 2, pp. 307-3 13.113Holmes, Scott and Nicholls, Des. “Modelling the Accounting Information Requirements ofSmall 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 onOrganizational Design, Intelligence, and Decision Making”, Academy of ManagementReview, Vol. 15, No. 1, January 1990.Ives, Blake, Olson, Margrethe H. and Baroudi, Jack J., “The Measurement of UserInformation Satisfaction”, Communications of the ACM, Vol. 26, No. 10, October1983, 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 ThePerformance Effects Of Information Technology Investment”, Proceedings of the TenthInternational 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, December1989, pp. 60-6 I.Lees, John D., ‘Successful Development Of Small Business Information Systems”, Journal ofSystems Management, September 1987, pp. 32-39.Lees, John D. and Lees, Donna D., “Realities Of Small Business Information SystemImplementation”, Journal of Systems Management, January 1987, pp. 6-13.Lefebvre, Elizabeth and Lefebvre, Louis A., Planning For Information TechnologyAcquisition: 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 HostileEnvironment”, Journal Of Systems Management, April 1990, pp. 11-14.Lind, Mary R. arid Zmud, Robert W., “The Influence of a Convergence in UnderstandingBetween 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 WithPartial Least-Squares Estimation”, Munich, 1984.Lucas, Henry C. Jr., “Empirical Evidence For a Descriptive Model of Implementation”, MISQuarterly, June 1978, pp. 27-41.114Luscombe, 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 InInformation 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, CarnegieMellon University, July 1990.Miller, Howard W., “Developing Information Technology Strategies”, Journal of SystemsManagement, September 1988, pp. 28-3 5.Millman, Zeeva and Hartwick, Jon, “The Impact Of Automated Office Systems On MiddleManagers And Their Work”, MIS Quartjy, December 1987, pp. 479-491.Moore, Gary C., The Examination of the Implementation of Information Teno1ogv For EndUsers: A Diffusion of Innovations Perspective, Unpublished PhD thesis, UBC 1989.Moore, Gary C. and Benbasat, Izak, “Development of an Instrument To Measure ThePerceived Characteristics of Adopting an Information Technology Innovation”, ThcInstitute 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 AndCommunication Channels On The Diffusion Of Innovation In A Data Base DevelopmentEnvironment”, 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, 1978.115Overbey, John T., Carland, Jo Ann, and Carland, James W., “Impact Of Microcomputers OnAccounting Systems”, Journal of Systems Management”, June 1987, pp. 20-27.Pavitt, K.; Robson, M.; and Townsend, J., “Technological Accumulation, Diversification AndOrganisation 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, foundin Bracker & Pearson (see above reference).Pendegraft, Norman; Morris, Linda; and Savage, Kathryn, “Small Business ComputerSecurity”, 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, December1989, 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 SmallBusiness’, MIS OuarteiJ, March 1985, pp. ???Reich, Blaize H. and Benbasat, Izak, “The Use Of Information Technology For CompetitiveAdvantage in Canada: An Examination Of Information Systems Linking Companies ToTheir Customers”, ASAC 1988 Conference, 1988, pp. 171-18 1.Richmond Business, “Small Business is Big”, Richmond Review (quoting from a reportprepared 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, 1986.116Sanders, G.L. and Courtney, JR. “A Field Study of Organizational Factors Influencing DSSSuccess”, 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 andCompany, 1978).Stulberg, Gregg, Focus On Office Technology, Financial Post, March 22, 1991, pp. 22Thompson, Ron L., “An Empirical Investigation Of Factors Affecting The Use Of PersonalComputers 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 SystemsMc.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 EmpiricalLiterature”, Management Science, Vol. 25, No. 10, October 1979, pp. 966-979.1/)118APPENDIX I-ARelative Advantage: the degree to which an innovation is perceived as being betterthan its precursorCompatibility: the degree to which an innovation is perceived as beingconsistent with the existing values, needs, and past experiences ofpotential adoptersEase of Use: the degree to which an innovation is perceived as being difficult(Complexity) to useTrialability: the degree to which an innovation may be experimented withbefore adoptionObservability: the degree to which the results of an innovation are observable toothersImage: the degree to which use of an innovation is perceived to enhanceones image or status in ones social systemVoluntariness: the degree to which use of the innovation is perceived as beingvoluntary, or of free willVisibility: the degree to which the innovation is apparent to the sense ofsightr—APPENDIX I-BINNOVATIVENESS: (Moore, 1989, pp. 133)Adoptive degree to which an individual is relatively early in adopting aninnovation.Implementation degree to which an individual puts an innovation to use within agiven use domain.Use degree to which an individual who has adopted the innovation usesit to solve novel problems, or in a new use domain.119(From Stone, 1978)VALIDITY ITEM DEFINITIONContent Validity Measurement items are representative sample ofdomain of items associated with variable beingmeasured.Construct Validity Appropriate operational definition cf anabstract variable (construct)Criterjon-related —— Use of scores obtained from one measureValidity (predictor) to infer individual’s probablestanding on another variable (criterion)Face Validity Item appears to measure what it claims tomeasure.Incremental Validity Item provides an improvement in predictivepower in conjunction with other measure(s)over the use of the other measure(s) alone.Convergent/Discriminant Scores on the measure correlate highly withValidity scores on other independent measures of thevariable and correlate low on measures ofother variables.APPENDIX I-CNote: There are several other Validity items, only the most commonlyused items are discussed above.APPENDIX H-A120121Appendix II- AlQuestionnaire122WELCOME!You are about to participate in a study of opinions about the usage of microcomputers in the accountingfunction. In some sections you will be asked questions about and see reference to the term CAS, whichstands for COMPUTERIZED ACCOUNTING SYSTEM. A CAS is defined for the purposes of thisstudy as a set of computerized tools for an individual, and usually consists of a personal ormicrocomputer with one or more software packages, such as an accounting program, and/or othersoftware such as spreadsheet, database, word processing, etc. in support of the accounting function. Thekey aspect of a CAS is that it is computer technology that you would use directly, as opposed to havingsomeone 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 pleasedon’t spend an excessive amount of time on each question.123INSTRUCTIONSIn the attached questionnaire, we ask questions which make use of rating scales with seven places; youare asked to place an ‘X’ in the place that best describes your opinion. For example, if you were askedto 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 unlikelyextremely quite slightly neither slightly quite extremelyIf you think that it is extremely likely that driving a car in winter is easy, you would make your markas follows:Driving a car in winter is easy.likely L x I I I I I unlikelyextremely quite slightly neither slightly quite extremelyIf you think that it is neither likely nor unlikely that driving a car in winter is easy, you would makeyour mark as follows:Driving a car in winter is easy.likely I I I X f unlikelyextremely quite slightly neither slightly quite extremelyIn addition to the “likely-unlikely” pairs, other pairs such as “disagree-agree” will also be used. Theyshould 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 unlikelyextremely quite slightly neither slightly quite extremelyTHIS NOT ThIS2. 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 lettercorresponding to a particular answer for a question. Please be careful to see that your circle goes aroundonly the letter or number which corresponds to your desired response.124TO 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, speeddialing, call waiting, etc.). (Place an ‘X’ beside the appropriate answer):___NOYESIf yes, which functions do you use? (Place an ‘X’ beside the appropriate functions):CALL TRANSFER(CONSULTATIONS) HOLDTHREE-WAY CONFERENCECALL FORWARDINGCALL PARKING- CALL PICKUPCALL WAITINGRING AGAIN/AUTOMATIC CALL BACKSPEED CALLINGLAST NUMBER DIALLEDSAVE NUMBER AND REPEAT125A-2 How often do you use the products listed below? (Place an ‘X’ under the appropriate columnfor each applicable area):About 1- MoreLess than 3 Times About 2- About than OnceNot at Once per per Once per 4 Times Once per per DayAll Month Month Week per Week Daya. Automated TellerMachineb. Programmable Calculatorsc. Home Computersd. Business_Computersc. Video Gainesf. Programmable MicrowaveOvensA-3 How often do you carry out the computer-related activities listed below; on paper, via electronicmail, on floppy disk, etc...? (Place an ‘X’ under the appropriate column for each applicablearea):About 1-Less than 3 Times About 2- About MoreNot at Once per per Once per 4 Times Once per than Onceall Month Month Week per Week Day per DayReceive computer output(reports/documents)Submit documents, etc. toothers for word processingSubmit data to others forcomputer analysis126A-4 What is your current keyboarding (typing) ability?a. Mark with an ‘X’:___HUNT & PECKTOUCH TYPEb. Place an ‘X’ in the place that best reflects your speedI I I I I I I IOwpm 1-15 16-30 31-45 46-60 61-75 >75wpm wpm wpm wpm wpm wpmA-5 How many educational courses (at any level) have you had about computers, but which did notinclude your personal hands-on use? (eg: “theory” courses)________COURSESA—6 How many educational courses have you had which required your personal hands-on use ofcomputers? (eg: “applied” courses)COURSESA-7 My firm receives non-computer support for the following areas (place an ‘X’ under theappropriate column for each applicable area):none constant1 2 3 some 5 6 7AccountingAuditBUSIneSS AdviceFinancial PlanningGov’t ComplianceMarketingTaxOther(please specify)127A-8 My firm receives non-computer support from the following sources external to the firm (placean ‘X’ under the appropriate column for each applicable source):none constant1 2 3 sOme 5 6 7Personal friend (nonemployee)Public accounting firmNon-Accountant computerconsultantNoneOther(please specify)A-9 I am satisfied with the current level of support for non-computer areas I receive from thefollowing sources external to the firm (place an ‘X’ under the appropriate column for eachapplicable source):satisfied unsatisfiedextremely quite slightly neither slightly quite f extremelyPersonal friend (nonemployee)Public accounting firmNon-Accountantcomputer consultantNoneOther(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 Iextremely quite slightly neither slightly quite extremely128A-il How knowledgeable do you feel you are of the uses of the CAS?unlimited I I I I Iextremely quite slightly neither slightly quite extremelyA-12 Have you ever used a CAS? (Place an ‘X’ beside the appropriate column):___CURRENTLY USE A CAS Please go on to the next pageHAVE NEVER USED A CAS Please go on to the next pageUSED TO USE A CAS BUT NO LONGER DO SOPlease 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 youstopped using it.STARTED______ ______MONTh YEARSTOPPEDMONTH YEARA-14 Please indicate which of the CAS functions below you used by indicating the number of monthsyou used them.Accounting Graphics Information Report Spreadsheet Statistical Text/word OtherSoftware Generation Retrieval Generation Analysis Processing (pleasespccit’)MONTHSA-i5 Could you please indicate very briefly why you no longer use the CAS.Please go on to the next page129FIRST WE WOULD LIKE TO GET YOUR IMPRESSIONS OF THE CAS. IN THEFOLLOWING, WE WILL PRESENT YOU WITH A NUMBER OF STATEMENTSEXPRESSING PARTICULAR VIEWPOINTS ABOUT THE CAS. WE WOULD LIKE YOU TOINDICATE HOW MUCH EACH STATEMENT REFLECTS YOUR PERSONAL VIEWPOINTBY PLACING AN ‘X’ IN THE APPROPRIATE PLACE ON THE DISAGREE-AGREE SCALESPROVIDED. ALTHOUGH THERE MAY APPEAR TO BE A NUMBER OF SIMILARSTATEMENTS, 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 agreestrongly quite slightly neither slightly quite stronglyU-2 Using a CAS is completely compatible with my current situation.disagree I I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-3 Using a CAS is compatible with all aspects of my work.disagree I I I I I Istrongly quite slightly neither slightly quite stronglyU-4 My superiors expect me to use a CAS.disagrecl I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-S I believe that a CAS is cumbersome to use.disagreel I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-6 Using a CAS improves my image within the organization.disagreel I I I I I I agreestrongly quite slightly neither slightly quite strongly130U-7 Using a CAS improves the quality of work I do.disagree I I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-8 Using a CAS makes it easier to do my job.dLsagreel I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-9 I think that using a CAS fits well with the way I like to work.disagree I I I I I I iagleestrongly quite slightly neither slightly quite stronglyU-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 agreestrongly quite slightly neither slightly quite stronglyU-li I have seen what others do using their CAS.disagree I I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-12 I’ve had a great deal of opportunity to try various CAS applications.disagree I I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-13 In my organization, one sees CAS on many desks.disagree I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-14 My boss does not require me to use a CAS.disagreel I I I I I I agreestrongly quite slightly neither slightly quite strongly131U-15 I would have no difficulty telling others about the results of using a CAS.disagreci I I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-16 I know where I can go to satisfactorily try out various uses of a CAS.disagreci I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-17 People in my organization who use a CAS have more prestige than those who do not.disagree I I I I I I recstrongly quite slightly neither slightly quite stronglyU-18 Although it might be helpful, using a CAS is certainly not compulsory in my job.disagreel I I I I I jagreestrongly quite slightly neither slightly quite stronglyU19 My using a CAS requires a lot of mental effort.disagreel I I I I agreestrongly quite slightly neither slightly quite stronglyU-20 Using a CAS is often frustrating.disagree I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-21 People in my organization who use a CAS have a high profile.disagree I I agreestrongly quite slightly neither slightly quite stronglyU-22 A CAS was available to me to adequately test run various applications.disagreej I I I I I I Istrongly quite slightly neither slightly quite strongly132U-23 I believe I could communicate to others the consequences of using a CAS.disagree [strongly quite slightly neitherU-24 I believe that it is easy to get a CAS to do what Idisagreel I I Istrongly quite slightlyU-25 Overall, I believe that a CAS is easy to use.disagree I I Istrongly quite slightly neitherU-26 Using a CAS improves my job performance.disagreeL I I Islightly quite stronglywant it to do.I I I II I I II I I IU-29 Before deciding whether to use any CAS applications, I was able to properly try them out.strongly quite slightlyU-30 Learning to operate a CAS is easy for me.I I I I I Ineither slightly quite stronglyslightly quite stronglyagreeagreeagreeagreeagreeagreestrongly quite slightly neither slightly quite stronglyU27 CAS are not very visible in my organization.disagreel I I Istrongly quite slightly neither slightly quite stronglyU-28 Overall, I find using a CAS to be advantageous to my job.disagreel j I I Istrongly quite slightly neither slightly. I Iquite stronglydisagree I I I I Idisagree Ineither slightly quite stronglyI I II jagreej agreestrongly quite slightly neither slightly quite strongly133U-3 1 Using a CAS enhances my effectiveness on the job.disagrec[ I I I I I I lagreestrongly quits slightly neither slightly quite stronglyU-32 Using a CAS fits into my work style.disagrcc I I I I I I Istrongly quits slightly neither slightly quite stronglyU-33 I would have difficulty explaining why a CAS may or may not be beneficial.disagrec I I I I I I agreestrongly quits slightly neither slightly quits stronglyU-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 agreestrongly quite slightly neither slightly quits stronglyU-35 Using a CAS gives me greater control over my work.disagrec I I I I 1 I 1 agreestrongly quite slightly neither slightly quits stronglyU-36 Using a CAS increases my productivity.disagreel I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-37 Having a CAS is a status symbol in my organization.disagree I I I I I I agreestrongly quite slightly neither slightly quits stronglyU-38 It is easy for me to observe others using CAS in my firm.disagrecl I I I I I lagreestrongly quits slightly neither slightly quits stronglyU-39 I have had plenty of opportunity to see the CAS being used.disagreel I I I I I Istrongly quite slightly neither slightly quite strongly134FINALLY, IN THIS SECTION WE WOULD LIKE TO ASK YOU A FEW QUESTIONSABOUT YOUR USE OF THE CAS.B-i Overall, my using a CAS in my job is (place an X on all four scales):wmenegativeextremely quite slightly neither slightly quite extremelyfoolishpo8itweB-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 I I I I I I I unlikelyextremely quite slightly neither slightly quite extremelyimprobableprobablegoodi I I I I I I Ihannfiilextremely quite slightly neither slightly quite extremelyI I I I I I Iextremely quite slightly neither slightly quite extremelybadbeneficialI I I I I I Iextremely quite slightly neither slightly quite extremelyI I I I I I I Iextremely quite slightly neither slightly quite extremely135B-3 Approximately when (month and year) did you first start using a CAS beyond any trial of it youmay have carried out?MONTH YEARB-4 How regularly do you now use a CAS? (Place an ‘X’ under the appropriate column):Less than About 1-3 About About More thanonce per times per once per 2-4 times once per once perNot at all month month week per week day dayI I IB-S On average, how frequently do you currently use the following functions (place an ‘X’ under theappropriate column):Less Aboutthan 1-3 About 2-4 About Moreonce times once times once thanNot at per per per per per onceall month month week week day per dayAccounting SoftwareGraphics GenerationInformation RetrievalReport GenerationSpreadsheetStatistical AnalysisText/word ProcessingOther (please specify)136B-6 For each of the following questions, place an ‘X’ under the appropriate column for eachapplicable function:a. On average how many hours per week do you spend using the CAS on the followingfunctions?b. Please indicate approximately how long (in months) you have been regularly using anyof the following functions.OtherAccounting Graphics Information Report Statisi.ical Texilword (pleascSoftware Generation Retneval Generation Spreadsheet Analysis Processing pecify)HOURSMONTHSB-7 Overall, how do you expect your usage of the CAS will change in the six months? (Placean ‘X’ in the appropriate column):increasel I I I I I I I decreasesignifi- some- marginally same marginally some- significantly what what cantlyB-8 Overall, how has your usage of CAS changed in the last six months? (Place an ‘X’ in theappropriate column):increasedi I I I I I I I decreasedaigniii- some- marginally same marginally some- significantly what what cantlyB-9 I have been using a CAS for (place an ‘X’ under the appropriate column):Less than About 1-3 About About More thanonce per times per once per 2-4 times once per once perNot at all month month week per week day day137B-1O When I started using my CAS, I received continuing support (training or help) for my CAS fromthe following sources (place an ‘X’ under the appropriate column for each source):About 1- MoreLess than 3 Time. About 2- About than OnceNot at Once per per Once per 4 Times Once per per DayAll Month Month Week per Week DayOther personnel from mycompanyPersonal friend (nonemployee)Public accounting firmNon-Accountant computerconsultantOther(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- MoreLess than 3 Times About 2- About than OnceNot at Once per per Once per 4 Times Once per per DayAll Month Month Week per Week DayOther personnel from mycompanyPersonal friend (nonemployee)Public accounting firmNon-Accountant computerconsultantOther(please specify)138B-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 ongoing1 2 J 3 some 5 6Other personnel from mycompanyPersonal friend (non-employee)Public accounting finnNon-Accountant computerconsultantOther(please specify)B-13 I plan on getting my future CAS help from the following sources (place an ‘X’ under theappropriate column for each applicable source):none ongoing1 2 3 some 5 6 7Other personnel from mycompanyPersonal friend (non-employee)Public accounting finnNon-Accountant computerconsultantOther(please specify)139B-14 I am satisfied with the current level of continuing support for my CAS that I receive from thefollowing sources (place an ‘X’ under the appropriate column for each applicable source):satisfied unsatisfied N/A(Don’treceiveanyauppoit)extremely quite slightly neither slightly quite extremelyOther personnel frommy companyPersonal friend (nonemployee)Public accountingfirmNon-Accountantcomputer consultantOther(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 ineffectiveextremely quite slightly neither slightly quite extremelyOther personnel frommy companyPersonal friend (non-employee)Public_accounting_firmNon-Accountantcomputer consultantOther(please specify)140B-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 ineffectiveextremely quite slightly neither slightly quite extremelyOther personnel frommy_companyPersonal friend (nonemployee)Public accounting firmNon-Accountantcomputer consultantOther(please specify)THANK YOU FOR YOUR PERSEVERANCE AND COOPERATION SO FAR. NOW,PLEASE GO ON TO THE NEXT PAGE.141IN THE FOLLOWING, WE WILL PRESENT YOU WITH A NUMBER OF STATEMENTSEXPRESSING PARTICULAR VIEWPOINTS ABOUT THE CAS. WE WOULD LIKE YOUTO INDICATE HOW MUCH EACH STATEMENT REFLECTS YOUR PERSONALVIEWPOINT PLACING AN ‘X’ IN THE APPROPRIATE PLACE ON ThE DISAGREE-AGREE SCALE. ALTHOUGH THERE MAY APPEAR TO BE A NUMBER OF SIMILARSTATEMENTS, 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 Istrongly quits slightly neither slightly quite stronglyN-2 Using a CAS would improve the quality of work I do.disagree I I I I I I I agreestrongly quits slightly neither slightly quite stronglyN-3 Using a CAS would be compatible with all aspects of my work.disagree I I I I I I agreestrongly quits slightly neither slightly quite stronglyN-4 My superiors expect me to use a CAS.disagrecl I I I I I I ] agreestrongly quits slightly neither slightly quits stronglyN-5 I believe that a CAS would be cumbersome to use.disagree 1 I I I I I agreestrongly quits slightly neither slightly quits stronglyN-6 Using a CAS would improve my image within the organization.disagrecf I I I I I I I agreestrongly quite slightly neither slightly quite stronglyN-7 Using a CAS would be completely compatible with my current situation.disagreei I I I I I I agreestrongly quits slightly neither slightly quits strongly142N-8 Using a CAS would make it easier to do my job.disagrecj I I I I I Istrongly quits slightly neither slightly quite stronglyN-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 agreestrongly quits slightly neither slightly quite stronglyN-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 agreestrongly quite slightly neither slightly quite stronglyN-Il I have seen what others do using their CAS.disagreel I I I I I I I agreestrongly quits slightly neither slightly quite stronglyN-12 I’ve had a great deal of opportunity to try various CAS applications.disagree I I I I I agreestrongly quits slightly neither slightly quits stronglyN-13 In my organization, one sees CAS on many desks.disagree I I I I I I agreestrongly quite slightly neither slightly quits stronglyN-14 My boss does not require me to use a CAS.disagree I I I I I I agreestrongly quits slightly neither slightly quits stronglyN-15 I would have difficulty telling others about the results of using a CAS.disagree I I I I I I agreestrongly quite slightly neither slightly quite strongly143N-16 I know where I can go to satisfactorily try out various uses of a CAS.disagrec I I I I I I agreestrongly quite slightly neither slightly quite stronglyN-17 People in my organization who use a CAS have more prestige than those who do not.disagreel I I I I I I Istrongly quite slightly neither slightly quite stronglyN-18 Although it might be helpful, using a CAS is certainly not compulsory in my job.disagrees I I I I I agreestrongly quite slightly neither slightly quite stronglyN-19 My using a CAS would require a lot of mental effort.disagrcc I I I I I I I agreestrongly quite slightly neither slightly quite stronglyN-20 Using a CAS would often be frustrating.disagreel I I I I I I I agreestrongly quite slightly neither slightly quite stronglyN-21 People in my organization who use a CAS have a high profile.disagree I I I I I I agreestrongly quite slightly neither slightly quite stronglyN-22 A CAS is available to me to adequately test run various applications.disagreel I I I I I I agreestrongly quite slightly neither slightly quite stronglyN-23 I believe I could communicate to others the consequence of using a CAS.disagree I I I I I agreestrongly quite slightly neither slightly quite strongly144N-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 agreestrongly quite slightly neither slightly quite stronglyN-25 Overall, I believe that a CAS would be easy to use.diaagreei I I I I I I I restrongly quite slightly neither slightly quite stronglyN-26 Using a CAS would improve my job performance.disagree I I I I I I I agreestrongly quite slightly neither slightly quite stronglyN-27 CAS are not very visible in my organization.disagree I I I I I I I agreestrongly quite slightly neither slightly quite stronglyN-28 Overall, I would find using a CAS to be advantageous in my job.disagreci I I I I I Istrongly quite slightly neither slightly quite stronglyN-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 agreestrongly quite slightly neither slightly quite stronglyN-30 Learning to operate a CAS would be easy for me.disagree I I I I I I agreestrongly quite slightly neither slightly quite stronglyN-31 Using a CAS would enhance my effectiveness on the job.disagree[ I I I I I I lagrecstrongly quite slightly neither slightly quite strongly145N-32 Using a CAS would fit into my work style.disagreel I I I Istrongly quits slightly neither slightly quite stronglyN-33 If I were to use a CAS, I would have difficulty explaining why using a CAS may or may not bebeneficial.disagree I I I I I agreestrongly quits slightly neither slightly quite stronglyN-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 Istrongly quite slightly neither slightly quite stronglyN-35 Using a CAS would give me greater control over my work.disagreel I I Istrongly quite slightly neither slightly quite stronglyN-36 Using a CAS would increase my productivity.disagreci I I I Istrongly quite slightly neither slightly quite stronglyN-37 Having a CAS is a status symbol in my organization.disagreel I I I Iquite stronglyI I Istrongly quite slightly neither slightlyN-38 It is easy for me to observe others using a CAS in my firm.disagreci I — I I Istrongly quits slightly neither slightlyN-39 I have had plenty of opportunity to see the CAS being used.disagree I I I I I IagreeIjagreejagreejagreeagreeagreeagreequite stronglystrongly quite slightly neither slightly quite strongly146FINALLY, 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 extremelyhannfulf I I I I I Iextremely quite slightly neither slightly quite extremelywme I I I I I foolishextremely quite slightly neither slightly quite extremelynegative L__.._. I I I I I I positiveextremely quite slightly neither slightly quite extremelyC-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?likelyunlikelyextremely quite slightly neither slightly quite extremelyimprobableprobableextremely quite slightly neither slightly quite extremely147C-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 Aboutthan 1-3 2-4 MoreOnce Times Once Times About than Did notNotat per per per per Once Once UseAU Month Month Week Week per Day per Day CASOther personnel frommy companyPersona] friend (nonemployee)Public accounting finnNon-Accountantcomputer consultantOther(please specify)C-4 How effective do you feel the following were in helping you use a CAS? (Place an ‘X’ underthe appropriate column for each applicable source):Less About Aboutthan 1-3 2-4 MoreOnce Times Once Times About than Did notNotat per per per per Once Once UseA]] Month Month Week Week per Day per Day CASOther personnel frommy companyPersonal friend (non-employee)Public accounting firmNon-Accountantcomputer consultantOther(please specify)148C-.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 columnfor each applicable source):Other personnel Non-accountantfrom my Professional computercompany Personal friend accounting firm consultant NoneTHANK YOU FOR YOUR PERSEVERANCE AND COOPERATION SO FAR. NOW,PLEASE GO ON TO THE NEXT PAGE.149In this last section, we would like to ask you some questions about yourself. Remember, all answers areconfidential, 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 ORDISAGREEMENT 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 1agstrongly quite slightly neither slightly quite strongly1-2 I rarely trust new ideas until I can see whether the vast majority of people around me acceptthem.disagreei I I I I .1 Istrongly quite slightly neither slightly quite strongly1-3 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 agreestrongly quite slightly neither slightly quite strongly1-4 1 am reluctant about adopting new ways of doing things until I see them working for peoplearound me.disagree I I I I I I agreestrongly quite slightly neither slightly quite strongly1-5 1 find it stimulating to be original in my thinking and behaviour.disagreel I I I I I I agreestrongly quite slightly neither slightly quite strongly1-6 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 agreestrongly quite slightly neither slightly quite strongly1501-7 I am challenged by ambiguities and unsolved problems.disagreef I I I I I I f agecstrongly quite slightly neither slightly quite strongly1-8 I must see other people using new innovations before I will consider them.aisagrecf I I I I I I (agreestrongly quite slightly neither slightly quite strongly1-9 1 am challenged by unanswered questions.disagree I I I I I I I agreestrongly quite slightly neither slightly quite strongly1-10 1 often find myself sceptical of new ideas.disagree I I I I I I I agreestrongly quite slightly neither slightly quite stronglyNEXT, WOULD YOU PLEASE INDICATE HOW LIKELY OR UNLIKELY EACH OFTHE FOLLOWING STATEMENTS ARE BY ONCE AGAIN PLACING AN ‘X’ IN THEAIPROPRIATE SPACE.S-i Most people who are important to me think I should use the CAS in my job.likely unlikelyextremely quite slightly neither slightly quite extremelyS-2 My close friends think that I should use the CAS in my job.likely I I I I I I unlikelyextremely quite slightly neither slightly quite extremelyS-3 My co-workers (peers) think that I should use the CAS in my job.likely I I I I unlikelyextremely quite slightly neither slightly quite extremely151S-4 My immediate superiors think that I should use the CAS in my job.likelyl I I I I I I unlikelyextremely quite slightly neither slightly quite extremelyS-5 Senior management thinks that I should use the CAS in my job.lik.eIyI I I I I I I I unlikelyextremely quite slightly neither slightly quite extremelyS-6 My subordinates think I should use the CAS in my job.likelyl I I I I I I I uldlyextremely quite slightly neither slightly quite extremelyS-7 Generally speaking, I want to do what most people who are important to me think I should do.likely unlikelyextremely quite slightly neither slightly quite extremelyS-8 Generally speaking, I want to do what my close friends think I should do.likely I I I I I I unlikelyextremely quite slightly neither slightly quite extremelyS-9 Generally speaking, I want to do what my co-workers think I should do.1iicly I I I I 1 I I unlilcelyextremely quite slightly neither slightly quite extremelyS-lO Generally speaking, I want to do what my immediate supervisors think I should do.likely I I I I I I J unlikelyextremely quite slightly neither slightly quite extremelyS-li Generally speaking, I want to do what senior management thinks I should do.likely I I I I unlikelyextremely quite slightly neither slightly quite extremelyS-12 Generally speaking, I want to do what my subordinates think I should do. 152likely I I I I I I I unlikelyextremely quite 8lghtly neither 8lightly quite extremelyFINALLY, 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):MALEFEMALEP-2 Your present age: yearsP-3 Your department:P4 Your job title:______________________________________________P-5 Years you have worked in your current department: yearsP-6 Years you have worked in this company: yearsP-7 What is the highest level of education that you completed? Place an ‘X’ beside the appropriatecolumn):GRADE SCHOOLSOME HIGH SCHOOLHIGH SCHOOL GRADUATE___SOME TECHNICAL COLLEGETECHNICAL COLLEGE GRADUATESOME COMMUNITY COLLEGECOMMUNITY COLLEGE GRADUATESOME UNIVERSITYUNIVERSITY GRADUATEPOSTGRADUATE153P-8 The job that best describes my organizational level is (place an ‘X’ beside the appropriatecolumn):___EXECUTIVE/TOP MANAGEMENTMIDDLE MANAGEMENTSUPERVISORYPROFESSIONALTECHNICAL__CLERICALOTHER (please specify)GENERAL BUSINESS INFORMATIONF-i Number of: Employees Accounting StaffFull TimePart TimeF-2 Annual sales last year:< $250,000 < $500,000 < $1,000,000 c $10,000,000 > $10,000,000F-3 Type of organization (e.g., profit, non-profit, CO-OP. etc...)F-4 What industry does your firm operate in?INDUSTRY:_____________________F-5 Does your firm plan to implement or expand a CAS in the next two years?YES NO DON’T KNOWIf YES, approximately how much do you expect your firm to spend on the CAS in this time?$______________154F-6 THANK YOU VERY MUCH FOR YOUR PARTICWATION!If you wish to add any comments or further observations, please use the space below or simplyattach them to this page.1 cI ._JAppendix II- A2Pilot Study156WELCOME!You are about to participate in a study of opinions about the usage of microcomputers in the accountingfunction. In some sections you will be asked questions about and see reference to the term CAS, whichstands for COMPUTERIZED ACCOUNTING SYSTEM. A CAS is defined for the purposes of thisstudy as a set of computerized tools for an individual, and usually consists of a personal ormicrocomputer with one or more software packages, such as an accounting program, and/or othersoftware such as spreadsheet, database, word processing, etc. in support of the accounting function. Thekey aspect of a CAS is that it is computer technology that you would use directly, as opposed to havingsomeone 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 pleasedon’t spend an excessive amount of time on each question.157INSTRUCTIONSIn the attached questionnaire, we ask questions which make use of rating scales with seven places; youare asked to place an ‘X’ in the place that best describes your opinion. For example, if you were askedto 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 unlikelyextremely quite slightly neither slightly quite extremelyIf you think that it is extremely likely that driving a car in winter is easy, you would make your markas follows:Driving a car in winter is easy.likely X I I I I unlikelyextremely quite slightly neither slightly quite extremelyIf you think that it is neither likely nor unlikely that driving a car in winter is easy, you would makeyour mark as follows:Driving a car in winter is easy.likely f I I I X I I I unlikelyextremely quite slightly neither slightly quite extremelyIn addition to the “likely-unlikely” pairs, other pairs such as “disagree-agree” will also be used. Theyshould 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 unlikelyextremely quite slightly neither slightly quite extremelyTillS NOT TillS2. 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 lettercorresponding to a particular answer for a question. Please be careful to see that your circle goes aroundonly the letter or number which corresponds to your desired response.158SECTION ATO BEGIN, WE WOULD LIKE TO ASK YOU ABOUT YOUR EXPERIENCE WITHCOMPUTERS AND OTHER HIGH-TECHNOLOGY PRODUCTS AND SERVICES.A-i Have you ever used a multi-function telephone (including such functions as call forward, speeddialing, call waiting, etc.). (Place an ‘X’ beside the appropriate answer):NO)L ysIf you use a multi-function phone, which functions do you use? (Place an ‘X’ beside theappropriate functions):CALL TRANSFER(CONSULTATIONS) HOLDTHREE-WAY CONFERENCECALL FORWARDINGCALL PARKING___CALL PICKUPCALL WAiTINGRING AGAIN/AUTOMATIC CALL BACK% SPEED CALLINGX LAST NUMBER DIALLEDSAVE NUMBER AND REPEAT159A-2 How often do you use the products listed below? (Place an ‘X’ under the appropriate columnfor each applicable area):About 1- MoreLess than 3 Times About 2- About than OnceNot at Once per per Once per 4 Times Once per per DayAll Month Month Week per Week Daya. Automated TellerMachineb. Programmable Calculatorsc. Home Computersd. Business Computers >c. Video Gamesf. Programmable MicrowaveOvensA-3 How often do you carry out the computer-related activities listed below; on paper, via electronicmail, on floppy disk, etc...? (Place an ‘X’ under the appropriate column for each applicablearea):About 1-Less than 3 Times About 2- About MoreNot at Once per per Once per 4 Times Once per than Onceall Month Month Week per Week Day per DayReceive computer outputx(reports/documents)Submit documents, etc. to Xothers for word processingSubmit data to others forcomputer analysis160A-4 What is your current keyboarding (typing) ability?a. Mark with an ‘X’:___HUNT & PECK< TOUCH TYPEb. Place an ‘X’ in the place that best reflects your speedI I I I>I I I IOwpm 1-15 16-30 31-45 46-60 61-75 >75wpm wpm wpm wpm wpm wpmA-5 How many educational courses (at any level) have you had about computers, but which did notinclude your personal hands-on use?COURSESA-6 How many educational courses have you had which required your personal hands-on use ofcomputers?COURSESA-7 My firm receives non-CAS support for the following areas (place an ‘X’ under the appropriatecolumn for each applicable area):none constant1 2 3 j some 5 6 7AccountingAuditBusiness AdviceFinancial PlanningGov’t Compliance—MarketingTax XOther(please specify)161A-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 constant1 2 3 1 some 5 6Personal friend (nonemployee)Public accounting finnNon-Accountant computerconsultantNoneOther(please specify)A-9 I am satisfied with the current level of support for non-CAS areas I receive from the followingsources external to the firm (place an ‘X’ under the appropriate column for each applicablesource):satisfied unsatisfiedextremely quite slightly neither slightly quite extremelyPersonal friend (noncmploy)Public accounting finnNon-Accountantcomputer consultantNoneOther(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 huntedextremely quite slightly neither slightly quite extremely162A-i 1 How knowledgeable do you feel you are of the uses of the CAS?unlimited I < I I I I I limitedextremely quite slightly neither slightly quite extremelyA-12 Have you ever used a CAS? (Place an ‘X’ beside the appropriate column):)( CURRENTLY USE A CAS_______________PLEASE SKIP TO SECTION BHAVE NEVER USED A CAS PLEASE SKIP TO SECTION CUSED TO USE A CAS BUT NO LONGER DO SOPlease 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 youstopped using it.STARTED______ ______MONTH YEARSTOPPEDMONTH YEARA-i4 Please indicate which of the CAS functions below you used by indicating the number of monthsyou used them.Accounting Graphics Information Report Spreadsheet Statistical Textlword OtherSoftware Generation Retrieval Generation Analysis Processing (pleasespecify)MONTHS[A-i5 Could you please indicate very briefly why you no longer use the CAS.PLEASE SKIP TO SECTION CSECTION B 163Please answer questions in this section only if you currently use the CAS.FIRST WE WOULD LIKE TO GET YOUR IMPRESSIONS OF THE CAS. IN THEFOLLOWING, WE WILL PRESENT YOU WITH A NUMBER OF STATEMENTSEXPRESSING PARTICULAR VIEWPOINTS ABOUT THE CAS. WE WOULD LIKE YOU TOINDICATE HOW MUCH EACH STATEMENT REFLECTS YOUR PERSONAL VIEWPOINTBY PLACING AN ‘X’ IN THE APPROPRIATE PLACE ON THE DISAGREE-AGREE SCALESPROVIDED. ALTHOUGH THERE MAY APPEAR TO BE A NUMBER OF SIMILARSTATEMENTS, 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 Istrongly quite slightly neither slightly quite stronglyU-2 Using a CAS is completely compatible with my current situation.disagreej I I I I I ><‘ I iagstrongly quite slightly neither slightly quite stronglyU-3 Using a CAS is compatible with all aspects of my work.disagree I I I I I X agreestrongly quitc slightly neither slightly quite stronglyU-4 My superiors expect me to use a CAS.1disagrce I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-5 I believe that a CAS is cumbersome to use.Idisagree X I I I I I agreestrongly quite slightly neither slightly quite stronglyU-6 Using a CAS improves my image within the organization.disagree I I I agreestrongly quite slightly neither slightly quite strongly164dagree I I I Istrongly quite slightly neither slightlyU-9 I think that using a CAS fits well with the way I like to work.disagrec I I I I Istrongly quite slightly neither slightly quiteU-lO My use of a CAS is voluntary (as opposed to required by my superiorsdisagreel I I <I I I Istrongly quite slightlyU-il I have seen what others do using their CAS.disagreci IU-12U-7 Using a CAS improves the quality of work I do.disagrec I I Istrongly quite slightlyU-8 Using a CAS makes it easier to do my job.I -neither slightly quitej agreeagreestronglyxquite strongly>( agrecstronglyor job description).agreeneither slightly quite stronglyU-13.lxistrongly quite slightly neither slightly quite stronglyI’ve had a great deal of opportunity to try various CAS applications.disagrec I I I Istrongly quite slightly neither slightly quite stronglyIn my organization, one sees CAS on many desks.disagree I Iagreeagreeagreeagreestrongly quite slightly neither slightly quite stronglyU-14 My boss does not require me to use a CAS.stronglydisagree)<} I I I I I Iquite slightly neither slightly quite stronglyIIU-17I would have no difficulty telling others about the results of using a CAS.disagrecf I I I I I Istrongly quite slightly neither slightly quite stronglyI know where I can go to satisfactorily try out various uses of a CAS.disagreestrongly quite slightly neither slightly quite stronglyPeople in my organization who use a CAS have more prestige than those who do not.disagrcc I I I I I Istrongly quite slightly neither slightly quite stronglyU-18 Although it might be helpful, using a CAS is certainly not compulsory in my job.disagree‘< I I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-19 My using a CAS requires a lot of mental effort.‘xdisagreestrongly quite slightly neither slightly quite stronglyUsing a CAS is often frustrating.I Idisagree I I I I Istrongly quite slightly neither slightly quite stronglyPeople in my organization who use a CAS have a high profile.disagreestrongly quite slightly neither slightly quite stronglyA CAS was available to me to adequately test run various applications.disagrees I I Istrongly quite slightly neither slightly quite stronglyagreeU-15U-16-1 165agreeagreeagreeU-20U-21U-22agreeagreeagree166U-23 I believe I could communicate to others the consequences of using a CAS.xdisagreestrongly quite slightly neither slightly quite stronglyU-24 I believe that it is easy to get a CAS to do what I want it to do.disagree_____________________________________________________strongly quite slightly neither slightlyU-25 Overall, I believe that a CAS is easy to use.disagree [__L I I I I I>cI I agreestrongly quite slightly neither slightly quite stronglyU-26 Using a CAS improves my job performance.disagreci I I Istrongly quite slightly neitherU-27 CAS are not very visible in my organization.disagreei‘<‘ I I I Istrongly quite slightly neither slightly quite stronglyU-28 Overall, I find using a CAS to be advantageous to my job.disagreef I I I I IU-29I agreeagreequite stronglyslightly quite stronglyJagrccjagreagreeU-30strongly quite slightly neither slightly quite stronglyBefore deciding whether to use any CAS applications, I was able to properly try them out.disagree f I > I I I I I agreestrongly quite slightly neither slightly quite stronglyLearning to operate a CAS is easy for me.disagree I I I I I I I agreestrongly quite slightly neither slightly quite strongly167U-31 Using a CAS enhances my effectiveness on the job.disagree I L Istrongly quite slightly neither slightly quite 8troflglyU-32 Using a CAS fits into my work style.disagree I I I I I I I agreestrongly quite slightly neither slightly quite stronglyU-33 I would have difficulty explaining why a CAS may or may not be beneficial.disagree I X I I I agreestrongly quite slightly neither slightly quite stronglyU-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 slightlyU-35 Using a CAS gives me greater control over my work.I I I I I II I Iquite stronglyneither slightly quiteI Istrongly quite slightly neither slightly quite 3tronglyU-37 Having a CAS is a status symbol in my organization.disagiestrongly quite slightly neither slightly quite stronglyU-38 It is easy for me to observe others using CAS in my firm.disagree I I Iagreedisagreestrongly quite slightlyU-36 Using a CAS increases my productivity.disagrcej Istrongly>cagreeagreeagreeagreeagree-II jstrongly quite slightly neither slightly quite strongly168[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 extremelyharmful f I I I I I X beneficialextremely quite slightly neither slightly quite extremelywiscj I I I I I Ifoolishextremely quite slightly neither slightly quite extremelynegative I I I I I potiveextremely quite slightly neither slightly quite extremelyB-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 unlikelyextremely quite slightly neither slightly quite extremely1improbableprobableextremely quite slightly neither slightly quite extremelyB-3 Approximately when (month and year) did you first start using a CAS beyond any trial of it youmay have carried out?10MOWTH YEAR169B-4 Overall, how many hours per week do you use a CAS? ‘0HOURSB-5 How regularly do you now use a CAS? (Place an ‘X’ under the appropriate column):Less than About 1-3 About About More thanonce per times per once per 2-4 times once per once perNot 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/minicomputer, on a microcomputer, on both, or on neither. (Place an ‘X’ under the appropriatecolumn):Mainframe?Mini Micro Both NeitherAccounting SoftwareGraphics GenerationInformation RetrievalReport GenerationSpreadsheetStatistical Analysis >(Textlword ProcessingOther (please specify170B-7 On average, how frequently do you currently use the following functions (place an ‘X’ under theappropriate column):Less Aboutthan 1-3 About 2-4 About Moreonce times once times once thanNot at per per per per per onceall month month week week day per dayAccounting SoftwareGraphics GenerationInformation RetrievalReport GenerationSpreadsheetStatistical Analysis >cText/word ProcessingOther (please specify)B-8 For each of the following questions, place an ‘X’ under the appropriate column for eachapplicable function:a. On average how many hours per week do you spend using the CAS on the followingfunctions?b. Please indicate approximately how long (in months) you have been regularly using anyof the following functions.OtherAccounting Graphics Jnformation Report Statistical Textlword (pleaseSoftware Generation Retrieval Generation Spreadsheet Analysis Processing specify)HOURS I I IMONTHS I 8- 8— 8—171B-9 Overall, how do you expect your usage of the CF\S will change in the p six months? (Placean ‘X’ in the appropriate column):increasedi I I I I I Idsigmfi- some- marginally same marginally some- significandy what what cantlyB-lO Overall, how has your usage of CAS changed in the last six months? (Place an ‘X’ in theappropriate column):increasedf I I I I fdecreasedsignifi- some- marginally same marginally some- significandy what what candyB-li I have been using a CAS for (place an ‘X’ under the appropriate column):Less than About 1-3 About About More thanonce per times per once per 2-4 times once per once perNot at all month month week per week day day.>(172B-12 When I started using my CAS, I received continuing support (training or help) for my CAS fromthe following sources (place an ‘X’ under the appropriate column for each source):About 1- MoreLess than 3 Times About 2- About than OnceNot at Once per per Once per 4 Times Once per per DayAU Month Month Week per Week DayOther personnel from mycompanyPersonal friend (nonemployee)Public accounting firmNon-Accountant computerconsultantOther(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- MoreLess than 3 Times About 2- About than OnceNot at Once per per Once per 4 Times Once per per DayAll Month Month Week per Week DayOther personnel from mycompanyPersonal friend (nonemployee)Public accounting firmNon-Accountant computerconsultantOther >{(please specify)173B-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 10times in total. Total may be less than 10.):‘‘1h123 41516 7.819110Personal friend (nonemployee)Public Accounting finn )(Non-Accountantcomputer consultantOther(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 constant1 2 3 some 5 6 7Other personnel from mycompanyPersonal friend (non-employee)Public accounting finnNon-Accountant computerconsultantOther(please specify)174B-16 I plan on getting my future CAS help from the following sources (place an ‘X’ under theappropriate column for each applicable source):none constant1 2 3 some 5 6 7Other personnel from mycompanyPersonal friend (non-employee)Public accounting firmNon-Accountant computerconsultantOther(please specify)B17 I am satisfied with the current level of continuing support for my CAS that I receive from thefollowing sources (place an ‘X’ under the appropriate column for each applicable source):satisfied unsatisfiedextremely quite slightly neither j slightly quite extremelyOther personiel frommy companyPersonal friend (non-employee)Public accounting firmNon-Accountantcomputer consultantNoneOther(please specify)175B-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 ineffectiveextremely quite slightly neither slightly quite extremelyOther personnel frommy companyPersonal fricnd (nonemployee) —_______Public accounting fri-rnNon-Accountantcomputer consultantOther(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 ineffectiveextremely quite slightly neither slightly quitc extremelyOther personnel frommy companyPersonal fiicnd (non-employee)Public accounting finn XNon-Accountantcomputer consultantOther(please specify)THANK YOU FOR YOUR PERSEVERANCE AND COOPERATION SO FAR. NOW,PLEASE SKIP TO THE FINAL SECTION, SECTION D.SECTION D 176In this last section, we would like to ask you some questions about yourself. Remember, all answers aieconfidential, 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 ORDISAGREEMENT WITH A NUMBER OF STATEMENTS; THIS TIME ABOUTYOURSELF. PLEASE PLACE AN ‘X’ IN THE APPROPRIATE SPACE.1-1 I am generally cautious about accepting new ideas.Adisagree agreestrongly quite slightly neither slightly quite strongly1-2 I rarely trust new ideas until I can see whether the vast majority of people around me acceptthem.I I I Istrongly quite slightly neither slightly quite strongly1-3 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 agreestrongly quite slightly neither slightly quite strongly1-4 I am reluctant about adopting new ways of doing things until I see them working for peoplearound me.disagrees I I xl I I I Istrongly quite slightly neither slightly quite strongly1-5 I find it stimulating to be original in my thinking and behaviour.disagree I I I I I Jagreestrongly quite slightly neither slightly quite strongly1-6 I tend to feel that the old way of living and doing things is the best way.disagree I I I I I I agreestrongly quite slightly neither slightly quite strongly1771-7 I am challenged by ambiguities and unsolved problems.disagree_____________________________________strongly quite sightly neither slightly1-8 I must see other people using new innovations before I willdisagreef I I Istrongly quite slightly1-9 I am challenged by unanswered questions.disagree I I Istrongly quite slightly1-10 I often find myself sceptical of new ideas.disagreci I I Istrongly quite slightly neitherNEXT, WOULD YOU PLEASE INDICATE HOW LIKELY OR UNLIKELY EACH OFTHE FOLLOWING STATEMENTS ARE BY ONCE AGAIN PLACING AN ‘X’ IN THEAPPROPRIATE 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 unlikelyextremely quite slightly neither slightly quite extremelyS-2 My close friends think that I should use the CAS in my job.likely I I ‘‘ I I I unlikelyextremely quite slightly neither slightly quite extremelyS-3 My co-workers (peers) think that I should use the CAS in my job.likely I I I I I I unlikelyI I I agreequite stronglyconsider them.agreeneither slightly quite stronglyneitherI I_i Islightly quite stronglyagreeagreeI I I Islightly quite stronglyextremely quite slightly neither slightly quite extremely178S-4 My immediate superiors think that I should use the CAS in my job.likely unlikelyextremely quite slightly neither slightly quite extremelyS-S Senior management thinks that I should use the CAS in my job.likely unlikelyextremely quite slightly neither slightly quite extremelyS-ó My subordinates think I should use the CAS in my job.likely unlikelyextremely quite slightly neither slightly quite extremelyS-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 unlikelyextremely quite slightly neither slightly quite extremelyS-8 Generally speaking, I want to do what my close friends think I should do.><likely unlikelyextremely quite slightly neither slightly quite extremelyS-9 Generally speaking, I want to do what my co-workers think I should do.likely I I I I I I unlikelyextremely quite slightly neither slightly quits extremelyS-1O Generally speaking, I want to do what my immediate supervisors think I should do.likely unlikelyextremely quite slightly neither slightly quite extremelyS-i 1 Generally speaking, I want to do what senior management thinks I should do.likely unlikelyextremely quite slightly neither slightly quite extremelyS-12 Generally speaking, I want to do what my subordinates think I should do.likely I I I ‘> I I I I unlikelyextremely quite slightly neither slightly quite extremely179FINALLY, WE WOULD LIKE TO ASK A FEW QUESTIONS ABOUT YOURSELFFOR STATISTICAL PURPOSES. COULD YOU PLEASE INDICATE: ——P-i Your sex (place an ‘X’ beside the appropriate coLn)KALE_______________FEMALEP-2 Your present age: —________ yearsP-3 Your department: i -r -iP-4 Your job title: CL-Dc)r--P-5 Years you have worked in your current department.1yearsP-6 Years you have worked in this coeçany. I “yearsP-7 What is the hi9hest LeveL of education that you coapLeted? (pLace an ‘X’ beside the appropriate colwn)GRADE SCHOOLSOME HIGH SCHOOl.HIGH SCHOOL GRADUATESOME TECHNICAL COLLEGETECHNICAL COLLEGE GRADUATESOME COMMUNITY COLLEGECOMMUNITY COLLEGE GRADUATESOME UNIVERSITYUNIVERSITY GRADUATEPOSTGRADUATEP-8 The job that best describes my organizationaL Level is (place an ‘X’ beside the appropriate coluT)_____EXECUTIVE/TOP MANAGEMENT_MIDDLE MANAGEMENTSUPERVISORYPROFESSIONAL/EXEMPTIc TECHNICAL/NON-PROFITCLERICALOTHER (pLease specify)GENERAL BUSINESS INFORMATION 180F-i Nuiter of: EirçLoyees Accounting staffFull time______-3Part timeF-2 Annual Sales last year (in thousands of dollars. k=1,000).< $250k I $250k-$SOOk I $500k-$1,000kF-3 Type of organization (eg. profit, non-profit, CO-OP, etc...)\Jj tF-4 Does your firm plan to iirçilement or expand a CAS in the next two years? (Yes or No)Yes No Don’t Know ‘7If Yes, approximately how nich do you expect to spend on the CAS in this time?$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 IAPPENDIX 11-B 181182Appendix II- BiClient Letter183YOUR FIRM’S LEtTERHEADDear Client:The University of British Columbia has contacted our firm about participating in a study on InformationTechnology (iT). They have also requested permission to contact our clients in order to ask you toparticipate in the study.Our firm has met with the researchers from UBC to find out more about the nature of the study. Webelieve that the results from this study would be important to both our finn and to our clients’ helpingus 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 questionnaireto the owner/manager, the chief accountant, and to any other accounting staff members interested inparticipating. 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 nothave access to your name or addrçss (unless you specifically include this information on thequestionnaire). All mailings are handled by our firm.2. Since you will be mailing the completed questionnaire back to the UBC researchers, no personnelfrom our accounting firm will have access to your responses.If you have not been provided with enough questionnaires, please call our office or photocopy sufficientadditional questionnaires.If you have any questions about this study please contact the UBC researchers at the phone number onthe attached letter.Our firm is not sponsoring or otherwise associated with either the research study or the UBCresearchers.184Appendix II- B2Partner LetterFaculty of Commerce Albert S. Dexter 185and Business Administration Associate Professor2053 Main Mall Management Information Systems_______Vancouver, B.C. Canada V6T 1Z2 Telephone: (604) 822-8380Fax: (604) 822-8489September 13, 1991Dear Sir/Madame:We are conducting a study at the University of British Columbia on Information Technology (iT). Wewould 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 beeninstalled successfully while others have not been very successful. The purpose of our research is todetermine what the difference is between firms that have successfully installed computer systems andthose that were not so successfully installed. We will obtain this information from a questionnaire thatasks respondents their opinions about using computers.We hope to use the results from this study to help owners and managers make sound business decisionsabout acquiring other iT in the future. It is undeniable that firms will be purchasing other IT in thefuture. Technology such as Teleconferencing, Networking, Image Processing, Desktop Publishing,Multimedia, etc., are currently becoming established as the newest forms of IT that many businesses arelooking at to improve their competitive position. Over the next five to ten years there will be other if’sthat 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 takeapproximately 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 aboutpersonally using computers. Other studies have shown that there is a link between what employees thinkand how an organization performs. Thus our results should enable organizations to better manage thespread of computers and other IT. As a token of our appreciation, once the stud is completed, wewould be pleased to send you a copy of our findings, conclusions and recommendations if you send usa card indicating your name and address. We hope to receive your completed questionnaire by the endof 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:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.831.1-0086714/manifest

Comment

Related Items