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

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

Item Metadata

Download

Media
[if-you-see-this-DO-NOT-CLICK]
ubc_1992_spring_laktin_richard_s.pdf [ 3.41MB ]
Metadata
JSON: 1.0086714.json
JSON-LD: 1.0086714+ld.json
RDF/XML (Pretty): 1.0086714.xml
RDF/JSON: 1.0086714+rdf.json
Turtle: 1.0086714+rdf-turtle.txt
N-Triples: 1.0086714+rdf-ntriples.txt
Original Record: 1.0086714 +original-record.json
Full Text
1.0086714.txt
Citation
1.0086714.ris

Full Text

ADOPTION OF iNFORMATION TECHNOLOGY IN ASMALL 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 thesisin partial fulfilment of the requirements for anadvanced degree at theUniversity of British Columbia, Iagree that the Library shall make it freelyavailable forreference and study. I further agreethat permission for extensive copyingof this thesis forscholarly purposes may be grantedby the head of my department orby his or herrepresentatives. It is understoodthat copying or publication of this thesisfor financial gainshall not be allowed without my writtenpermission.Department of Commerce andBusiness AdministrationThe University of BritishColumbia1956 Main MallVancouver, British ColumbiaCanadaV6T 1Y3Date: 30 April, 199211AB STRACTMany small businesses are turningto Information Technology as a means ofcompetitive advantage and survival in today’s tougherbusiness climate. The PublicAccounting profession portrays itself in therole of Information Consultant to small businesswhen it comes to information technology. Therole that Public Accountants play in theinformation technology adoptionprocess is poorly understood. The purpose of this researchwas to examine more closely the rolethat information consultants play in the adoptionprocess, with particular emphasison the public accountant.The Dffusion of Information Technology model (Moore, 1989)was used as thetheoretical foundation for this study. TheDiffusion of Information Technology model isllgrounded in theory and is supported byMoore’s research results.The major research questions answeredare:1. What role do independent informationconsultants such as accounting firms play in theDffiision ofInformation Technologyprocess?2. Is the Diffiis/on ofInformation Technology modela general model?A cross-sectional survey using a questionnaire wasissued to small business clients ofpublic accounting firms. Profiles of informationtechnology users and non-users weregenerated from questionnaire data. Theseprofiles were subject to regression analysis andstructural equation modelling using PLS (PartialLeast Squares). The analysis provided someanswers to the role accountants play inthe information technology adoption processas well assupporting the Diffusion of InformationTechnology model in a small business domain.111TABLE OF CONTENTSABSTRACTiiTABLE OF CONTENTSiiiLIST OF TABLESivLIST OFFIGURESvACKNOWLEDGEMENTSviiiCHAPTER 1: INTRODUCTION AND OVERVIEW OF RESEARCH11.1 RESEARCH STUDY RATIONALE 11.2 RESEARCH DIRECTION 31.3 THE COMPUTERIZED ACCOUNTING SYSTEM 41.4 TOWARDS A SMALL BUSINESS ORIENTATION51.5 THE ROLE OF INFORMATION TECHNOLOGY IN ORGANIZATIONS 7CHAPTER 2: LITERATURE REVIEW10CHAPTER 3: ADOPTION OF INFORMATIONTECHNOLOGY 133.1 DIFFUSION OF INNOVATIONS143.2 THE THEORY OF REASONED ACTION153.3 DIFFUSION OF INFORMATION TECHNOLOGY153.4 SIGNIFICANCE OF THE DIFFUSION OF INFORMATION TECHNOLOGYMODEL15CHAPTER 4: TECHNOLOGY TRANSFER- RESEARCH QUESTIONS 174.1 RESEARCH HYPOTHESES18CHAPTER 5: INSTRUMENT DEVELOPMENT22SECTION A - INTRODUCTION225.1 GENERAL225.1.1 RELIABILITY225.1.2 VALIDITY235.1.3 QUESTIONNAIRE SELECTION24SECTION B: QUESTIONNAIRE DESIGN - PILOT STUDY245.2 CLIENT/DIFFUSION OF INFORMATION TECHNOLOGYQUESTIONNAIRE 245.2.1 PERCEIVED CHARACTERISTICSOF INNOVATIONS 245.2.2 SYSTEM USAGE265.2.3 CLIENT COMPUTERIZED ACCOUNTING SYSTEMSUPPORT 30SECTION C: FINAL SURVEYS - SCALE RELIABILITIES305.4 GENERAL305.5RESULTS31SECTION D: QUESTIONNAIRE DESIGN315.6 GENERAL315.7 FORMAT325.7.1 PAMPHLET325.7.2 QUESTION LAYOUT335.7.3 COVERING LETTER33CHAPTER 6: DATA COLLECTION ANDANALYSIS 34SECTION A: DATA COLLECTION AND CONDITIONING346.1 INTRODUCTION346.2 SURVEY SAMPLE346.2.1 TARGET POPULATION SELECTION346.2.2 PROBLEMS ENCOUNTERED366.2.3 RESPONSE RATES376.3 CLIENT FIRM’S SURVEY386.3.1 RESULT DEMONSTRABILITY396.4 CONDITIONING THE DATA406.4.1 GENERAL406.4.2 ACCURACY OF INPUT DATA406.4.3 MISSING DATA40iv6.4.4 OUTLIERS AN]) SKEWNESS .416.4.5 NON-LINEARITY AND HOMOSCEDASTICITY 41SECTION B: DESCRIPTIVE STATISTICS426.5 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 GENERAL536.14.2 ATTITUDE. SUBJECTIVE NORM AND VOLUNTARINESS ONINNOVATIVENESS536.14.3 PERCEIVED CHARACTERISTICS OF INNOVATING.SUBJECTIVE NORM,AND VOLUNTARINESS ON INNOVATIVENESS556.14.4 OTHER REGRESSIONS57SECTION D: PATH MODELING596.15 CHOICE OF PATH MODEL COMPUTER IMPLEMENTATION - LISRELvsPLS606.15.1 DESIGN OF PLS PATH MODEL616.15.2 ANALYSIS OF SAMPLE SIZE REQUIREMENTS626.15.3 GOODNESS OF FIT DETERMINATION646.15.4 ASSESSMENT OF HYPOTHESES TESTING666.16 SUMMARY OF RESULTS: PATH ANALYSIS69SECTION F: SUMMARY OF DATA ANALYSIS706.17 GENERAL706.18 SUMMARY OF DESCRIPTIVE STATISTICS706.19 SUMMARY OF HYPOTHESES TESTING70CHAPTER 7: CONTRIBUTIONS. IMPLICATIONS AND LIMITATIONS737.1 INTRODUCTION737.2 SUMMARY OF THE RESEARCH PROCESS737.3 THE RESEARCH QUESTIONS ANSWERED747.3.1 QUESTION TWO747.3.2 QUESTION ONE757.4 CONTRIBUTIONS767.5 LIMITATIONS OF THE STUDY777.6 CONCLUSION78TABLES80FIGURES100BIBLIOGRAPHY108APPENDICES117VLIST 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: MOORE82TABLE 5(a) - DEMOGRAPHIC BACKGROU1D OFSURVEY RESPONDENTS 83TABLE 5(b) - DEMOGRAPHIC BACKGROU1D OF SURVEYRESPONDENTS 84TABLE 6(a) - SURVEY VARIABLES - DESCRIPTIVE STATISTICS85TABLE 6(b) - SURVEY VARIABLES - DESCRIPTIVESTATISTICS 86TABLE 7(a) - USERS VERSUS NON-USERS87TABLE 7(b) - USERS VERSUS NON-USERS88TABLE 8- REGRESSION RESULTS89TABLE 9- REGRESSION RESULTS90TABLE 10(a) - REGRESSION RESULTS91TABLE 10(b) - REGRESSION RESULTS92TABLE 11(a) - REGRESSION RESULTS93TABLE 11(b) - REGRESSION RESULTS94TABLE 12(a) - REGRESSION RESULTS95TABLE 12(b) - REGRESSION RESULTS96TABLE 13(a) - SUIVIMARY RESULTS OF HYPOTHESESTESTING 97TABLE 13(b) - SUMMARY RESULTS OF HYPOTHESESTESTING 98TABLE 14 - GENERAL PLS STATISTICS FOR TESTEDMODELS 99viLIST OF FIGURESFIGURE 1 - DIFFUSION OF INFORMATION TECHNOLOGY MODEL101FIGURE 2- DIFFUSION OF INNOVATIONS MODEL102FIGURE 3- INNOVATION DECISION MODEL103FIGURE 4- STAGES OF TNE INNOVATION DECISION PROCESS MODEL104FIGURE 5- NON-CAS USERS: RESULT DEMONSTRABILITY105FIGURE 6- DIFFUSION OF INFORMATION TECHNOLOGYMODEL: 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, whowas 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 helpfulcriticismand comments. I would also like to thank my fellow graduatestudents at UBC and the facultymembers in the Management Information Systems area for their supportand encouragement.I would like to express my appreciation to Wynne Chin from the Universityof 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 putup with the irritations, grumpiness,and frequent late night disturbances resulting from theprinting out of various drafts of thispaper. In addition she was instrumental in helping out with thedata verification process.Special thanks go out to my cousin Ann who was very helpful withdata verificationand reviewing this paper for grammatical errors.Thank you Lindsay, for your encouragement andinsightful suggestions on the finalversions of this paper.Finally, I would like to thank my parents, Cyriland Doris, for their financial andmaterial support provided at various times throughout thetime spent completing the graduateprogram. Without their help this thesis would nothave been completed.1CHAPTER 1: JNTRODUCTION AND OVERVIEW OFRESEARCH1.1 RESEARCH STuDY RATIONALEMost small firms have limited access to information sources on informationtechnology (IT). As a result, informationis often sought from an external informationconsultant (Goodson, 1990). The role of the external information consultantas an informationsource to small firms is an important research area.In the role of information consultantsprofessional accountants have been involved with manycomputer systems that have beenconsidered successful by their users, and several thathave 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 seemto be not only a product of planning buta product of fortune as well. To an accountant, working ina profession that sells informationand methods of generating informationas products, unsuccessful implementation ofcomputerized accounting systems isto be avoided. Maintaining good client relations is thebottom line to professional accounting organizations and failures(perceived or otherwise) areunacceptable, as small businesses cannot afford theemotional and monetary costs of anunsuccessfully implemented computerized accountingsystem. Research that can illuminate theinteraction between small firmsand their public accountant may provide the accountingprofession with an understanding of howto better deliver the current information technologyservices it already provides to small business clients.Equally as important to small businesses are suggestionsfor coping with informationtechnology and finding ways to increase productivitygiven the scarcity of trained and skilledspecialists. There is a growing belief thatinformation technology will be the most importanttechnology to change business and societyin the 1990’s as Canada moves from an economythat is resource based to one that is servicebased (Gunning, 1992). Small businesses may endup in the unenviable position of relying on informationtechnology much more than theycurrently are, and unable to find ready assistance (inthe form of skilled labour) to implementand manage the information technology theyrequire.2For the public accountant, this research should helpreinforce the need to beadequately trained in areas that will be called upon increasinglymore often by current andfuture clients, such as information technology.Public accountants are finding themselvesmore and more in the position of being InformationConsultants to their small business clients.The respective institutes (CICA and CGA)are portraying their members as computer (IT)professionals in national ads. This researchshould provide results that show if the message isgetting through to the public as wellas to the professional accountant.For the purposes of this paper, the term informationconsultants is broadly defined as“professionals who use their knowledge of inforniationtechnology to help individuals (i.e.clients/customers) obtain sufficient knowledge/skill levelin the use of an informationtechnology to become independentof further extensive professional aid in usingtheinformation technology”. This definition includesinformation centers, DP departments,computer consultants, and pjjlic accountants.The public accountant is often relied uponby the small business manager for help ininstalling computerized accountingservices to ensure that the system will meet theaccountant’s requirements as wellas the manager’s. This expectation arises from the publicperception of the accountant’s expertise with informationtechnology. Public accountants nowfind that some 95% of their auditclients have information technologyinstalled (Walker,1991). Often, however, accountantsare 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 publicand private humiliation that adverseheadlines tend to bring, as well as the subsequentlawsuits and loss of business, research isrequired that will aid the professionalaccountant in helping his client successfullyadopt anynew information technology.Despite the good reputation ofinformation consultants, failures still occur. Practicaladvice based on solid research,designed to minimize the risk of failure,would be verywelcome. Also, new types ofinformation technologies are continuallybeing developed.Inevitably, the new information technologywill find its way into business. Theskills to cope.3with the introduction of the information technology needto be defined in an attempt to avoidany trepidation on the part of the client, basedon past experience, that may otherwise occur.There is a general reluctance to adopt new information technology inthe publicaccounting profession (Batch et al, 1989) as well as in other professions(Newman, 1990). Ifthese information technology specialists are resistantto learning and adopting newerinformation technology, it should be no wonder that the informationtechnology specialistsexperience user resistance to the introduction of even basic information technology. Thisresearch should provide motivation for the information consultantto continue on the arduoustask of bringing his clients intothe 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 majorityof research in MISis carried out on large organizations (Attewell, 1989).The result is similar for studies on howinformation technology affects organizationsas well. It can be easy to fall into the trap ofthinking that results from these studies apply equally wellto small organizations. However, ithas been shown that small firms differ fromlarge firms in many areas, including job creationand growth which in turn affect many other organizational characteristics(Attewell, 1989).For example, research on the role ofinformation consultants, such as the Information Center(IC), is generally carried out on large firms (for a typical large firmstudy see Brancheau &Wetherbe, 1990). However, there are few (if any) IC’s or similarentities in small firms. Therehas been little empirical research thathas looked at the role of information consultants in theadoption of information technology in a smallbusiness setting.The role of the information consultant in the diffusion ofinnovations process will beexamined. For small business managers this is animportant issue as small firms usually lackthe resources to develop necessaryexpertise in-house. These businesses often lookto theirprofessional accountant for advice ontheir information requirements. For professionalaccountants this is also an importantissue as their associations are attemptingto transform4their members into information specialiststo 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 ACCOUNHNGSYSTEMThe Computerized Accounting System is the specific informationtechnology ofinterest to the accounting profession and smallbusiness in general. The ComputerizedAccounting System is a special subset ofthe Personal Work Station which Moore studied.The Personal Work Station as definedby Moore consists of a set of computerizedtools designed for an individual;is used on a microcomputer or terminal connectedto aminicomputer or mainframe; is accompaniedby appropriate software; and is used directly(hands on) (Moore & Benbasat, 1991). The PersonalWork Station is general and not functiondependent. A Personal Work Station canbe 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 theindividual.A Computerized Accounting System for thepurposes of this research is defined as aset of computerized tools for an individual, andusually consists of a personal ormicrocomputer with one or more softwarepackages, including an accounting programand/orother software such as a spreadsheet,database, word-processing, etc. in support of theaccounting function. A Computerized AccountingSystem is similar to the Personal WorkStation defined by Moore. The major differencesbetween a Computerized Accounting Systemand Personal Work Station are thatthe use of a Computerized AccountingSystem(hardware/software) is usually an organizationaldecision and a Computerized AccountingSystem supports the accountingfunction primarily.1.4 TOWARDS A SMALL BUSiNESS ORIENTATIONResearch into information technology, now entering its thirddecade, has primarilyfocused on large organizations. Although there are severalissues regarding whether or not itis necessary to study small businesses separately fromother businesses, the main issue iswhether the organizational factors foundin small firms are sufficiently similar to those oflarger firms. If the main factors of interest are commonacross firms then it is appropriate andeconomically prudent to limit research studies to large firms andextrapolate the results to allother firms, given the difficulty in obtaining results fromsmall firms. If these factors aredissimilar, then we as researchers have been omittinga significant group of organizations fromour studies and we cannot claim with confidencethat our results are generalizable across allfirms.This orientation towards big business is natural,as larger firms tend to operate incomplex conditions. Understanding the environmentaland internal factors that influence howa firm will behave is important to the enterprise andto society. This understanding isnecessary because large firms have high publicprofiles, are large employers, and make largecontributions to local economies, research institutes, andgovernments in the form of taxes ordonations. Large firms are properly viewedas being very important to our economy.Small businesses are also important to the economy.A study on small businesses inCanada, commissioned by the FederalBusiness Development Bank (FBDB) in 1986 andreleased in 1987, found some unexpected results.Small businesses (defined as firms withsales under $2 million and typically withless than 20 employees) accounted for 25% of ourGNP, 96% of all business organizations (over 700,000),created the greatest employmentopportunities for women and youngpeople (under 25 years old), had less of a wage gapbetween men and women, employed 32% of allworkers (excluding farm, professionals,fishing and commission sales people)and over the period 1978-1982 created over 52% of allnew jobs (FBDB, 1987). More recentdata confirms the impact of small firms on job creation,as a study commissioned by the CanadianOrganization of Small Business found smallbusinesses created over 98% of the newjobs in the period 1984-1987 (Small Business6Magazine, October 1989). The increasing importance of smallbusinesses 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 newjobs (Richmond Business, 1990). Similargrowth has occurred all across Canada duringthis 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 innovationsduring the period1970-1979 (Pavitt et al, 1989)and the portion of innovating small firms (under 200employees) has been increasing significantly overthe period 1945-1983 (Pavitt et a!, 1989).The importance of small firms to theeconomies of Western countries is obvious.The above statistics hide the sensitivityof small firms to economic fluctuations. Evenin boom times many small firms experiencea rocky road. The Canadian experience in theperiod 1978-82, for firms employing 5or less full time employees, indicated that for every100 net new jobs created: 52 were incurrently existing firms; 106 were for newly createdfirms which survived; and 58 were lostfor new firms that didn’t survive (FBDB, 1987). Dueto this sensitivity to the economic environment, smallerfirms 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 importantto public accountants, and viceversa. There is a special, symbiotic relationshipbetween these two groups. This relationship,while acknowledged, is not well understoodand varies from country to country. It appearsthat many small businesses in Canadarely on their public accountants for more than theiraccounting and tax knowledge (Goodson, 1990; Delenteet al, 1990; Hamilton, 1989), whilemost small firms in Australia still seekmainly year end accounting and tax servicesfrom theiraccountants (Holmes & Nicholls, 1989). A recent Canadianstudy on small firm’s relationshipwith their accountants found thatone of the reasons small firms initiallyengaged their1The researcher has encountered several smallfirms that have experienced most, if not all, of theaboveproblems through his own involvement in accountingpublic 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 computersystems 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 lossof 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 becomea real concern (Jenish, 1992;BYTE, August 1991; Rockburn, 1990; Kunz & Maingot, 1989). Whilemost larger firms haveinternal resources to help overcome these problems (in-house expertise, financialresources toacquire adequate information technology) most small firms remain at riskdue to their lack ofresources.Factors contributing to the problem of unmanaged informationtechnology includeignorance of the full potential of the information technologyby 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 impacton himself (Baronas & Louis, 1988);management ignorance of the skills the organization hasavailable for using the informationtechnology (Benson, 1983); and management reluctanceor inability to provide adequate usertraining (Buckler, 1990 and others). For large and small firmsthe information technology useris often unsophisticated because the technologyis new to the firm and personnel familiar withit would be relatively few (Lees & Lees, 1987). Tolearn to use the information technology theuser has the options of relying on informationconsultants (Melone & Bayer, 1990; Stieren,1990), other staff (Melon & Bayer, 1990;Nilakanta & Scamell, 1990; McFarlan&8McKenney, 1983), or on the user’s own abilities. The extentof 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 problemfaced by many small business managers isthat they attempt to manage information technologybased 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 tendsto impose on an organization(Miller, 1988) as is the case with the initial introductionof an information technology. Mostlarge firms have experienced these major changesseveral years (or decades) ago and will bemore familiar in dealing with change thantheir smaller counterparts. In large firms users oftenhave skilled resources to fall back onsuch 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 sourcesare finding it increasingly difficult to keepup (Geliman, 1991; Gotleib, 1990) with the resultthat large firms will turn to specialists(consultants) if necessary (Gotleib, 1990; Boynton& Zmud, 1987). Users in small businesseson the other hand have much fewer resourcesto fall back on (Willits, 1990; Delone, 1988;Lees & Lees, 1987). Often they must rely on externalskilled specialists, helpful friends, orthemselves (Lefebvre & Lefebvre, 1990; Gable, 1989;Delone, 1988; Lees, 1987). In manycases hiring the external information specialist is muchcheaper than hiring full time EDP staff(Arter, 1988) with the result that external informationconsultants are commonly used by smallfirms (Bracker & Pearson, 1985).For the small business the specialist is often theirprofessional advisor - their public accountant (Delenteet. al, 1990; Peat et al, 1984). Recentstudies show that in Canada there isa growing shortage of skilled information technologyspecialists (Buechert, 1992). While this shortageposes problems from businesses in general, itprovides an opportunity for public accountantsto fill this void. Partly in response to this trend,organizations such as the CICA have exhibitedplans to expand their involvement ininformation technology on a large scale(Brown, 1992).9It has been suggested that the reasonsa small firm seeks outside help for managinginformation technology are similar to those used for seeking outside help inbusiness planning(Gable, 1989). If this is true, thenthe professional accountant is the person to whom thebusiness manager will turn as the accountant often has providedthe business planning adviceinitially. However, success in providinga business plan doesnt ensure success regarding theadoption of information technology. The failures of informationsystems installed with thehelp of information consultants have been well documented inthe media. This is particularlytrue for accountants (e.g. see Babcock, 1986) and thefear of lawsuits over malpractice forproviding information systems or adviceis a real and growing threat (Dragich, 1989; Walton& Durham, 1988). While there is research to supportthe claim that external accountingservices help small firms to be successful (Bracker & Pearson, 1985), there arealso researchresults that claim using external informationconsultants, including accountants, provide lessthan satisfactory results for a small business (Hamilton,1989; Baker, 1987; Lees, 1987; Lees& Lees, 1987; Bracker & Pearson, 1985). Some ofthese studies indicated that highersatisfaction could be achieved if the consultant provideda 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 asa group. They are theones who force upon the industrial situation the concernwith numbers, with exchangeablemoney, with tangibles rather than intangibles, with exactness, withpredictability, withcontrol, with law and order generally, etc. ... Andy Kay [then presidentof the company]pointed out that the accountants have the lowest vocabulary scoresof any of theprofessional groups. I added that the psychiatrists thinkof them as being the mostobsessional of any group. From what I know of them, they alsoattract to the schools ofaccounting those who are number bound, those who are interestedin small details, thosewho are tradition bound, and the like.” [Maslow, 1965quoted by Davidson, 1991].A1D NOW“My own research ... found that members of professional accounting firmsare very bright,with an average intelligence levelat 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, independentand impulsive.” [Davidson &Dalby, 1991].From an organizational perspective, there is a growing realization thatinformation canbe considered as an asset (Frarnel, 1990; Ahituv, 1989), albeit an intangibleasset. Many firms(large and small) are turning to information technologydue to the increasingly complex andcompetitive business environment and the recenttechnological 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 accountantscan attest) that little is done to protect(Bradbard et al, 1990; Alavi & Weiss, 1986) or exploittheir data. A recent BYTE survey ofits readers (including large and small firms) foundthat 53% of respondents had suffered lossof critical data costing an average of$14,000 (BYTE, August 1991).The wide spread diffusion of information technologyhas left many firms open to theissue of security. Many small firmsappear to be ignorant of the necessity of information11technology security (Pendegraft et a!, 1987). For the small business it has beensuggested thatsecurity is even more important than for large firmsdue 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, particularlymicrocomputersoftware (Bradbard et al, 1990; Overbeyet 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 provideslarge 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) environmentora PULL (individual) environment.A PUSH environment exists when events externalto the user (or the firm) forceinformation technology on the user. Firms acquire information technologydue 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 necessaryto acquire informationtechnology due to his own work environment. Employees may acquireinformation technologyfor higher job satisfaction (Kraut et al, 1989; Pentland,1989; Millman & Hartwick, 1987).Additionally, non-IS employees may acquire information technologydue to frustration withthe IS department for delays in developing user requiredsystems (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 witha system that he developedhimself (Gremillion & Pyburn, 1983); but he is also responsible for theimplementation (Davis,1981).12Research of successful adoption of information technology has focused onmeasurableattributes associated with success. Over the past decade or so, the definition of success hasevolved from a one dimensional point of view (i.e. see studiesby 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, 1979and others). As a result of theincreasing knowledge on informationtechnology adoption processes, currently success isviewed as a relative term (Gallupe, 1989). In other words, success isdependent on how wellthere is a match between the user’sexpectation of what the information technology issupposed to accomplish, and what the information technology actuallydoes.User attitudes have increasingly been seen as animportant 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 dueto the increasedemphasis on theory based constructs such as attitudes(from the social and cognitivepsychology domain - for an overview of currentthought on attitudes, see Pratkanis et al,1989), where in the IS domain the concept of userattitude 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 adoptionofinformation technology, also called technology transfer(Bouldin, 1989). The study ofinformation technology using an adoption of innovationsapproach 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 whichwas 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 INFORMATIONTECHNOLOGYIt must be considered that there isnothing more difficult to carry out, nor more doubtful ofsuccess, nor more dangerous to handle, thanto 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 lukewarmnessarises partlyfrom fear of their adversaries, who havethe laws in their favor, and partly from theincredulity of mankind, who do not truly believe inanything 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 researchhas been the lack of anadequate theory of IS (Goodhue, 1986). There is considerable confusionon the issue of whata successful ny’brmation system is (Goodhue, 1986).The recent research on informationsystem attitudes and adoption of innovations has begunto clear up this confusion. The currentview of information system which incorporates these conceptshave been described by Boon& Pienaar (1989, pp. 122):“Technology is not an end in itself but merelya 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 informationtechnology, would result inappropriate and successful application of the technology’(emphasis added).The issue of knowing what the knowledge workersare doing is addressed by Moorein his study. Moore (1989) has developeda general model, the Diffusion of InformationTechnology model (see Figure 1 - Diffusion of InformationTechnology), that explains theadoption and use of information technologies byindividuals.This model was integrated from conceptscontained in the Diffusion of Innovationsmodel by Rogers (1983) (see Figure 2) and theTheory of Reasoned Action by Ajzen andFishbein (1980) (see Figure 3), to explainthe adoption of information technology byindividuals. In developing this model, Moore has attemptedto overcome the previously notedweaknesses (ie. lack of theory, measuring information systemsuccess) in research in thisarea. The Moore model is the mostcomprehensive and theory backed work to dateoninformation technology diffusion and adoption.143.1 DIFFUSION OF INNOVATIONSThe Diffusion of Innovations work by Rogers is well supportedby research. Rogers’Dffusion ofInnovations model is used to explain the rate of adoptionof innovations (Rogers,1983), which included five perceived attributesof innovations; type of innovation-decision(individual or collective decision); communication channels (mediaor interpersonal contact);nature of the social system (social norms, interconnectednessof the communication network);and extent of change agent’s (product champion or opinionleader) 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 areall conceptually distinct fromeach other (Relative Advantage - the degreeto which an innovation is perceived as beingbetter than its precursor; Compatibility - thedegree to which an innovation is perceived asbeing consistent with the existing values,needs, and past experiences of potential adopters;Complexity - the degree to which aninnovation is perceived as being difficultto use;Trialability - the degree to whichan innovation may be experimented with before adoption;and Observahility - the degree to which the results ofan innovation are observable to others).Moore added an additional two attributes (Image - thedegree to which use of an innovation isperceived to enhance one’s image orstatus 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 perceivedas being voluntary, or of free will),was also added by Moore and the variable Complexitywas 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 summaryof a complete list of Perceived Characteristics ofInnovativeness variables and Voluntarinessdefinitions.153.2 THE THEORY OF REASONED ACTIONIn the Theory ofReasoned Action, which is well supportedby research studies, Ajzen& Fishbein identified the relationship between intentions, beliefs,attitudes, and behaviours(Ajzen & Fishbein, 1980). The basic premise isthat 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 attitudetowards the behaviour) and hisSubjective Norms (the individual’s perception ofwhat other people think about the behaviour)determine the individual’s behavioural intention. Theseattitudes and subjective norms are afunction of the individual’s belieJ. The basic premisesof the Theory of Reasoned Action areillustrated in Figure 3.3.3 DIFFUSION OF iNFORMATIONTECHNOLOGYThe link between the Diffusion of Innovations model and theTheory of ReasonedAction can be seen in Figure 1. The synthesized DiffusionofIn!hrmationTechnology modeldeveloped by Moore can be described as follows (Moore& Benbasat, 1990,pp.3):“Innovations diffuse because of the cumulative decisionsof individuals to adopt them. Thus,it is not the potential adopters’ perceptions ofthe innovation itself, but rather theirperceptions of using the innovation thatare key to whether an innovation diffuses.”To test the Diffusion of Information Technology model,a questionnaire wasdeveloped and administered in a cross sectionalstudy involving individuals in six Canadianorganizations. The questionnaire results supported all eight PerceivedCharacteristics ofInnovation variables as being factors in explaining the diffusion ofPersonal Work Stations,which was the particular innovation being investigated(Moore & Benbasat, 1990).3.4 SIGNIFICANCE OF THE DIFFUSION OF INFORMATIONTECHNOLOGYMODELTheDiffusion ofInformation Technology modelattempts to predict, explain andinfluence individual behaviourtowards the adoption of information technology. TheDiffusion of Information Technology model is alsodesigned to be a general model (Moore,1989). As a general model, the DiffiTsionof Information Technology model should apply toa16specific information technology other than the Personal WorkStation, such as ComputerizedAccounting Systems. The Diffusion of InformationTechnology model should also applyequally well to small businesses and large businesses.These observations about the Diffusion of Information Technologymodel arise froman inspection of the theory on which themodel is based. Because the Diffusion of InformationTechnology model is based on theoretical models, it will contain the characteristicsof theunderlying models. An important characteristic of the Theory of ReasonedAction is the abilityto predict, explain and influence individual behaviour (Ajzen& Fishbein, 1980). The Theoryof Reasoned Action is generalizable and isapplicable to all people. The Diffusion ofInnovations model focuses on the adoption of innovations. The Diffisionof Innovationsmodel should be generalizable across all innovations, includinginformation technology.It is important to examine whether the Diffusion of Information Technology modelissufficiently robust to include small businesses aspart of the population it encompasses.Research models that are of help to small firms are fewand far between. This model couldprovide a means for explaining whya particular innovation, such as installation of acomputerized accounting system, succeeds in one firm andnot another. It could also be usedfor predicting if the innovation is likely to succeed, before significanttime and resources arecommitted to a project, by determining if the firm has an adequate mixof similar attitudes andbeliefs as those found for the successful adopters.17CHAPTER 4: TECHNOLOGY TRANSFER - RESEARCHQUESTIONS“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 beingused 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 Technologymodelhas the ability to predict and explain individual behaviour towardsthe adoption ofinformation technology. An underlying reason for this currentstudy 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 publicaccountant(information consultant) as well as to their small business clients. For the smallbusinessmanager or information technology specialist, it is expectedthat by understanding the factorsthat lead to successful adoption of information technologya systematic approach can bedeveloped to influence individual behaviour to adopt new information technology in thefuture.Besides attempting to extend the Diffusion of Information Technology modelto thesmall business domain, this study attemptedto 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, coupledwith the Extentof Change Agent’s Promotion Efforts section (Figure 2)and the Connnunications Networksection of Fishbein & Ajzen’s model (Figure 3) essentially representthe same concept -information gathering/exchange (for convenience the term Communication Channels willbeused in the remainder of this thesis). Moore’s research did not investigatethis area. However,the importance of communications channels in theadoption process should not beunderestimated, as it has been pointed out that18“before a business unit can adopt anduse a technology, members of the business unit mustbecome knowledgeable of the technology andbe able to propose ideas for its use. Thisawareness results from communication behaviors ... wherebya ‘technology provider’ familiarwith the technology interacts witha potential ‘technology user’ not familiar with thetechnology” (Lind & Zmud, 1990,pg. 4).For a small business, the technology provider, likely anexternal consultant (Gable,1989), is often the accountant (Goodson,1990; Hamilton, 1989). The role of externalconsultants as information sources on informationtechnology has not been well established inthe Diffusion of Information Technologyliterature (Gable, 1989). Unlike most externalconsultants, accountants are often consideredto be an integral part of their client’smanagement team (Delente et al., 1990; Goodson, 1990). For manysmall business managers,the opinion of their accountants are highlyregarded and persuasive (Goodson, 1990).4.1 RESEARCH HYPOTHESESThe Diffusion of Information Technology modelprovides a means to determine justwhat the characteristics of a successful interaction between theuser and a specific informationtechnology are. Interactions between users and an informationtechnology are registered bymeans of a questionnaire that Moore hasdeveloped and validated. Moore’s questionnaire didnot focus on the role of external informationconsultants, probably as a result of his focus onlarge business adoption of information technology wherethe necessary expertise would beavailable in-house through the Information Centre orsimilar department. As small firms donot have a similar body of in-house informationexpertise, the role of external informationconsultant becomes more important. This specific item mayprovide an important researcharea for small firms.As validating Moore’s results regarding the Diffusionof Information Technologymodel is one goal of this study, a summary ofthe Moore hypotheses (modified to reflect theComputerized Accounting System) is provided below.Hj: One attitude towards using a Computerized AccountingSystem will influence one’sinnovativeness with respect to Computerized AccountingSystem usage.19H2: Relative Advantage will have a contribution morethan any other PerceivedCharacteristics of Innovation on one’s attitude towards adoptingComputerizedAccounting Systems.H3: Computer Avoidance will have a contribution less than anyother PerceivedCharacteristics of Innovation on one’s attitude towardsadopting ConiputerizedAccounting Systems.114: The Subjective Norm willinfluence one’s innovativeness with respect toComputerized Accounting System usage.H5. The Subjective Norni will influence one’s attitude toward adoptingthe ComputerizedAccounting System.H6: Voluntariness is negatively relatedto one’s innovativeness with respect toComputerized Accounting System usage.H7: Voluntariness i’ill be negativelyrelated to one’s attitude towards usingComputerized Accounting System.An important research question for small business managersarises concerning the rolethat Support groups, especially external informationconsultants such as accountingprofessionals, play in the process ofinformation technology diffusion. The researchhypotheses related to this question are developed in thefollowing paragraphs.It has been shown, in Chapter 1, thatsmall and medium firms rely on externalconsultants more than large firms. Because small firmshave little in-house expertise ininformation technology, especially for an importantinformation technology such as aComputerized Accounting System, the involvementof an external source of information andguidance should contribute to thesuccess of the introduction and adoption of theComputerized Accounting System.H8. The involvement of a Support Group i’ilicontribute to a successful adoption ofComputerized Accounting Systems.20As a Support Group is made up of different components, it follows that eachof thesecomponents should contribute to a successful Computerized AccountingSystem. 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 contributeto a successful ComputerizedAccounting System.H1]•The involvement of an external Accountant will contribute toa successfulComputerized Accounting System.Hp. The involvementqf an external Consultant will contribute to a successfulComputerized Accounting System.An investigation of the Moore, Fishbein & Ajzen, and Rogers modelsindicate that thepresence of a communications channel will influence other areasof the Diffusion ofInformation Technology model as wellas Innovativeness. In this study, communicationschannels is represented by the Support Group. The Fishbein& Azjen model (Figure 3) showsdirect links from Communications Network to SubjectiveNorm and Attitude. These linkssuggest the following two hypotheses:H13.The involvement of a Support Group will have a positive influenceon SubjectiveNorm.H14.The involvement ofa Support Group will have a positive influenceon Attitude.Also, while not explicit in the Fishbein & Ajzen model, it is possiblethat the PerceivedCharacteristics of Innovation variables may alsobe influenced by the communicationschannels. This link is suggested from a review of theadoption process indicated in Roger’sStages of the Innovation Decision Processmodel (Figure 4), where theKnot ‘ledge/Persuasion cycle (incorporatingthe communications channels) impacts the21Decision cycle (which incorporates the behavioural intention, whichare 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. SupportGroup) perceiveor present information technology. Fromthese 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 questionnairesused in the study will bediscussed. Reliability results for both the pilot study and the finalstudy 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 theonly feasible way aresearcher can obtain sufficient volume of data in an economical manner. It is controversialasthe method is susceptible to a number ofsources of error that could render any resultssuspect. A good questionnaire must therefore strikea balance between its length andcomplexity, presenting to respondents a form that isn’t intimidating, while obtainingdata thatis reliable and valid.Moore spent a considerable amount of time establishing the reliabilityand validity ofhis questionnaire. The changes to the Diffusion of Information Technologyquestionnairediscussed in the next section were of a typeto potentially call into question its reliability butnot its validity. The changes made were generally cosmetic, substitutingComputerizedAccounting System for Personal Work Station and cleaningup terminology to be moreconsistent with a small business environment. These changes were notexpected 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 isa reliability issue. As a resultreliability issues will be dealt with in more detail than validityissues.5.1.1 RELIABILITYReliability is defined as “the degree to which the results of measurementare free oferror” (Stone, 1978). Generally, there are two componentsto any measurement, a “true”component and an “error” component. A reliable measurementinstrument 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 reliablewhen independent but comparable measures forthe item of interest provide similar results (Churchill,1979).The appropriate level of reliability isa factor of the goals of the researcher andpublished criteria for the type of research being done. Reliabilitynumbers range from 0 to 1and are usually presented asdecimal fractions, where the higher the fraction the better thereliability. The general rule of thumb fora 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 thatis in the early stage of theory testing (Nunnally, 1978)and is also an acceptable itt/c of thumb levelfor PLS analysis (Barclay et. al., 1991). It shouldbe stressed that .70 is the lower bound for an acceptable reliabilitylevel.5.1.2 VALIDITYValidity is defined as ‘the degree to which a measureactually measures what itpurports to” (Nunnally, 1967,pp.75). In other words, the differences observed aretruedifferences for the characteristics being investigated and nota result of some other source(Churchill, 1979). There are several items comprising validitywhich are summarized inAppendix I-C. It should be noted that not all ofthese factors may be an important issue withany given questionnaire, but they should at leastbe considered upon preparation.Validity is not considered to be a problem in this researchas the questionnaire usedwas previously validated by Moore. Changes made to the questionnairefor this study did notfundamentaly alter what the questions were meantto measure. For example, questions meantto measure Image still measured Image, only the Imagebeing measured was for aComputerized Accounting System (Modified questionU-6: Using a CAS improves my imagewithin the organization) and not a Personal WorkStation (Original question U-6: Using aPWS improves my image within the organization ). Thissubstitution of CAS for PWSoccured for all 39 questions.245.1.3 QUESTIONNAIRE SELECTIONThe research issues being investigated indicated that two separatequestionnaires wererequired. One questionnaire to test the Diffusion of InformationTechnology model andsimultaneously gather data on users (clients) information technologyinformation sources, thesecond questionnaire to elicit data from accountants.The development of each of these questionnaires is discussed in thefollowing sections.SECTION B: QUESTIONNAIRE DESIGN - PILOTSTUDYAll references to questions in this section refer to the Pilot Studyquestionnaire.5.2 CLIENT/DIFFUSION OF INFORMATION TECHNOLOGYQUESTIONNAIRE5.2.1 PERCEIVED CHARACTERISTICSOF INNOVATIONSOne goal of this research study isto 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 validitycriteria (see Moore& Benbasat, 1991). It was determined that redeveloping an alternatequestionnaire would beredundant, fitile, and not contribute to a cumulative discipline.Therefore, Moore’squestionnaire was adopted with some minor modificationswhich are discussed below.In Moore’s study, the measurement of Perceived Characteristics ofInnovations wasobtained through the use of an interval scale (ranging from1 to 7) consisting of 50 questions.These 50 questions were used to measurethe 9 Perceived Characteristics of Innovationvariables that Moore considered integralto the Diffusion of Information Technology model.Based on subsequent analysis of the Diffusion ofInformation Technology model usingLISREL (Linear Structural RelationsModel), Moore was able to determine that only 8Perceived Characteristics of Innovation variables weresignificant factors. Moore also wasable to determine that the PerceivedCharacteristics of Innovation questions could betrimmed down to 38 from 50 without significantly affectingthe results (Moore & Benbasat,1991). In this paper, references toMoore’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 Systemto focus the study on theinformation technology Computerized Accounting System. Altering the questionnaireintroduced the risk that the instrument no longer measured what it was supposedto measure.The modified questionnaire was tested bya pilot study on a sample of small businesses andcompared to Moore’s results to establish thatthe modifications did not fundamentally alter thereliability of the questionnaire in relationto Moore’s Diffusion of Information Technologymodel. The major risk inherent in this approach is if the pilotstudy does not producestatistically similar results, it will be difficult to determine if the results arefrom the changes tothe instrument or from difference betweenlarge and small firms. Due to this potentialproblem, an additional pilot study was contemplatedto be carried out on a relativelyunmodified version of Moore’s questionnaire. The only modificationto this questionnairewould be the substitution of Computerized Accounting Systemfor Personal Work Station.The results from these two pilot studies would becompared to each other and to Moore’sresults to ensure that the overall integrity of the questionnaire wasnot damaged. Anydifferences between the two pilot studies could be attributedto changes in the questionnaire,while differences between the pilot studies andMoore’s study could be attributed todifferences between large and small firms.As it turned out the results for the pilot study were statistically similarenough toMoore’s findings to dispense with the second pilotstudy. 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 .60and higher wereconsidered as acceptable as reliability scores tend to increase withlarger sample sizes(Nunnally, 1978). This pilot study had all Perceived Characteristicsof Innovation variablesexcept Visibilty (.28) and image (.59) reporting scores above.60 (see Table 1). ThePerceived Characteristics of Innovation variable Visibilityhad a reliability score much lowerthan the minimum acceptable and was examinedmore closely. Upon reviewing Moore’s26rationale for using a subset of his original questionnaire it was decided that Visibilitycould beimproved by adding an additional question tothe questionnaire, bringing the Diffusion ofInformation Technology subset of the questionnaire upto 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 pilotstudy questionnaire would be used in the actual study. ThePerceived Characteristics of Innovation questions were labeledU-i to U-39 for ComputerizedAccounting System users and N-i to N-39for 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, measurementof adoption is difficult andsurrogate items are often used, such as system usage. After reviewingthe literature it becameevident that usage was commonly measured by using one or two items. Thisis disturbing asreliability is impossible to establish based ona 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 variablesthis is not as major a drawback as it is for independentvariables. As validating Moore’s model is an important partof this study, it was determined touse similar usage measures as those used by Moore.Adoption is measured by determining the usage of the informationtechnology (thePersonal Work Station). The usage measuresare called Innovativeness. There are threeaspects of innovative behaviour that were measured in hisstudy, these are AdoptiveInnovativeness - degree to which an individual isrelatively early in adopting an innovation,Use Innovativeness - degree to which an individualputs an innovation to use within a givenuse domain, and Implementation Innovativeness - degreeto which an individual who hasadopted the innovation uses it to solve novel problems, or ina new use domain (Moore,1989). The Innovativeness measures anddefinitions 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 questionnairetomeasure 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 developedbyconverting 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 notberequired.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 useof 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 singlenumber by summing the hours of eachfunction used, in order to make the two measures similar in nature. A reliabilityscore of .80(see Table 1) was achieved. While the reliability score was acceptable, a reviewof thequestionnaires indicated that there were problems that some respondents hadin answeringthese questions consistently. The basic problem was that theprocess 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 weeklyusage scale (B-4). Often th totalsresulting from adding hours reported in question B-8a wereconsiderably higher than the28overall number reported in question B-4. It was reasoned that individuals areprobably morelikely to accurately remember how much they use individual ComputerizedAccountingSystem 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 weekdo 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 thefollowing functions?...).It wasalso decided not to develop a replacement question for the item droppedas the bestalternative would have been to obtain actual usage figures. This alternative wasnot feasible asthe researcher had no access to the respondents’ place of work to measureusage due toconfidentiality. Judging fiom the researcher’s own experience working withsmall businesses,it was also unlikely such records existed in small firmseither.Frequency of Use in the Pilot study was measuredby 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 usinga CAS for...)measuredComputerized Accounting System usage in an overall manner. The third questionmeasuredfrequency 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 1-3times per month, 4=About once perweek, 5=About 2 to 4 times per week, 6=About once per day, 7=More thanonce per day).After reviewing the responses to these questions it was decided notto use the questionmeasuring use by individual Computerized AccountingSystem 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 (indicatedby a “4” on the scale for B-7) yet reportusing a CAS more than once per day (indicated by a “7” on the scale forB-5 or B-i 1). Thesedifferent reponses could arise due to the timingof 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 includedtodetermine 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 regularlydo you now use a CAS?) inthe present tense and B-i 1 (I have been usinga CAS for...)in the past tense. The inclusionof the same 7 point ranking scale (discussed above) should have caused respondentsto answerthe questions similarily. Dropping one of these questions (B-il) wasconsidered; however itwas decided to retain this question to see if similar results would occur in the fullstudy.Use InnovativenessThe Pilot study included four questions designed to measure system usage, calledUseInnovativeness 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 manyhours 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 calculatedby taking the averagenumber of functions used for each question. The reliability score was found tobe low, .45 (seeTable 1). Further reliability calculations were performed ona reduced subset of questions andit was found that by dropping the question did the firm usea mainframe or micro (B-6) thescore improved to .77. The Pilot study indicated that all respondentsonly usedmicrocomputers, which made sense for a small business environment. It wasdecided to dropB-6 from the final questionnaire for the above reasons.305.2.3 CLIENT COMPUTERIZED ACCOUNTINGSYSTEM SUPPORTGeneralAn important part of this study was to examine the role of thesupport group inComputerized Accounting System adoption. A series of questionswere asked regarding themakeup of the support group and the role they play in helping the clientwith the use of theclient’s Computerized Accounting System.Current SupportFor Computerized Accounting System users, there were fivequestions designed tomeasure the composition of the support group. Thesequestions were currently receivecontinuing support (B-13), ...iast JO source(s,) of CASsupport (B-14), ...where to go fneedComputerized Accounting System help (B-i 5), ...ratingof satisfaction with support group(B-17), and ...rating qf effectiveness of supportgroup (B-19). Because each questionmeasured different aspects of support, the results were transformedto a binary measure foreach support group (i=support, O=no support). Thistreatment resulted in a reliability score of.94.Based on follow up conversations with some respondents itappeared that B-i4 wasconfusing. A reliability measure of .93 was obtainedon the other four questions. As therewere several comments about the length of the questionnaire, itwas decided to drop B-i4from the final questionnaire, resulting ina shorter questionnaire and only a minor reduction onreliability.SECTION C: FINAL SURVEYS - SCALE RELIABILITIES5.4 GENERALAlthough full details of the full study are providedin the next chapter, the reliabilityscores for the various measures are summarizedin Table 2 found at the end of this chapter.For the Perceived Characteristics ofInnovation variables, all 75 respondents are included. Forthe scales measuring Innovativenessand 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 inthe 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 tosee 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 thetwoquestions 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 usedby Moore. Thisapproach was considered the most appropriate asMooresquestionnaire design was based onthe Total Design Method which had been designed andtested by Diliman (1978). Thismethod was reported to have resulted in very high response rates. There weresome 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 colouredpages separating themajor sections of the questionnaire. A covering letterfrom UBC was also attached to thefront of the questionnaire. Moore had chosen this format in orderto improve the overallappearance of the questionnaire in an attempt to make it appear moreprofessional and worthyof a good response (Moore, 1989).After presenting a copy of the questionnaireto the Partners in one of the accountingfirms participating in the study, and discussing thepossible distribution of a similarquestionnaire to their client base, it was determinedthat some changes would have to bemade. The Partners considered the questionnairetoo long in appearance and that many oftheir clients would simply not fill itout, even though the covering letter stated that not all ofthe questionnaire was to be filledout. It was decided to split the questionnaire into two parts,one part for Computerized Accounting Systemusers and one for non-ComputerizedAccounting System users. This approach wasused for a number of reasons. First, it wasexpected that there would be differences between ComputerizedAccounting System users andnon-Computerized Accounting System users.Separating the questionnaire based on thisconsideration was consistent with the objectivesof the research. Secondly, the researcherwould not have direct access to the client base of participating Accountingfirms. Because thePartners or someone knowledgeable in each Accountingfirm were to do the distribution totheir clients, they would know if the intended recipientwas a user or non-user and distributethe appropriate questionnaire. Finally,the questionnaire each potential respondent was toreceive would be approximately half the sizeas originally designed which should enhancewillingness to participate. These factors made thesplitting 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 questionssothat they were horizontally oriented and not vertically oriented. Thisapproach was takenbecause of early feedback received on the apparent length of the questionnaire, even aftersplitting it into two parts. The Partnersused 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 howto answer that specificquestion. At the end of each section encouragement was providedto 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 purposeof the research as well as theconfidentiality of the replies received. The second letter was prepared on theletterhead of theparticipating accounting firm and explained that the firm was not sponsoringthe study butbelieved the results would be useful. Encouragementto participate and confidentiality werestressed in this letter also.Both of these covering letters (see Appendix IT-B) were designedafter extensiveconsultation with Partners from different accounting firms and with thethesis supervisor. Itwas emphasized to the Partners that the wording of the second coveringletter (the accountingfirm letter) was a suggestion only and that they were freeto make changes as they chose. Therationale behind this approach was to win Partnersupport 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 encouragethem 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, thedata integritychecks performed, the statistical analysis and results will be discussed insome detail. Beforeproceeding with this discussion a brief summary of the goals forthis study are presented.The prime objective of this study is to establish the role that public accountants playinthe introduction and adoption of information technology in small businesses.This type ofinformation is vital, as several research studies have shown that publicaccountants are notgetting the message out, to their members and to thesmall business community, thataccountants are skilled information technology specialists (see Hamilton,1989; Batch, 1989).As part of this analysis, the Diffusion of Information Technology model developedby Moorewill be examined in a Small Business setting, using Computerized AccountingSystem as theinformation technology of interest. This will be dine in orderto evaluate whether (i) theDffusion of Information Technology model is generalizableacross firm size and (ii) differentinformation technologies than those examined by Moore whendeveloping this model. Recallthat the major differences betweena Computerized Accounting System and Personal WorkStation are that the use of a Computerized AccountingSystem is usually an organizationaldecision, and a Computerized Accounting System supportsthe accounting function primarily,whereas the use of a Personal Work Station is often a personal decision anda 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 practiceaccounting firms. Most smallbusinesses use a public accountant for tax purposes orfor preparation of financial statements.35However, not all firms decide to use a public accountant. Theremay 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 southwesternB.C. werecontacted to elicit interest in the study as these accounting firms werethe most likely to havelarge numbers of small business clients. Dueto the method of selecting the sample certainbiases may have been introduced that mayaffect the generalizability of the results.A potential regional bias may restrict thegeneralizability 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 informationtechnology, in southwestern B.C.,relative to the rest of Canada. Because the Diffusionof Information Technology modelmeasures individuals’ attitudes, any bias wouldaffect the results. Other regional biases mayexist at the firm level as southwestern B.C. mayhave a larger than average number of smallbusinesses concentrated in specific industries. Theseindustries could have their own peculiarrate of adoption, independent of anindividual’s propensity to adopt. Also, there could be abias between small businesses in large citiesand 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 towardsintroducing information technology to theirclients.These potential biases inherent in this study should notgreatly affect the objectives ofthis study (i.e. generalizability). One objective of thisstudy is to provide a predictiveinstrument that can be used to help small firms successfullyintroduce 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 anyother province; therefore regional biases should not2Although there is no reliable information on the numberof firms that don’t use public accounting firms, thisnumber is generally accepted to be small. Firms that fallinto this category include inactive or nearly inactivecompanies. The inclusion of these firms in the study wouldcause misleading results as IT is not likely to beapriority with low activity firms. Public accounting firms arenot likely to be interested in inactive businesseseither, as these firms are not likely to become clients norpay their accounting fees.36be an issue. Also, members of public accounting firms (CA and CGA) must all takeCanadawide exams as well as continuingProfessional Development courses. All of theseprofessionals will have a similar educational exposureto information technology which shouldhelp reduce regional differences amongst public accounting firmslevel of knowledge aboutComputerized Accounting System.Data collection involved the use of survey instruments, withdata analysis performedon self-reported data. Directed interviews were consideredas a multi-method approach isconsidered appropriate for generating more assurance onthe validity of the findings.However, the multi-method approach provedto not be feasible and the directed interviewapproach was abandoned.6.2.2 PROBLEMS ENCOUNTEREDNo different than any other research project, this one had itsshare of problems fromthe onset. Due to the volume and variety of problems encountered itwas considered justifiableto devote a seperate section discussing these problems and their impact on thestudy.The sample size of directed interviews could not be increased beyond fiveor six due tothe promise of confidentiality made to all participants,especially clients. At one point,arrangements were made with selected accounting firmpersonnel for follow up interviews, butconditions in the working world interfered with the followup 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 theyreturned it was considered thattoo long a time period had elapsed toput confidence in their responses. Additionally, someparticipating accounting firms (along with participating personnel) decidedto back out of theircommitments. It was too late to recruit new participantsas the remaining accounting firmshad already distributed questionnairesto their personnel and clients.Coupled with the problems of holidays and attrition of participatingaccounting firms,an untimely mail strike hampereddata collection efforts severely. It appears that manyquestionnaires that were delivered to clients duringthis time were either not filled out or37mailed in. As no facility to follow up on non respondents was available, theselost respondentscould not be recovered. Also, by the timethe strike was over, the participating accountingfirms had entered the start of their busy season and distributionof questionnaires was givenlow priority. Regaining the initial enthusiasm exhibitedby participating accounting firmsproved to be difficult. Data collection became a tedious taskas researcher phone calls wouldoften not be returned and promised actionswould not be delivered.6.2.3 RESPONSE RATESA total of 283 questionnaires were distributedto accounting firms and other contactpeople for distribution. Of these, 120 were returnedby contacts who had decided to endparticipation in the survey, resulting in a total of 163 questionnairesbeing distributed tovarious clients. A total of 56 usable questionnaires were returned (nobreakdown is availableon how many client firms responded) resulting ina response rate of 34%. This response ratewas lower than expected. A higher response rate was expectedas the contact people hadagreed to solicit agreement to participate fromtheir clients before distributing thequestionnaires. Based on follow up discussions with some of thesecontacts it appears thatsome firms sent the questionnairesout without consulting with the clients, while otherscontacted the clients first and then sentout the questionnaires. It also seems that some clientsdid not fill out the questionnaires even though theyhad told their contact that they would.Additionally, some contacts may not have distributedall of the questionnaires allocated tothem. This lower than expected response rate resultedin a change in approach to analysing theDiffusion of Information Technology modelby using FLS instead of LISREL. It was decidedthat the 19 responses from the pilotstudy would be included in the data analysis in order tohave enough questionnaires to use PLS. All resultsreported for the final survey, includingreliability results, included the pilot questionnaires.The pilot questionnaires were included asthere were only minor differences in the twoquestionnaires for the research issues in question.A convenience sample of clients of B.C. publicaccounting firms was used due tovarious constraints. Face to face contact withindividuals 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 database.Questionnaires were sent to both Computerized Accounting System users and non-Computerized Accounting System users. It is importantto include non-ComputerizedAccounting System users as it has been pointed out that one should not leaveout the “zerovalue” or control group when exploring the effects of an intervention(Attewell, 1989). Theintervention being controlled for in thiscase, the Computerized Accounting System, isconsistent with Moore’s approach. However, it is difficult to control for theintervention ofthe public accountant by acquiring data from firms whodo not use accountants for anyreason. As discussed earlier, these firms may not exist or wouldbe 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. Thislimitation to the scope ofgeneralizability is not severe, as it has been previously mentioned that most firmsuse PublicAccountants.An initial sample size goal of 200-3 00 responses wasset in order to accommodate theobjective of testing Moore’s Diffusion of Information Technology model usingLISREL.However, as stated earlier, several unexpected problems arose that dramaticallyreduced thenumber of questionnaires that could be expected tobe returned. As a result of these datacollection problems it was decided to use PLS instead ofLISREL as PLS is widely consideredan acceptable alternative to LISREL (Barclayet al, 1991).6.3 CLIENT FIRM’S SURVEYThe results from the full study indicated general reliability support forthe scales usedto describe the variables in Moore’s Diffusion of Information Technologymodel. All of thePerceived Characteristics of Innovating variables had reliability scoresat the .70 level andhigher except for Result Demonstrability, which droppedfrom a reliability score of .62 in thepilot study (Table 1) to a reliability score of .43in 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 comparableto 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 reflecta difference between sampledomains (firm size) or result from use of a subset (3 questions) of the 4questions used tooriginally define this Perceived Characteristics of Innovating variable. A closer lookat theresponses of non-Computerized Accounting System users indicates that the majority of theseindividuals perform non-accounting tasks. This visual analysis is substantiatedby Mann-Whitney tests, which confirm that there is a statistical differencebetween ComputerizedAccounting System users and non-users for Result Demonstrability(discussed later in thischapter, also see Table 7(b)). Additional reliability figures were obtainedby obtaining abreakdown between Computerized Accounting System users and non-users. ResultDemonstrability reliability improves to . 71 (Table 3) whenComputerized Accounting Systemuser data only is used. A graph of Result Demonstrability non-Computerized AccountingSystem users was generated to determine why no reliability figure couldbe calculated for thisvariable. Inspection of this graph (Figure 5) shows that the three scalesused to measureResult Demonstrability (U 15, U23, and U33) received very inconsistentresponses. Normally,a graph with scales that are highly reliable would have the scores for eachscale moving in thesame direction for each response. The graph in Figure 5 shows thatthe scores for each scalemove in opposite directions for each response, in mostcases.Further inspection of Table 3 indicates that noother variable showed an obvioussimilar variability in responses by non-ComputerizedAccounting System users, althoughVoluntariness (alpha=.43) indicated that non-usersdid appear to have some difficulty with40this measure also.It is not clear why non-users would record responses thatwere as inconsistent as thoseobserved for Result Demonstrability (and possibly Voluntariness).6.4 CONDITIOMNG THE DATA6.4.1 GENERALBefore the data could be analysed, severalsteps were required to ensure that theresults would be meaningful. These include checking thedata for accuracy, dealing withmissing data, and dealing with outliers. Theseare discussed below.6.4.2 ACCURACY OF INPUT DATAThe data was originally input intoa spreadsheet program by the researcher, who thenrechecked large sections of each questionnaire. A printedcopy of the input was thencompared to the original questionnaireby two independent persons (the data checkers).Differences between the two were notedby each data checker on the print-out. Theresearcher then compared the items identifiedas being incorrect to the correspondingquestionnaire and made appropriate correctionsto the spreadsheet. Very few errors weredetected by the data checkers. With a relatively smallsample it is unlikely that there would bemany undetected errors. After these errorchecking 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 notpossible to adopt one globalapproach in treating the data for missingvalues. Questionnaires that were missing data forlarge sections of the questionnaire were notused at all (there were 2 of these). Multi-itemscales, such as those used to definea Perceived Characteristics of Innovating variable, wouldhave the scale mean inserted if only one itemwas 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 arethose 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 Characteristicsof Innovating questions weremarked neutral (4 on a 7 point interval scale) indicating the respondent had nottaken time tounderstand or read the questions. Descriptive statistics were also reviewedto determine ifthere were any other cases of outliers. Except for the non-usei responsesto ResultDemonstrability (discussed in a previous section), no others were found.A search for skewness is usually done to determine if the data distribution is normalaswell as whether there may be more outliers. Moore found that his data wasgenerally skewedbut that transformations were not practical due to the design of the questionnaire(Moore,1989). Transforming his data did not provide results differentfrom the original data (Moore,1989). Given the small sample size and the relatively large impactremoving anyquestionnaires could have, whether they were outliers or not, it wasdetermined 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 betweentwovariables show linearity (straight line) and homoscedasticity (variabilityin scores areapproximately equivalent for all values of the two variables). Both of these,revealed by thepresence of an oval shaped scattergram, are requiredassumptions for multivariate regression.42Scattergrams were produced for the variables of interest and no significant violations of thesetwo assumptions were detected. Thus the data appeared tobe 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 descriptivestatisticswere generated for the research variables including themean, standard deviation, andmaximum and minimum reported values. These are summarized in Table 6. Acomparison ofComputerized Accounting System users vs. non-Computerized AccountingSystem users isprovided in Table 7(a) and 7(b), including Mann-WhitneyU test results. These results will bediscussed in detail later in this chapter. The Mann-WhitneyU test is used to determine if thereare differences between Computerized AccountingSystem users and non-ComputerizedAccounting System users. The Mann-Whitney (M-W) test isused in order to avoid relyingupon the t-test and its assumptions (normal distribution). The M-W test is a conservativetest.This test was also used by Moore as part of hisdata analysis. Regression analysis results onthe variables of interest are provided in Table 8 throughTable 11. Regression results arediscussed in the following sections.General comparisons will be made to Moore’s study, based on whetherthe resultssupport the hypothesis that is being tested. Specificnumerical comparisons will be made toMoore’s study where the results from this study differ from Moore’s. A summaryof resultsfrom hypothesis testing for this study can be foundin Table 13(a), and for Moore’s study inTable 13(b).References to question numbers will refer to the finalquestionnaire (Appendix Il-Al)unless otherwise noted.436.6 DEMOGRAPHICSDemographic data is summarized in Table 5 with Adjusted Frequency figuresused(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 theaccounting area. The remaining48.5% were distributed throughout other departments, including Administration(19%) andOther (29%) - “Other” consisted of areas not falling into Accounting orAdministration. 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 remaining23% 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 educationbeyondhigh school (18%) while another 10% received some training froma trade school. Theremaining 72% received some College/University education, including 8% withPostgraduatedegrees. [From Moore: High School= 12%; TradeS chool= 19%; College/University69%;Postgraduate 18%]. It appears that for small/medium sized firms that the levelof 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 togeneralize to any large extent as themethods of gathering the above information differ and maycause the perceived differencesnoted (ie. Moore had respondents gathered into a roomto fill out the questionnaire, somepotential respondents may have had to stay behind to “run theshop” and these may have beenthe younger employees). However, there appearto be definite differences in the age groups of44employees working in smaller firms.The SEX profile is also in sharp contrastto Moores study. This study had 33% malerespondents and 67% female, while Moore had 63% males and 37% femalesrespond. Again,definite differences in smaller firms. The smaller sample size in thisstudy may contribute tosome of this difference.The overall demographic profile of this study indicates sharpdifferences from Moor&ssurvey. Respondents are generally younger, more likely tobe female, and have less formaleducation than in larger firms. These findings generally support earlierstudies (discussed inChapter 2) on demographic characteristics of people employed in smallto medium sized firms.6.7 ATTITUDE TOWARDS INNOVATTNGThe dependent variable Attitude was generated froma four item semantic differentialscale (good-bad; harmful-beneficial; wise-foolish; andnegative-positive) in response to thequestion Overall, my using a C’AS in my job is (B-i). Variousdescriptive statistics weregathered on Attitude. These statistics are based on all 75questionnaires. On a seven pointscale (lmost positive, 7=most pessimistic) anoverall average of 2.2 (Table 6(a)) indicatesthat attitudes are generally quite positive towards theComputerized 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) betweenComputerized Accounting System users(mean 1.8) and non-users (mean = 3.3). TheseM-W results provide a method ofdetermining to what extent the overall mean of 2.2 is influencedby users and non-users. Thedescriptive statistics results in general, and M-W results for usersspecifically, indicate supportfor Hj [One attitude towards i/sing Computerized AccountingSystems will influence one’sinnovativeness with respect to ComputerizedAccounting System usagej. The claim forsupport of Hj is based on the assertion thatthe 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-innovativepeople), 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 exceptfor Voluntariness (3.1) had amean score of 4 (neutral) or higher (Table 6(a)). The most positive PerceivedCharacteristicsof 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-usersat thep<.O5 level (Table 7(b)) except Ease of Use (.14). All of Moore’s PerceivedCharacteristics 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 opinionsaboutComputerized Accounting Systems. There is support for H2LfRelativeAdvantage will have acontribution more than any other Perceived Characteristicsof Innovating on one’s attitudetowards adopting Computerized Accounting SystemsJbased 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 providedas 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 1to 7) in a manner similar to themethod used for Perceived Characteristics of Innovating, with higher scoresindicating amore positive response. As discussed previously in this section, Voluntarinesshad 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 (Table7(b)) showsthat Computerized Accounting System users means (2.72) are significantly lowerthan non-Computerized Accounting System users means (4.13), indicatingsupport for H646[Voluntariness is negatively related to one’s innovativeness with respectto ComputerizedAccounting System usagej. This finding is the sameas Moore’s.6.9 SUBJECTIVE NORMSValues for Subjective Norm scores were calculated by multiplying the NormativeBelief (ranging from 1 to 7) by the Motivationto Comply (ranging from -3 to +3). The rangeof scores could vary from -21 to +21. The mean scores reportedin this study (Table 6(a)) aremixed and range near zero which is neutral (-2.8 to 1.5). Moore’sranged from 1.7 to 7.3.Based on the M-W tests (Table 7(b)), the only significant differencesbetweenComputerized Accounting System users and non-Computerized Accounting Systemusers, atp.O5, arise from Senior Management (.0i9) and Subordinates (.003).In general, H4 [TheSubjective Norm it//I injinence oiies innovativeness with respectto Coniputerized AccountingSystem usage] is not supported. This differs from Moore’sstudy where H4 was supported (allof Moore’s Subjective Norm variables showed significant differences between usersand non-users). These M-W results are quite different fromMoore’s and again seem to indicatedifferences between large and small firms. In smaller offices, employeesare more likely tointeract with people in other functional areas (cross-pollination of ideas) andthe influence ofreference groups would be more uniform. Large firmswould likely have less uniform opinionsabout reference groups due to the lack of interaction with people in otherfunctional 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, andnumber of functions used. Descriptivestatistics for each of these measures can be foundin Table 6(b). Because Innovativenessinformation was only gathered for Computerized AccountingSystem users, M-W tests couldnot be run on Innovativeness variables and N’A appearsfor the boxes where statistics are notapplicable in Table 7(b). As a result of this data gatheringapproach, 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 offirstCAS use (B3) and (‘AS use by function, in months (B-6(b)). An average of 56 months (Table7(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 readilyapparent. Traditionally it has beenheld that larger firms adopt information technology before smaller firms. Perhapstheparticular information technology of interest, Computerized Accounting Systems, diffuseearlier than the other Personal Work Station items that Moore examined. Itshould be notedthat no statistical tests were done to determine if the values for both studieswere significantlydifferent. If such tests were run it is possible that they could show no statistical differenceinadoption periods between the two studies.Hours of use of Computerized Accounting System per week is calculatedby 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 AccountingSystem userand an overall average was calculated from the total hours calculated,for all ComputerizedAccounting System users. An average of 21.6 hours per week (Table7(b)) is more than the15.9 hours reported by Moore. This average indicatesthat 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 significantlydifferent.Frequency qf Computerized Accounting System useis calculated in two ways. In thefirst method, a general frequency of use is calculatedby taking an average of the results forthe two questions which ask how long the GAS user hasbeen using the CAS (B-4 and B-9) asthese two items ask the same question. Both items consist ofa seven point scale, and anaverage of 6 (about once per day, Table 7(b)) was calculated. In the secondmethod,frequency values for a question that asked forfrequency of use by function (B-5), wereobtained by summing the coded values from a sevenpoint scale (1=not at all, 4about onceper week, 7=more than once per day), for eachof the eight applications. Ranges of values for48an individual Computerized Accounting System user couldvary from53(didn’t use anyComputerized Accounting System applications) to 56 (used all eight applications morethanonce per day). An average value of 27.8 functions was calculated using the secondmethod.This was lower than Moore’s result of 35; however Moore’s Personal WorkStation listed 12functions to the 8 Computerized Accounting System items identified in thisstudy. As noted inthe footnote, the average value reported in thisstudy may be understated as well. Nostatistical tests were performed to determine if the values for both studies weresignificantlydifferent.The number qf functions used is calculatedby averaging the responses to thequestions asking the frequency(?fuse byjiinction (B-5), how many hours per week each CASfunction was used (B-6(a)), and how many months eachCAS function ivas used (B-6(b)).Theoretically, if one of these questions received an answer then all three questionsshouldhave had an answer. Each question was codeda zero (0) for no response or a one (1) for aresponse. By averaging the responses to each function forthe three questions, effects frommissing data was likely to be minimized. An average of 4.5Computerized Accounting Systemfunctions (Table 7(b)) are used, compared to 5.9 Personal Work Station functionsfor Moore.There is a higher proportion of Computerized Accounting System functionsused (4.5/8) thanPersonal Work Station functions used (5.9/12). It is not clear if this differenceis due to theselection of functions. The Computerized AccountingSystem functions are basically a subsetof the Personal Work Station functions and the most popularfunctions may have been chosen.Alternatively, the nature of the task, accounting, maycontribute to heavier use of informationtechnology. No statistical tests were performed to determineif 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. creatinga potential ‘missing data” problem. Themethod chosen to record the responses resulted in a “MinimumScore” of 4 instead of the theoretical 8discussed. This approach may result in understated Frequency ofuse 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 Systemusers 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 Table7(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=zlessthan 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 involvementof otherEmployees will contribute to a .s’uccessfui Computerized Accounting SystemJ, Hjj [Theinvolvement of an external Accountant will contribute to a successfiuiComputerizedAccounting 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. Theresults arediscussed in this section. PLS results are discussed in the next section.506.13 THE EFFECT OF PERCEIVED CHARACTERISTICSOF INNOVATIVNESS ANDVOLUNTARINESS ON ATTITUIEThe initial regression model analysed was the seven Perceived CharacteristicsofInnovating variables and Voluntariness on Attitude. The procedure followed paralleled that ofMoore (1989). A STEPWISE regression was run, with the F-value probabilityset at p<.O5 forentry andp>.10 for removal of a variable once in the equation. Followingthis regression, asecond regression was run where all variables were forced into the equationin the same orderas the STEPWISE regression. The end result of the forced entry procedureis 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 andan adjustedR2=.749 (Table 8), indicating that the Perceived Characteristics ofInnovating variables aresignificant in the formation on Attitude towards using computerizedaccounting systems.. Theregression results indicate that the various Perceived Characteristicsof Innovating variableshave different effects on Attitude. Only Re/alive Advantage is highlysignificant (p.OO,R2=.73). Visibility (p=.O4, incremental R2=03 [=76-73]) is marginallysignificant. None ofthe other variables contribute to R2 in any meaningful way. These resultsare summarized inTable 8 where part I lists the results for the forced stepregression 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 Innovationon one’s attitude trniards adoptingComputerized Accounting Systems], as Relative Advantage’scontribution to R2=.73 while thefull equation had anR2=.78. This result was similarto Moore’s.There was no support for H7 [Voluntariness will be negatively relatedto one’sattitude towards using Computerized AccountingSystems], 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 regressionwasto 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 andSeniorManagement. Individual scores for each reference group could range from-21 to 21 (this wasdiscussed in Chapter 5). The composite measure calculated for Subjective Normwill 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 maynot be able todifferentiate between these two possible explanations. A larger sample sizewould indicatesignificance, if there was a trend in the scores in the same direction for thereference groups,or if one (or more) reference group was clearly dominant and the remaininggroup scoreswere near neutral. A larger sample size would not help if two reference groupswith oppositescores were dominant.Sample selection, and possibly firm size may be a factor as larger firms tendto haveestablished cultures and prevailing opinions on information technology use.Moore’s sample52consisted of individuals from six large firms, whereas this study contained responsesfrommany more (smaller) firms, possibly 20 or more. Moore’srespondents would likely show amore dominant culture affecting Attitude as, at most, thereare six different cultures andpossibly less. A dominant culture could emphasize one reference group orcombination overanother. It would be difficult to determine if Moore’s significant results for SubjectiveNormare due to the small sample size of large firms or the presenceof 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 forSubjective Normwere found in this study, there is a likelihood that there is no dominantcorporate 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 correctlyrejected (oraccepted).A further regression was run with Computerized Accounting System Support,Perceived Characteristics of Innovating, Vohintariness and SubjectiveNorm on Attitude.This equation is similar to the previous regression with the addition ofComputerizedAccounting System Support. The purpose of this regression wasto determine the influence ofComputerized Accounting System Support on Attitude in orderto examine the extensionsmade to the Diffusion of Information Technology model. This regressionwas run withComputerized Accounting System Support included as a composite score(individual supportgroups consisted of Friend, Employee, Accountant, and Consultant). Theregression resultsindicated that the composite Computerized AccountingSystem Support score was notsignificant in the regression equation (p=.3 1). These results are summarizedon Table 9, PCLSN & SUPPORT column. H14 [The involvement of a Support Group wi/i havea positiveinfluence on Attitude], was not supported.The regression results on Support clearly indicate that this variable has noeffect onAttitude. This result is unexpected and maybe 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 tobe the case(Moore, 1989). The unexpected lack of significance of Support on Attitude may bea 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 ina subsequent regression run toexamine the influence of this variable on Innovativeness. A different regression on individualPerceived characteristics of Innovating variables and Subjective Norm variableswas alsorun. This second regression omitted the intervening variables AttItudeand the overallSubjective Nonii measure. Once again Support was added in a subsequent runto measure itsimpact on Innovativeness.6.14.2 ATTITUDE, SUBJECTIVE NORM AND VOLUNTARINESS ONINNOVATIVENES SFour regressions were run, one for each of the dependent Innovativenessmeasures(Number of Functions Used, Frequency of Use, MonthsSince Adopted, and Hours of UsePer Week). The dependent variables for each regression runwere Attitude, Subjective Norm54and Voluntariness. The results of the regressions, summarizedin 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 toComputerized Accounting System usage],is not supported. This is consistent with Moore’s findings.Voluntariness is not significant in any of the regressions, thus rejectingH7[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 thebetas 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 toComputerizedAccounting System usagej, is not supported. Moore’s resultssupported H6.The adjusted R2 values for the four regression equations range froma low of .193 to ahigh of .223. The variations in R2 indicate that the independent variablescapture differentdegrees of the variance in the different forms of innovativeness. This low range ofadjusted 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 wererun including Computerized AccountingSystem Support as an independent variable in the aboyeregression 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 aComputerizedAccounting Systenij.With the addition of Computerized Accounting System Support, the influence of theother independent variables on the various Innovativeness variables changed. Withtheaddition of the Support Group variable (compared to the regressions without theSupportGroup variable), Voluntariness became significant for Hours (p=.O5 vsp=. 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 vsp.OO)].Thus, theinclusion of Computerized Accounting System Support has weakened the importance ofAttitude on Innovativeness (ie. weaker support for Hj). This effect indicatesthatComputerized Accounting System Support has an influence in the Diffusion of InformationTechnology model.6.14.3 PERCEIVED CHARACTERISTICS OF INNOVATING, SUBJECTIVENORM,ANT) VOLUNTARINESS ON INNOVATIVENESSIndividual Perceived Characteristics of Innovating variables and individual SubjectiveNorm variables were regressed on the dependent Innovativeness variables in orderto measurethe magnitude of their direct effects on adoptive behaviour. As each individual PerceivedCharacteristics of Innovating and Subjective Norm measure were expectedto have differentinfluences on the dependent variables, STEPWISE regression was used with the probabilitytoenter a variable into the equation set atp.O5.As in the previous regression runs,Computerized Accounting System Support was added as an independent variabletosubsequent regression runs. In this case, the individual Support Group measures wereusedinstead of the composite scale in order to examine the influence eachscale 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 variablesweresignificant only for the Innovativeness variable Number Of FunctionsUsed (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).4It 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 thatSupport Group has asignificant effect on Innovativeness. The regression results support H8 [The involvement of aSupport Group will contribute to asuccess!;’,!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 consistentto 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” thandid the compositescales designed to represent “Attitude”. When Moore used LISRELto 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 therewasany effect between these variables. The results aresummarized 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 thatcaused difficultyinterpreting results for the Di/flision of Information Technologymodel for both studies theAttitude variables generated may not have captured the constructAttitude. It also appearedthat the construct Subjective Norm (Subjective Norm) maynot have been appropriatelyspecified, because the regression analysis indicatedthat individual Subjective Norm variablesaccounted for more variance on the dependent variables thandid Subjective Norm itselfAdditionally, the construct Innovativeness could notbe generated using normal regressiontechniques due to the differences in the scales of the Innovativenessvariables (eg. months vshours vs functions used). These factors indicated thatan alternate method to regressionanalysis would assist in constructing Subjective Norm, Attitudeand Innovativeness from theirindividual components. One such method is known alternativelyas causal modeling (Barclayet a!, 1991; Bagozzi, 1982), structural equation modeling(Fournell et a!, 1982) or pathmodeling (Wold, 1985). For convenience the term pathmodeling will be used throughout thispaper.60Path modeling utilizes second generation5multivariate analysis techniques in ordertoobtain statistical information that cannot generallybe 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 andmanifest variables)and psychometrics (latent variables) (Wold, 1985), all path models have in common thetraitsof 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 strengthsand weaknesses foranalysing models. The choice of which program to use depends on thestage of theorydevelopment being tested and the goals of the researcher. PLS is generallyused in the earlystages of theory development while LISI?EL is better suited to modelsbased on welldeveloped theory. LISREL is based on assumptions of multivariate normality indata whereasPLS requires no such assumptions. LISREL requires large sample sizes while PLScan be usedwith much smaller sample sizes (Barclay et al, 1991).Afier analysing the various characteristics of this study it was determined thatthe useof PLS would be most appropriate. This study is examining the Di/,/iisionof InformationTechnology model first developedby 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 hasyet to be firmly established.Additionally, preliminary results from regression analysis indicated that multivariatenormality5The term “second generation” is used to denote the useof more sophisticated mathematical models and statisticalcomputer programs. A second generation multivariate technique mustmeet four requirements (Fournell, C.,ASecond Generation of Multivariate Analysis Methods, 1982. cited in Barclayet. a!.. 1991): the technique must 1)analyze multiple criterion and predictorvariables: 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 proceduresare special cases of second generation techniques.Multiple regression. multiple discriminant analysis. analysisof variance and covariance, and principal componentsanalysis are all special cases of canonical correlation ... whichitself 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 notbe 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 (outerdesign matrix); the secondequation describing the path connecting LVs to each other (inner design matrix).TheDiffusionof Information Technology model, with iWVs’ and LVs identified, canbe seen inFigure 6. This model design is comparable with the LISREL modelused 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 theorysupportingthe model.(1) LV Subjective Norm has as its indicators MVs Supervisors, Peers, SeniorIvianagement,Subordinates, Friends and Perceived Voluntariness. The MV, Friends, whichwasomitted from the Moore’s LISREL analysis, was included in the PLS analysisin order tofully analyse the Df/iision(?fInformation Technology model The MV indicatorPerceived Vohintariness has been includedbecause the regression results indicated astrong interaction with Subjective Norm. The inclusion of this All7 is consistentwith 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. TheseMVs are actuallyPerceived Characteristics ofInnovatingindicators 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 dependon the underlyingpremise of the model as set out by the model builder (Lohmoller, 1984). In thecurrent modelthe scales are standardized and path loadings between LVc and MVs’ representthe relativeimportance of the composite scale score to the LV. The path loadingsbetween LVs can alsorange from zero to one. The higher the loading the more important therelationship/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, becausethe 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 forniativeconstruct (or LatentVariable in PLS terminology) is an LV that isa construction, or composite, of its MVs(Barclay et. al., 1991; Lohmoller, 1984). [Reflective construct’s on the otherhand are L Vs withlviVs that are products or reflect the underlying construct ofthe LV (Barclay et. al., 1991;Lohmoller,1984)]. An endogenous construct is an LV that is predictedby 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 sizeversusparameters to estimate. PLS can deal with this situationbecause ... the iterative algorithmbehind PLS estimates parameters in only small subsets ofa model during any given iteration.The subset estimation process consists of simple and multipleregressions so that the samplerequired is that which would support the most complex multipleregression encountered.”(Barclay et al, 1991,pp.15).The determination of whether MVs are formative or reflective in regardsto theirassociated LV depends on the researchers prior experience with the model and theunderstanding of the real world situation being studied. Ifthe constructs are not welldeveloped then the IvIVs for that construct are considered formative. For thepurposes of FLSanalysis of the Dffusion of Information Technologymodel (Figure 6) only the SubjectiveNorm MVs will be treated as formativeindicators, while Subjective Norm andCommunication Network will be treatedas 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 exogenousL Vs, while Attitudeand Innovativeness are endogenous LVs.The largest number of formative indicators is five(Subjective Norm) while the endogenous LV with thelargest 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 NormMVs).In Figure 7, Voluntariness and Communications hannei areexogenous L Vs, whileSubjective Norm, Attitude and innovativeness are endogenousL Vs. The largest number offormative indicators is five (Subjective Norm) while theendogenous LV with the largestnumber of predictor LVs is Innovativeness withfour (Communications Channel, SubjectiveNorm, Voluntariness, and Attitude). This would indicatea 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,= 1SSEn/ 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 isa test of predictive relevance.Mathematically the formula is similar to Communality except that it appliesto 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 Innovatingvariables(R2=. 12, H2=.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 R2results. 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). Thevarious 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 thesehypotheses change dependingon which model is used.Hj. One’s attitude towards using Computerized Accounting System i’ill influence one’sinnovativeness t’ith re5pect toComputerized Accounting Systeni usage. Thehypothesis indicates that a positive coefficient is requiredto increase innovativeness.For the original model (.624) and the extended model (.440) thepath coefficient ispositive thusHiis supportedH2: Relative Advantage wi/i have a contribution more thanany other PerceivedCharacteristics oJ Innovating on one’s attitude towards adopting ‘omputerizedAccounting Systems. Path loading for MV Relative Advantage onLV Attitude is thelargest for the original model (.9141) and the extended model(.9180), supportingH2.H3: C’omputer Avoidance i’iii have a contribution lessthan any other PerceivedCharacteristics of Innovating on one’s attitude towardsadopting ComputerizedAccounting Systems. This hypothesis is not explored in thisstudy.114: The Subjective Norm will influence oneinnovativeness with respect toComputerized Accounting System usage. For theoriginal model (-.06 1) and for theextended model (.056) the path coefficient havea very small loading value indicatingno support for H4,H5: The Subjective Norm i’iii influenceone’s attitude toward adopting the ComputerizedAccounting System. The original model (.122) andextended model (.053) have verysmall path coefficients, which indicate no support forH5.68“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 Technologymodel (see Figure7).‘‘8The 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 contributeto the overallSupport Group (Communications Channel) LV path loading.H9. The involvement ofa Friend i ill contribute to a success/lu ComputerizedAccountingSstem. As the path coefficient is small (.0694), this suggests that the hypothesis isnot supported.Hj.The involvement of other Employees will contribute to a successfulComputerizedAccounting 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 contributeto a successfulComputerized Accounting System. As the path coefficient is smalland 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 onPerceivedcharacteristics ofInnovating 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, provideda marginally (possiblynot statistically different) better fit indicators than the original Diffusion of InformationTechnology model. However, the better indicator scores did not appearto 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 includedacomparison of the descriptive statistics between users and non-users of computerizedaccounting system adopters. Next, regression analysis was performed on thedata to examinethe effect of various independent variables on the dependent variables. Finally,path analysiswas used to examine the theoretically derived Diffusion of InformationTechnology modeldeveloped by Moore and compare this model to other versions of this modelto determinewhich model had the best fit to the data.6.18 SUMIVIARY OF DESCRIPTIVE STATISTICSWhile there were significant differences between Computerized AccountingSystemusers and non-users on several of the variables, there were fairly uniform Subjective Normsbyall respondents. Overall, 71% of the sample were identifiedas Computerized AccountingSystem users, which indicates that non-users have a large impact on the overallresults. Theaverage time elapsed since initial adoption is just under fiveyears. Computerized AccountingSystems are used fairly often, with over 4 computerized accounting systemfunctions beingused for 22 hours per week.6.19 SUMMARY OF HYPOTHESES TESTTNGHj. One attitude towards using a Computerized Accounting 5steni will influenceoneinnovatii’eness i i/h respect to Computerized AccountingSystem usage.This hypothesis was supported by descriptive statistics, not supported by regressionanalysis, and partially supported by PLS. Regression analysis indicated that whileAttitude wassignificant in the adoption process for all Innovativeness variables, all of thebetas were alsonegative. PLS results indicated that the loadings were negativein value for the standard modeland positive for the extended model. The confusing Attituderesults (Table 10(b)) may havebeen an artifact of the scales used to measure Attitude.Substituting the PerceivedCharacteristics ofInnovating variables for Altituderesulted 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, suchas accounting firms, play in theDffusion ofInformation Technology process?2. Is the DfJiision ofInformation Technology modela general model?Before the first question could be answered, the second question had tobe 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 wereapplied. Theseincluded analysing the differences between computerized accounting systemusers 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 modela general model?Results showed overall support for the general model. The roleof informationconsultants was not very significant when applied to the general modelbut did show someeffect on individual components of the model. The answer, therefore, isa qualified ‘yes”.Based on the results of hypotheses testing for H1to H7 (excluding H3 which was not tested),no individual hypothesis was fully supported across all three statisticaltests applied (H1received some support using all three methods). However, each hypothesis received eitherpartial support to definite support from at least one of thetests.The regression results indicated that a larger sample sizemay have obtained moresignificant results for some of the variables. What is clear from the results isthat 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 allof the hypotheses. In this studymost of the variables were not significant and at best, moderatesupport 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 ConsultantsandPersonnel have both a direct and an indirectaffect 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 Attitudewhich has a direct affect on adoption(Innovativeness).The role of the Public Accountant is significant indirectly on Innovativenessthroughits influence on the construct Attitude. The Accountantwas 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 abilityto predict ifan information technology will be adopted for a given organization. They willalso be able torecommend to clients a methodology to maximize thesuccess 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 questionnairecaptures PerceivedCharacteristics of innovating variables adequately, but does not capture theconstructAttitude. Although the development of suitable Attitudescales would normally be arecommendation, the use of path analysis programs like PLSto indirectly synthesize thisvariable suggests that further scale development for Altitude may notbe warranted.The inclusion of Communications Channel (ie. SupportGroup) as an extension to theDffusion of Information Technology model is an attemptto improve the robustness of this77model. Statistical analysis show that this extension does reveal some interactionsin 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 thata 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 witha largersample. It was difficult to directly compare results with those obtained by Mooreas he had amuch larger sample size (600) and nearly everything was significant for hisstudy. 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 confidentialitytorespondents, resulted in a loss of control of sample selection. The sample shouldbe 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 andnon-users78for several categories. The relatively large proportion ofnon-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 attemptsatimproving reliablity were made (asking the same question more than once), thegeneralproblems 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 fromthisstudy, 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 shouldbe noted that suchinferences and extrapolations are made at the risk of over-interpreting theresults 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 isa strong association betweenthe presence of information consultants and the successfuladoption of a computerizedaccounting systems, many small businesses do notrely on this support group to help themwith new information technology. It appears that computerized accounting systemusers whohave used a computerized accounting system fora 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 leadsto long and versatile computerizedaccounting system use, or if the experience gained dueto the passage of time and/or heavyuse has convinced users to seek outside help. If the lattercase is true, then PublicAccountants have a lot of work to do to getthe 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 extendedDiffusionofInformation Technology model does provide some insight into this process.TABLES80ocH C H H tn (1)C) mIt•ij -IjH HC HCr11 — r11H Hrn > H > > Z HC)C)—CcH—tTlcC —z 0 CD 0 0CD 0—H<tncmCo-rn1rn>cHr-(IDt>c,)>ci>C)r’iHH<tn<Cl)— m‘Hci)z(I)(/) HZ H >> C) pm H .-<z— H*CD Cl) Cl) CD CD 0 -t CDCD -t (ID 0 Cl) CD Cl, C) C) C) -t C) Li. -tC) -t C) 0 0 -q, -t CD C,) 0 Cl) C) C,)H 4-C)LILiLi004..-.—(ID— H C H O H(ID H2H C C4-Li.LI.L.)004-C)4-11)2CZ000000LC(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 BACKGROUNDOF SURVEY RESPONDENTSRelative AdjustedNumber Frequency FrequencyDEPARTMENT OF EMPLOYMENTAdministration13 17.3%1 8.6%Accounting / Finance37 49.3% 52.9%Other20 26.7% 28.5%Missing5 6.7%Total75 100.0%100.0%ORGANIZATION LEVELExecutive15 20.0% 20.8%—Middle Management13 17.3%18.1%Supervisory11 14.7% 15.3%Professional12 16.0% 16.7%Technical4 5.3% 5.6%Clerical/Support15 20.0% 20.8%Other2 2.7% 2.7%Missing3 4.0%Total75 100.0% 100.0%EDUCATIONSome High School3 4.(>% 4.0%High School Graduate10 13.3%13.3%Some Technical College4 5.3% 5.3%Technical College Graduate3 4.0% 4.0%Some Community College7 9.4% 9.4%Community College Graduate6 8.0% 8.0%Some University14 18.7% 18.7%University Graduate22 29.3% 29.3%Postgraduate6 8.0% 8.0%Missing0 0.0%Total75 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 2736.0% 38.0%40to49ycarsold 13 17.4%18.3%50 years old and older4 5.3% 5.7%Missing 4 5.3%Total 75 100.0%100.0%SEXMale26 34.7% 34.7%Female 49 65.3%65.3%Missing 0 0.0Y0Total 75100.0% 100.0%OTHER Minimum MaximumAverage Firm Size (Sales) $500k-<$250k >$ 10,000k$L000kAverage Firm Size (Full Time Employees) 2692Avg. Accounting Staff (Full Time Employees) 31285TABLE 6(a)SURVEY VARIABLES - DESCRTPTWESTATISTICSMAXIMUM MINIMUM#SCALE MEAN STANDARDREPORTED REPORTEDITEMS SCORE DEVIATION SCORESCOREPERCEIVED CHARACTERISTICS(Scale Range: 1 to 7)Compatibility4 5.427 1.577 7.0001.000EaseofUse 65.118 1.003 7.0002.167Image4 3.977 1.366 7.0001.000Relative Advantage 85.503 1.560 7.0001.000Result Demonstrability3 5.187 1.1797.000 2.667Trialability5 4.128 1.426 7.0001.000Visibility 54.856 1.484 7.0001.000Voluntariness 43.130 1.480 5.7501.000ATTITUDE 42.200 1.214 6.2501.000(Scale Range: I to 7)SUBJECTIVE NORMS(Scale Range: -21 to 21)Friends1 -1.067 — 6.00321.000 -18.000Peers1 1.467 5.512 — 21.000-8.000Supervisors1 .547 5.194 21.000-8.000Senior Management1 -2.320 5.403 21.000-15.000Subordnatcs1 -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 27.7928.725 49.000 4.000(Scale range: 4 to 56)Months Since First Use 255.955 27.560 120.500 2.000(Scale range: 1 to 199)Hours of Use per Week 121.584 12.606 40.000 3.000(Scale range: ito 40)Number of Functions Used 3 4.3971.790 7.667 1.000(Scale_range:_0_to_8)CAS SUPPORT(Scale range: I to 7)Personnel from Firm 3 2.6291.150 5.333 1.000Friend 32.157 .993 5.000 1.000Accountant 3 2.3081.025 4.667 1.000Consultant S2.277 1.031 4.333 1.000-—C)C-CDC)CD—raCI)—)— C C — C z Crfj 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 ANDTESTS FOR DIFFERENCES (M-WTESTS)NON- U-TESTUSERS USERS Z-SCORE SIGNIFPERCEIVED CHARACTERISTICS(M-W)Compatibility6.10 3.80 -4.86 .0000Ease of Use5.25 4.81 -1.47 .1420Image4.35 3.08 -3,45 .0006Relative Advantage 6.183.87 -4.81 .0000Result Demonstrability 5.674.02 -5.59 .0000Trialability4.33 3.65 -1.9() .0575Visibility5.31 3.76 -3.25.0011Voluntariness2.72 4.13 -3.74 .0002ATTITUDE 1.763.25 -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 55.96N/A N/A N/AHours of use per week21.58 N/A N/A N/AFrequency of use - generalonce/day N/A N/AN/AFrequency of use - detail 27.79N/A N/A N/ANumber of functions used4.50 N/A N/A N/ACAS SUPPORTPersonnel from Firm2.63 N/A N/AN/AFriend2.16 N/A N/A .N/AAccounting Firm2.31 N/A N/AN/AConsultant2.28 N/A N/AN/A89TABLE 8REGRESSION RESULTSPERCEIVED CHARACTERISTICSAND VOLUNTARINESS ON ATTITUDEI. SUMMARY OF STEPPED FORCEDENTRY OF VARIABLESII. STA ‘ISTICS FOR VARIABLESIN TI-IF FINAL EOI ATIONSTEP VARIABLE INBETA IN R2F (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 .000IVARIABLEBETA 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 ExplainedAdjusted R2 .74990TABLE 9REGRESSION RESULTSPCI’S, VOLUNTARINESS,SN AND SUPPORT ON ATTITUDE(IUatiOII PCI Only PCI andSN PCI,SN & SUPPORTBeta Weights BetaSig. F Beta Sig.F Beta1Sig. FRelative Advantage -.632 .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 ExplainedR2 =776R2 =780 R2 =.783Adj R2 =749Adj R2 =750 Adj R2 =75091TABLE 10 (a)REGRESSION RESULTSATTITUDE, SN, AND VOLUNTARINESSON INNOVATIVENESSVARIABLESDEPENDENT INDEPENDENTBeta Sig. F Adj. FF 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 INDEPENDENTBeta 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 .3099.260 .0000HOURS OF USE Voluntariness -.2 14 .054PER WEEK Attitude -.182.110Subjective Norm -.084 .366Support .492 .000 .42514.669 .000093TABLE 11 (a)REGRESSION RESULTSPCI AND SUBJECTIVENORMS ON INNOVATIVENESSEQUATION 1: DEPENDENTVARIABLE - NUMBER OFFUNCTIONS USEDEntry Independent VariableFinal Beta Sig. FAd. R2 FStep1 Result Demonstrability.447 .0002 SN Peers-.297 .0013 SN Subordinates.268 .0044 Relative Advantage.217 .049 .50420Weakly Significant:Ease of Use -.186.061EQUATION 2: DEPENDENTVARIABLE - FREQUENCY OFUSEEntry Independent VariableFinal Beta Sig. F Ad. R2FStepI Relative Advantage403 .0002 Result Demonstrability.365 .001 461 33Weakly Significant:SN Subordinate .158072Ease of Use-.172 .091EQUATION 3: DEPENDENTVARIABLE - MONTHS SINCE CASFIRST ADOPTEDEntry Independent VariableFinal Bela Sig. FAdj. R2 FStepI Result Demonstrability.597 .000 .34740Weakly Significant:Relative Advantage .22-I.058Voluntariness -. 179.065Coinpatibility .210.078EQUAT ON 4: DEPENDENT VARIABLE- HOURS OF USE PER WEEKEntry Independent VariableFinal Beta Sig. F Adj.R2 FStepI Result Demonstrability445 .0252 Visibility.268 .013 .3622294TABLE 11(h)REGRESSION RESULTSPCI, SUBJECTIVE NORMS, & SUPPORTON INNOVATIVENESSEQUATION 1: DEPENDENT VARIABLE - NUMBEROF FUNCTIONS USEDEntry Independent VariableFinal 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 - FREQUENCYOF USEEntry Independent Variable Final BelaSig. 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 - MONTHSSINCE CAS FIRST ADOPTEDEntry Independent Variable FinalBeta Sig. F Adj. R2 FStep1 Result Demonstrability.497 .0002 Consultant.221 .035 .378 23Weakly Significant:Voluntariness -. 169 .076EQUATION 4: DEPENDENT VARIABLE - HOURS OF USEPER WEEKEntry Independent VariableFinal Beta Sig. F Adj. R2 FStep1 SUPP Consultant.436 .0002 Result Demonstrability .312 .0013 Voluntariness—.225 .010 .510 2795TABLE 12(a)REGRESSION RESULTSCAS SUPPORT ON OTHER DEPENDENTVARIABLESEQUATION 1: DEPENDENT VARIABLE- SUBJECTIVE NORM (COMPOSITE)Dep. Variable Independent VariableFinal Beta Sig. F Adj. R2FSNc No Significant Variables---EQUATION 2: DEPENDENT VARIABLE -SUBJECTiVE NORM (COMPONENTS) +VOLUNTARINESSDep. Variable Independent VariableFiiial 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 VariableFinal 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 OTHERDEPENDENT VARIABLESEQUATION 3: DEPENDENT VARIABLE- INNOVATIVE (USE) VARIABLESDep. Variable IndependentVariable FinalBeta Sig. F Ad.R2 FFrequency SUPP Consultant.424 .000SUPP Personnel.481 .000 .60658*********Functions SUPP Consultant.445 .00011SUPP Personnel.383 .000 .50338* ** * *** * *Hours SUPP Consultant.522 .000SUPP Personnel.181 .092 .38424*********Months SUPP Consultant.327 .008SUPP Personnel.234 .053 .218 11EQUATION 5: DEPENDENTVARIABLE - VOLUNTARINESSDep. Variable IndependentVariable FinalBeta Sig. FMi.R2 FVoluntariness SUPP Personnel-. 267 .020 .059697TABLE 13 (a)—SUMMARY RESULTS OFHYPOTHESIS TESTINGHYPOTHESESADOPTERS VS. REGRESSIONPLSNON-ADOPTERS ANALYSIS ANALYSISHi: ATTITUDE -> INNOVATIVENESSSUPPORTED NOT SUPPORTEDSUPPORTEDH2: RELATIVE ADV> OTHER PCISUPPORTED SUPPORTED SUPPORTEDH3: AVOIDANCE < OTHER PCIN/A N/AN/AH4: SN -> INNOVATIVENESSNOT NOTNOTSUPPORTED SUPPORTEDSUPPORTEDH5: SN -> ATTITUDEN/A NOT NOTSUPPORTED SUPPORTEDH6: VOLUNTARY -> INNOVATIVENESSSUPPORTED NOTNOTSUPPORTED SUPPORTEDH7: VOLUNTARY -> ATTITUDEN/A NOT SUPPORTEDSUPPORTEDH8: SUPPORT -> INNOVATIVENESSNOT SUPPORTED SUPPORTEDSUPPORTEDH9: FRIEND -> INNOVATIVENESSNOT NOT NOTSUPPORTED SUPPORTED SUPPORTEDH1O: EMPLOYEE -> INNOVATIVENESSNOT SUPPORTEDSUPPORTEDSUPPORTEDHi 1: ACCOUNTANT -> INNOVATIVENOT NOTNOTSUPPORTED SUPPORTED SUPPORTEDH12: CONSULTANT -> INNOVATIVENOT SUPPORTEDSUPPORTEDSUPPORTEDHi 3: SUPPORT -> SNN/A NOTSUPPORTEDSUPPORTEDH14: SUPPORT -> ATTITUDEN/A NOT SUPPORTEDSUPPORTEDHI 5: SUPPORT -> PCVSN/A SUPPORTED SUPPORTED98TABLE 13(b)SUMMARY RESULTS OFHYPOTHESIS TESTING(MOORE)HYPOTHESESADOPTERS VS. REGRESSIONPLSNON-ADOPTERS ANALYSISANALYSISHI: ATTITUDE -> INNOVATIVENESSSUPPORTED SUPPORTEDSUPPORTEDH2: RELATIVE ADV> OTHER PCINOT SUPPORTED NOTSUPPORTED SUPPORTEDH3: AVOIDANCE < OTHER PCISUPPORTED NOT SUPPORTEDSUPPORTEDH4: SN -> INNOVATIVENESSSUPPORTED NOT SUPPORTEDSUPPORTEDH5: SN -> ATTITUDEN/A SUPPORTED SUPPORTEDH6: VOLUNTARY -> INNOVATIVESUPPORTED SUPPORTEDSUPPORTEDH7: VOLUNTARY -> ATTITUDEN/A SUPPORTED SUPPORTED99TABLE 14GENERAL PLS STATISTICSFOR TESTED MODELSMODELMULTIPLE R AVG.COMMUN. AVG. REDUND.. (R2)(H2) (F2)All data points:Full Model.1837 .5275.1097Full - Voluntariness.1281 .4636 .0584Full - Voluntariness - SN1357 .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—FIGURE1DiffusionofInformatIonTechnologyModel-Moore,1989TOWARDSSUBJECTIVENORMVOLUNTARINESSATTITUDEINNOVATIVENESSADOPTINGFIGURE2DiffusionoiInnovationsModel-Rogers,1983PerceivedAttributesCommunicationChannelsofInnovations-MassMedia-RelativeAdvantage-Interpersonal-Compatibility-Trialability-Complexity-ObservabilityTypeofInnovation-DecisionR4TEOFADOPTIONExtentofChangeAgents’-OptionalrsPromotionEfforts-CollectiveOFINNOVATIONS-AuthorityI____________________________NatureofSocialSystem-Norms -DegreeofInterconnectednessFIGURE3COMMUNICATIONS NETWORKInnovationDecisionModel-Fishbein&Ajzen,1975(AdaptedbyMoore,1989)OBJECTIVE CHARACTERISTICS OFINNOVATIONTOWARDS ADOPTINGINTENTIONINNOVATIONDECISIO))SUBJECTIVE-1PERSONAL CHARACTERISTICS OFADOPTERSNORMBEHAVIOURALN. 1/BEHAVIOUR(ADOPTION/REJECTIO1ATTITUDEOBJECTIVE CHARACTERISTICS OFPRECURSORCFIGURE4StagesoftheInnovationDecisionProcessModelRogers,1983(AdaptedbyMoore,1989)KNOWLEDGE PERSUASION___________DECISIONADOPTIONREJECTIONffiMATIONJlFIGURE5NON-CASUSERSRESULTDEMONSTRABILITYRs Ac TA NL GE0100010101030104010501060107016802030204020602070208RESPONDENT0209021002110216021702190233P124P1257 64 3 0-A—IIIIIIIIIIIIIII•U15U23AU331FIGURE6DIFFUSTIONOFINFORMATIONTECHNOLOGYMODELPLSLOADINGSONORIGINALMODELC=LATENTVARIABLE(LV)II=MANIFESTVARIABLE(MV)INNERMATRIX=OUTERMATRIX-REFLECTIVEOUTERMATRIX-FORMATrVESubordinates8Z5.3.91417092.90946q9“73FIGURE7DIFFUSIONOFINFORMATIONTECHNOLOGYMODELPLSLOADINGSONEXTENDEDMODEL.2935=LATENTVARIABLEII=MANIFESTVARIABLE.9136CC109BIBLIOGRAPHYAhituv, Niv, “Assessing TheValue Of InformationProblems and Approaches”,Proceedingsof the Tenth Internationalconference on InformationSystems, December 4-6, 1989,pp.3 15-3 25.Ajzen, Icek and Fishbein,Martin, UnderstandingAttitudes and Predicting Behavior,Prentice-Hall Inc., Englewood Cliffs,NJ, 1980.Alavi, Maryam and Weiss,Ira R., “Managing The RisksAssociated 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., “StrategiesFor End-User Computing:AnIntegrative Framework”, Journalof MIS, Vol. 4, No. 3, Winter 1987-88,pp.28-49.Alexander, Mary B., The AdoptionAnd Implementation OfComputer Technology InOrganizations: The ExampleOf Database Machines, UnpublishedPhD thesis, IndianaUniversity, 1989.Allen, Brandt, “An UnmanagedComputer System CanStop You Dead”, Harvard BusinessReview, November-December1982,pp.77-8 7.Arter, TerrenceJ., “Contract The Independent ComputerConsultant”, Journal of SmallSystems Management, January 1988,pp36-37.Arthur Anderson & Co. “EntrepreneursSpeak Out”, BC Business, April, 1990,pp.73-77.Attewell, Paul, “Survey& Other Methodologies Applied ToIT. Impact Research”, Paperpresented at the Workshopon Survey Research in MIS, Irvine,California, February 10-11, 1989.Babcock, Charles, “Big 8 FirmTied to 4GL Snafu”, ComputerWorld, July 28, 1986,pp.1,4.Bagozzi, Richard P.“Causal Modeling: A GeneralMethod For DevelopingAnd TestingTheories In Consumer Research”,American Marketing Association,1982,pp.195-20 1.Bailey, James E. andPearson, Sammy W., “Developmentof a Tool For MeasuringandAnalyzing Computer UserSatisfaction”, ManagementScience, Vol. 29, No. 5, May1983,pp.530-545.Baker, William H.,“Status Of Information ManagementIn Small Business”, Journal ofSystems Management. April,1987,pp.10-15.Barki, H. and Huff,Sid L., “Measuring DecisionSupport System Success”,ASAC 1984Conference.Barclay, D.W., Higgins, C.A.,& Duxbury, L.E., An Introductionto the Partial Least SquaresApproach to Causal Modeling,University of Western Ontario,WP-91-08, 1991.110Baronas, Ann-MarieK. and Louis, MerylReis, “Restoringa Sense of Control DuringImplementation; HowUser InvolvementLeads to System Acceptance”,MIS Quarterly,March 1988,pp. 111-123.Baroudi, JackJ., Olson, Margrethe H.,and Ives, Blake,“An Empirical StudyOf The ImpactOf User InvolvementOn System Usage AndInformationSatisfaction”, Communicationsof the ACM, Vol.29, No. 3, March 1986,pp.232-238.Batch, CR.; Fisher, J.;and McIntyre,R,, “Accountants TakeTo IT In A BigWay”,IhcBottom Line, November1989,pp.22-24.Benjamin, RobertI., “Information Technologyin the 1990s: ALong Range PlanningScenario”, MIS Quartey,June 1982,pp.11-31.Benjamin, RobertI. and Scott Morton,Michael S., “InformationTechnology, Integration,And OrganizationalChange”, Interfaces,Vol. 18, No. 3, May-June1988,pp.86-98.Benson, David H.,“A Field Studyof End User Computing:Findings and Issue&’,MIQuarterly, December1983,pp.35-45.Boon, J.A. and Pienaar,H., “Some Ideas OnThe Microcomputer AndThe InformationIKnowledge Workstation”,Microcomputers forInformation Management,Vol. 6, No. 2,pp.113-122.Bouldin, BarbaraM., Agents of Change,by Edward Press ComputingSeries, 1989.Boynton, AndrewC. and Zmud, RobertW., “InformationTechnology Planning in the1990’s:Directions for Practiceand Research”, MISQuarterly, March 1987,pp.59-71.Bracker, JeffieyS. and Pearson, JohnN., “The Impact OfConsultants On Small FirmStrategic Planning”,Journal of Small BusinessManagement, July1985,pp.23-30.Bradbard, David A.;Norris, Dwight R.; andKahai, Paramjit H.,“Computer Securityin SmallBusiness: An EmpiricalStudy”, Journal of SmallBusiness Management,Vol. 28, No. 1,January 1990,pp.9-19.Brancheau, JamesC., The Diffusion OfInformation Technology:Tejg And ExtendingInnovation DiffusionTheory In The ContextOf End-User Computing,Unpublished PhDthesis, University ofMinnesota, 1987.Brancheau, JamesC., and Wetherbe,James C., “TheAdoption Of SpreadsheetSoftware:Testing InnovationDiffusion TheoryIn The Context ofEnd-User Computing”,Information SystResearch, Issue 1: Vol.2,pp.115-141.Brown, DonaldA., “So Far, SoGood”, CA Magazine,April 1992,pp.56-58.111Bryman, Alan, and Cramer, Duncan,Quantative Data Analysis For SocialScientists,Routledge, London UK, 1990.Buckler, Grant, “Boost Your Training Buck”,Small Business, March 1990,pp. 65-66,7 1.Buechert, Dennis “Skilled Labour ShortageChronic, List Shows”, The VancouverSun,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 withthe researcherduring early June, 1992.Churchill, Gilbert A., “A Paradigmfor Developing Better Measures of MarketingConstructs”,Journal of Marketing Research,Vol. XVI, February 1979,pp.64-73.Clemons, Eric K. and Row, Michael,C., “Information Technology And EconomicReorganization”, Proceedings of TenthInternational Conference on InformationSystems, December 1989,pp.341-35 1.Cooper, Randolph B. and Zmud, RobertW., “Information Technology ImplementationResearch: A Technological DiffusionApproach”, Management Science, Vol.36, No. 2,February 1990,pp.123-139.Cox, Gary H., “Technology’s RewardsWithout The Risks”, Datamation, February1, 1990,pp.69-73.Davidson, WA., “Straw Accountants”, CAMagazine, September 1991,pp.43.Davidson, R.A. and Dalby, J.T., “PersonalityProfiles of Public Accountants”, WorkingPaper,University of Calgary, January 1991,from CA Magazine, September 1991,pp. 43, 45.Davis, Gordon B., “Caution: User-DevelopedDecision Support Systems Can Be DangerousTo Your Organization”, WorkingPaper No. MISRC-WP-82-04, ManagementInformation Systems Research Center, Schoolof 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. “Determinantsof Success for Computer Usage in SmallBusiness”.MiQuarterly, March 1988.Dillman, Don A., “Mailand Telephhone Surveys: The Total DesignMethod”, Wiley andSons, New York, 1978.Dimnik, Tony “An Introductionto LISREL”, The University of WesternOntario, WorkingPaper Series No. 86-21, 1986.112Dragich, MarthaJ., “Information Malpractice:Some ThoughtsOn The PotentialLiability OfInformation Professionals’,Information Technologyand Libraries, September1989,pp.265-272.Ein-Dor, Phillipand Segev, Eli, “OrganizationalContext and MISStructure: Some EmpiricalEvidence”, MISOuarterly, September 1982,pp.55-68.FBDB, “Small BusinessFacts and Figures”,Federal Business DevelopmentBank, 1987.Fournell C., TellisG.J., and ZinkhanG.M., “ValidityAssessment: A StructuralEquationsApproach Using PartialLeast Squares”, Proceedings,American MarketingAssociationEducators’ Conference,1982,pp.1-5.Frame!, John E.,“ManagingInformation Costs AndTechnologies As Assets”,Journal ofSystems Management,February 1990,pp.12-19.Gable, Guy G. “ConsultantEngagement forFirst Time Computerization:A Pro-Active ClientRole in Small Enterprises”,ASAC 1989 Conference,pp.57-60.Gallupe, R. Brent,“End-User Computing:Individual Differencesand Successful EndUsers,An ExploratoryCase Study”, ASAC ConferenceMontreal, 1989,pp.61-75.Geilman, Harvey, “SoaringWith Eagles”, OfficeSystems and Technology,Vol. 37, No. 2,March 1991,pp.28-42.Goodhue, Dale,“IS Attitudes: TowardTheoretical And DefinitionClarity”, ProceedingsofSeventh InterritionalConference on InformationSystems, December 1986,pp.181-194.Goodson, Leslie, “What’sWrong With ThisPicture?”, CA Magazine,August 1990,pp.20-28.Gotleib, Leo, “ISMay Be An ‘HonestBroker’ BetweenVendors, Users”,Globe & Mail,March 6, 1990,pp. C15.Gremillion, LeeL. and Pyburn,Philip, “Breakingthe Systems DevelopmentBottleneck”,Harvard Business Review,March-April, 1983,pp.130-137.Gunning, Ken “AccountingFor Canada”, CAMagazine, April 1992,pp.3 9-42.Hamilton, Marilyn,Small Business ViewsOf’ The Accounting Professionin Canada - Survey9, Dobson MarketingServices, 1989, Abbotsford,B.C. Portions of thissurvey werereproduced in Sial1usinessMagazii.ç,January 1990.Hill, Thomas; Smith,Nancy D. and Mann,Millard F., “Roleof Efficacy ExpectationsinPredicting the Decisionto Use AdvancedTechnologies: TheCase of Computers”,Journal of AppliedPsychology, Vol.72, No. 2,pp.307-3 13.113Holmes, Scottand Nicholls,Des. “Modellingthe AccountingInformationRequirements ofSmall Business”,Accountingand Business Research,Vol. 19, No. 74,1989,pp. 143-150.Huber, 0. P.,“A Theory ofthe Effects ofAdvanced InformationTechnologies onOrganizationalDesign, Intelligence,and Decision Making”,Academy of ManagementReview, Vol.15, No. 1,January 1990.Ives, Blake,Olson, MargretheH. and Baroudi,Jack J., “TheMeasurementof UserInformation Satisfaction”,Communicationsof the ACM, Vol.26, No. 10, October1983,pp.785-793.Jenish, D’Arcy,“A ‘Terrorist’ Virus”,Maclean’s, March16, 1992,pp.48-51.Kauffman, RobertJ. and Weill, Peter, “AnEvaluative FrameworkFor Research On ThePerformance EffectsOf InformationTechnology Investment”,Proceedings ofthe TenthInternational ConferenceonInformatiolLSystems, December4-6, 1989,pp.377-388.Kraut, Robert;Dumais, Susan;and Koch, Susan,“Computerization,Productivity, AndQuality Of Work-Life”,Communicationsof the ACM, Vol.32, No. 2, February1989,pp.220-23 8.Kuntz, M.A.,and Maingot, M.,“Viruses Are Aliveand Well”, CAMagazine, December1989,pp.60-6 I.Lees, John D.,‘Successful DevelopmentOf Small BusinessInformation Systems”,Journal ofSystems Management,September 1987,pp. 32-39.Lees, JohnD. and Lees,Donna D., “RealitiesOf Small BusinessInformationSystemImplementation”,Journal ofSystems Management,January 1987,pp.6-13.Lefebvre, Elizabethand Lefebvre,Louis A., PlanningFor InformationTechnologyAcquisition: TheCase Of Small Business,unpublished workingpaper, the UniversityofQuebec at Montreal,1990.Lin, Engmingand Ashcraft,Phillip, “ACase Of SystemsDevelopment InA HostileEnvironment”, JournalOf SystemsManagement, April1990,pp.11-14.Lind, MaryR. arid Zmud,Robert W., “TheInfluence ofa Convergence in UnderstandingBetween TechnologyProviders and Userson InformationTechnologyIrinovativeness”,Org. Science, Vol.2, No. 4, 1990.Lohmoller, Jan-Bernd,“LVPLS 1 .6Program Manual: LatentVariables PathAnalysis WithPartial Least-SquaresEstimation”, Munich,1984.Lucas, HenryC. Jr., “EmpiricalEvidence Fora Descriptive Model ofImplementation”,MISQuarterly, June1978,pp. 27-41.114Luscombe, N. “Representationor Fragmentation?”, CAMagazine, May 1990,pg. 3.McFarlan, Warren F. andMcKenny, James L.,“Corporate InformationSystems Management:The Issues Facing SeniorExecutives”, Richard D. IrwinInc., USA, 1983.McGill, Penny, “FocusedApproach Helps WhenDesigning Firm’s InformationSystems”,Globe & Mail, March 6, 1990,pp. C12.McKeen, James D.,“Successful DevelopmentStrategies for Business ApplictionSystems”,MIS Quarterly, September1983,pp.47-65.McLean, E. R., “EndUsers As ApplicationDevelopers”, MIS Quarterly,December 1979,pp.37-46.Melone, Nancy P., “ATheoretical AsessmentOf The User-SatisfactionConstruct InInformation Systems Research”.,Management Science,Vol. 36, No. 1, January 1990,pp.76-91.Melone, Nancy P. andBayer, Judy, “A DynamicModel Of The ImpactOf Communication-Channel Use On TheTechnology-Transfer Process”,GSIA WP# 1990-21,CarnegieMellon University, July 1990.Miller, Howard W.,“Developing InformationTechnology Strategies”,Journal of SystemsManagement, September 1988,pp.28-3 5.Millman, Zeeva andHartwick, Jon, “The ImpactOf Automated OfficeSystems On MiddleManagers And Their Work”,MIS Quartjy, December1987,pp.479-491.Moore, Gary C., The Examinationof the Implementationof Information Teno1ogvFor EndUsers: A Diffusion ofInnovations Perspective,Unpublished PhD thesis,UBC 1989.Moore, GaryC. and Benbasat, Izak,“Development of anInstrument To MeasureThePerceived Characteristicsof Adopting an InformationTechnology Innovation”,ThcInstitute of ManagementSciences, Sept. 1991,pp.192-222.Nelson, R.R. and Cheney,Paul H., ‘Training EndUsers: An ExploratoryStudy”,MISQuartiy, December 1 987.Newman, Michael, “ChangingThe Change Agent:An Admissions SysemCase Study”,Journal of SystemsManagement, January1990,pp.6-12.Nilakanta, Sree andScamell, RichardW., “The Effect Of InformationSources AndCommunication ChannelsOn The DiffusionOf Innovation In AData Base DevelopmentEnvironment”, ManagementScience, Vol. 36, No.1, January 1990,pp.24-40.Nunnally, JumC., Psychometric Theory,McGraw Hill, New York,1967.Nunnally, JurnC., 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 PersonalComputing: An Empirical Test”,Proceedings of the Tenth International Conferenceon 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 MISSuccess in the Context of SmallBusiness’, MIS OuarteiJ, March 1985,pp. ???Reich, Blaize H. and Benbasat, Izak, “TheUse Of Information Technology For CompetitiveAdvantage in Canada: An Examination Of InformationSystems 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 andEconomic Development in 1990),October 3, 1990,pp.3.Rivard, Suzanne and Huff, Sid L., “Factorsof Success for End-UserComputing”,Communications of the ACM, Vol. 31, No. 5, May1988,pp.553-561.Rockart, John F. and Flannery, LaurenS., “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 BePrevalent But is Also Preventable”, Globe& Mail, March 6, 1990,pp. C6.Rogers, Everett M., Diffusion ofInnovations, 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 Instituteof Chartered Accountants, 1986.116Sanders, G.L. and Courtney,JR. “A Field Study of Organizational FactorsInfluencing DSSSuccess”, MIS Quarterly, March 1985.Sein, Maung K., Bostrom, RobertP.; and Olfman, Lorne, “Training End UsersTo Compute:Cognitive, Motivational AndSocial Issues”, INFOR, Vol. 25, No. 3, August1987,pp.236-255.Small Business Magazine, October 1989,title & author unknown.Smith, Charlie. “The DominantSolution”, Equity, April 1989,pp25.Steiren, Carl, “Firms Find On-SiteInstruction Inexpensive and Convenient”, Globe& Mail,March 6, 1990,pp. C4.Stone, Eugene, Research Methods in OrganizationalBehavior, Scott, Foresman andCompany, 1978).Stulberg, Gregg, Focus On Office Technology, FinancialPost, March 22, 1991,pp.22Thompson, Ron L., “An EmpiricalInvestigation Of Factors Affecting The Use Of PersonalComputers By Knowledge Workers”, ASAC 1989 Conference,1989,pp.141-152.Walker, Robert, “Let’s Go For IT”, CAMagazine, April 1991,pp.33-35.Walton, Charles and Durham, Ashley, “InformationSystems 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 ByPartial Least Squares”, in Measuring The Unmeasurable,NATA AS! Series, Martinus NijhoffPublishers, Boston, 1985,pp.22 1-252.Zmud, Robert W., “IndividualDifferences and MIS Success: A Reviewof the EmpiricalLiterature”, Management Science,Vol. 25, No. 10, October 1979,pp.966-979.1/)118APPENDIX I-ARelative Advantage: the degreeto which an innovation is perceivedas being betterthan its precursorCompatibility: the degreeto which an innovation is perceivedas beingconsistent with the existing values,needs, and past experiences ofpotential adoptersEase of Use: the degree to which aninnovation is perceived as being difficult(Complexity) to useTrialability: the degree to which aninnovation may be experimented withbefore adoptionObservability: the degreeto which the results of an innovation are observabletoothersImage: the degree to whichuse of an innovation is perceivedto enhanceones image or status in ones social systemVoluntariness: the degreeto which use of the innovation is perceivedas beingvoluntary, or of free willVisibility: the degreeto which the innovation is apparentto the sense ofsightr —APPENDIX I-BINNOVATIVENESS: (Moore, 1989,pp.133)Adoptive degree to which anindividual is relatively early in adoptinganinnovation.Implementation degree to which an individualputs an innovation to use within agiven use domain.Use degree to which an individualwho has adopted the innovation usesit to solve novel problems, or ina new use domain.119(From Stone, 1978)VALIDITY ITEM DEFINITIONContent ValidityMeasurement items are representative sampleofdomain of items associated with variablebeingmeasured.Construct Validity Appropriateoperational definition cf anabstract variable (construct)Criterjon-related —— Use of scoresobtained from one measureValidity (predictor)to infer individual’s probablestanding on another variable (criterion)Face ValidityItem appears to measure what it claimstomeasure.Incremental Validity Item providesan improvement in predictivepower in conjunction with other measure(s)over the use of the other measure(s) alone.Convergent/Discriminant Scores onthe measure correlate highly withValidity scores onother independent measures of thevariable and correlate low on measuresofother variables.APPENDIX I-CNote: There are several other Validityitems, 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 sevenplaces; youare asked to place an ‘X’ in the place that best describes your opinion. For example, if youwere askedto rate “Driving a car in winter is easy” on such a scale, it would appearas follows:Driving a car in winter is easy.likelyI I I I I Iunlikelyextremely quite slightly neither slightly quite extremelyIf you think that it is extremely likely that driving a car in winter is easy, you would makeyour markas follows:Driving a car in winter is easy.likelyLxI I I I Iunlikelyextremely quite slightly neither slightly quite extremelyIf you think that it is neither likely nor unlikely that driving a carin winter is easy, you would makeyour mark as follows:Driving a car in winter is easy.likelyI I IXfunlikelyextremely quite slightly neither slightly quiteextremelyIn 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, pleaseremember the following points:1. Place your marks in the middle of spaces, NOT ON TILE BOUNDARIES.likelyIXIunlikelyextremely 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 willbe asked to circle a number or lettercorresponding to a particular answer for a question. Pleasebe careful to see that your circle goes aroundonly the letter or number which corresponds to your desiredresponse.124TO BEGIN, WE WOULD LIKE TO ASK YOU ABOUTYOUR EXPERIENCE WITH.COMPUTERS AND OTHER HIGH-TECHNOLOGY PRODUCTSAND SERVICES.A-i Have you ever used a multi-functiontelephone (including such functions as call forward, speeddialing, call waiting, etc.). (Placean ‘X’ beside the appropriate answer):___NOYESIf yes, which functions do you use? (Place an ‘X’ besidethe 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 eachapplicablearea):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 46-6061-75 >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 personalhands-on use ofcomputers? (eg: “applied” courses)COURSESA-7 My firm receives non-computer support for the following areas (placean ‘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 externalto 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 receivefrom thefollowing sources external to the firm (place an ‘X’ under the appropriate columnfor eachapplicable source):satisfied unsatisfiedextremely quite slightly neither slightly quitefextremelyPersonal 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.nütedI I I I I Iextremely quite slightly neither slightly quiteextremely128A-il How knowledgeable do you feel you are of the uses of the CAS?unlimitedI 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 CASPlease go on to the next pageHAVE NEVER USED A CASPlease 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 YEARSTOPPED____ ____MONTH 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.disagreeI I I I I I _iagreestrongly quite slightly neither slightly quite stronglyU-2 Using a CAS is completely compatible with my current situation.disagreeI I I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-3 Using a CAS is compatible with all aspects of my work.disagreeI I I I I Istrongly quite slightly neither slightly quite stronglyU-4 My superiors expect me to use a CAS.disagreclI I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-S I believe that a CAS is cumbersome to use.disagreelI I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-6 Using a CAS improves my image within the organization.disagreelI I I I I Iagreestrongly quite slightly neither slightly quite strongly130U-7 Using a CAS improves the quality of work I do.disagreeI I I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-8 Using a CAS makes it easier to do my job.dLsagreelI I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-9 I think that using a CAS fits well with the way I like to work.disagreeI I I I I Iiagleestrongly quite slightly neither slightly quite stronglyU-lO My use of a CAS is voluntary (as opposed to required by my superiors or job description).disagreclI I I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-li I have seen what others do using their CAS.disagreeI I I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-12 I’ve had a great deal of opportunity to try various CAS applications.disagreeI I I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-13 In my organization, one sees CAS on many desks.disagreeI I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-14 My boss does not require me to use a CAS.disagreelI I I I I Iagreestrongly quite slightly neither slightly quite strongly131U-15 I would have no difficulty telling others about theresults of using a CAS.disagreciI I I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-16 I know whereI can go to satisfactorily try out various uses of a CAS.disagreciI I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-17 People in my organization who use a CAShave more prestige than those who do not.disagreeI I I I I Irecstrongly quite slightly neither slightly quite stronglyU-18 Although itmight be helpful, using a CAS is certainly not compulsory in my job.disagreelI I I I Ijagreestrongly quite slightly neither slightly quite stronglyU19 My using a CAS requires a lot of mental effort.disagreelI I I Iagreestrongly quite slightly neither slightly quitestronglyU-20 Using a CAS is often frustrating.disagreeI I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-21 People in my organization who use a CAS have a high profile.disagreeI Iagreestrongly quite slightly neither slightlyquite stronglyU-22 A CAS was available to me to adequately testrun various applications.disagreejI I I I II 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 IdisagreelI I Istrongly quite slightlyU-25 Overall, I believe that a CAS is easy to use.disagreeI I Istrongly quite slightly neitherU-26 Using a CAS improves my job performance.disagreeLI 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.disagreelI I Istrongly quite slightly neither slightly quite stronglyU-28 Overall, I find using a CAS to be advantageous to my job.disagreel jI I Istrongly quite slightly neither slightly. I Iquite stronglydisagreeII I I IdisagreeIneither slightly quite stronglyI I IIjagreej agreestrongly quite slightly neither slightly quite strongly133U-3 1 Using a CAS enhances my effectiveness on the job.disagrec[I I I I I Ilagreestrongly quits slightly neither slightly quite stronglyU-32 Using a CAS fits into my work style.disagrccI 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.disagrecI I I I I Iagreestrongly 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.disagreelI I I I I I Iagreestrongly quite slightly neither slightly quitsstronglyU-35 Using a CAS gives me greater control over my work.disagrecI I I I 1 I 1agreestrongly quite slightly neither slightly quits stronglyU-36 Using a CAS increases my productivity.disagreelI I I I I Iagreestrongly quite slightly neither slightly quite stronglyU-37 Having a CAS is a statussymbol in my organization.disagreeI I I I I Iagreestrongly quite slightly neither slightly quits stronglyU-38 It is easy for me to observe others using CAS in my firm.disagreclI I I I Ilagreestrongly quits slightly neither slightly quits stronglyU-39 I have had plenty of opportunity to see the CAS being used.disagreelI I I I I Istrongly quite slightly neither slightly quite strongly134FINALLY, IN THIS SECTION WEWOULD LIKE TO ASK YOU A FEW QUESTIONSABOUT YOUR USEOF THE CAS.B-i Overall, my using aCAS in my job is (place an X on all four scales):wmenegativeextremely quite slightly neither slightlyquite extremelyfoolishpo8itweB-2 Assuming that any decision to use the CASis totally up to you, how would you rate yourpotential use of the CAS in the next sixmonths?likelyI I I I I IIunlikelyextremely quite slightlyneither slightly quite extremelyimprobableprobablegoodiI I I I II Ihannfiilextremely quite slightly neither slightly quiteextremelyI I I I II Iextremely quite slightly neither slightlyquite extremelybadbeneficialI I I II I Iextremely quite slightly neither slightlyquite extremelyI I I II I I Iextremely quite slightly neither slightlyquite extremely135B-3 Approximately when (month and year) did youfirst start using a CAS beyond any trial of it youmay have carried out?MONTH YEARB-4 How regularly do you now use aCAS? (Place an ‘X’ under the appropriate column):Less than About 1-3 AboutAbout More thanonce per times per once per 2-4 timesonce per once perNot at all month month week perweek day dayI I IB-S On average, how frequently do youcurrently use the following functions (place an ‘X’ under theappropriate column):Less Aboutthan 1-3 About 2-4 About Moreonce times once timesonce thanNot at per per per perper onceall month month weekweek 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):increaselI I I I I I Idecreasesignifi- 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):increasediI I I I I I Idecreasedaigniii- 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 myCAS fromthe followingsources (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 myCAS 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 canget support from the following sources(place an ‘X’ under the appropriate column for each source):noneongoing1 2J3 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 yourcurrent 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 PRESENTYOU WITH A NUMBER OF STATEMENTSEXPRESSING PARTICULAR VIEWPOINTSABOUT THE CAS. WE WOULD LIKEYOUTO INDICATE HOW MUCH EACHSTATEMENT REFLECTS YOUR PERSONALVIEWPOINT PLACING AN ‘X’ IN THEAPPROPRIATE PLACE ON ThEDISAGREE-AGREE SCALE. ALTHOUGH THEREMAY APPEAR TO BE A NUMBEROF SIMILARSTATEMENTS, PLEASE PROVIDE A RESPONSETO EACH ONE.N-i Using a CAS would enable meto accomplish tasks more quickly.disagreciI I II Istrongly quits slightlyneither slightly quite stronglyN-2 Using a CAS would improvethe quality of work I do.disagreeI II I I IIagreestrongly quits slightly neitherslightly quite stronglyN-3 Using a CAS would be compatible withall aspects of my work.disagreeI I I II Iagreestrongly quits slightlyneither slightly quite stronglyN-4 My superiors expect meto use a CAS.disagreclI I I II I ]agreestrongly quits slightly neither slightlyquits stronglyN-5 I believe that a CAS wouldbe cumbersome to use.disagree1 I I I IIagreestrongly quits slightly neither slightlyquits stronglyN-6 Using a CAS would improvemy image within the organization.disagrecfI I I II I Iagreestrongly quite slightly neither slightlyquite stronglyN-7 Using a CAS wouldbe completely compatible with mycurrent situation.disagreeiI I I II Iagreestrongly quits slightly neitherslightly quits strongly142N-8 Using a CAS would make it easier to do my job.disagrecjI 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.disagreciI I I I I I Iagreestrongly quits slightly neither slightlyquite stronglyN-IO My use of a CAS is voluntary (as opposedto required by my superiors or job description).disagrccI I I I I I Iagreestrongly quite slightly neither slightly quite stronglyN-Il I have seen what others do using their CAS.disagreelI I I I I I Iagreestrongly quits slightly neither slightly quite stronglyN-12 I’ve had a great deal of opportunity to try various CAS applications.disagreeI I I I Iagreestrongly quits slightly neither slightly quits stronglyN-13 In my organization, one sees CAS on many desks.disagreeI I I I I Iagreestrongly quite slightly neither slightly quits stronglyN-14 My boss does not require me to use a CAS.disagreeI I I I I Iagreestrongly quits slightly neither slightly quits stronglyN-15 I would have difficulty telling others about the results of using a CAS.disagreeI I I I I Iagreestrongly quite slightly neither slightly quite strongly143N-16 I know where I can go to satisfactorily try out various usesof a CAS.disagrecI I I I I Iagreestrongly quite slightly neither slightlyquite stronglyN-17 People in my organization who use a CAS have more prestige thanthose who do not.disagreelI I I I I I Istrongly quite slightly neither slightly quitestronglyN-18 Although it might be helpful, using a CAS is certainly not compulsory in myjob.disagreesI I I I Iagreestrongly quite slightly neither slightly quite stronglyN-19 My using a CAS would require a lot of mental effort.disagrccI I I I I I Iagreestrongly quite slightly neither slightly quite stronglyN-20 Using a CAS would often be frustrating.disagreelI I I I I I Iagreestrongly quite slightly neither slightly quite stronglyN-21 People in my organization who use a CAS have a high profile.disagreeI I I I I Iagreestrongly quite slightly neither slightly quite stronglyN-22 A CAS is available to me to adequately test run various applications.disagreelI I I I I Iagreestrongly quite slightly neither slightly quite stronglyN-23 I believe I could communicate to others the consequence of usinga CAS.disagreeI I I I Iagreestrongly quite slightly neither slightly quite strongly144N-24 I believe that it would be easy to get a CASto do what I want it to do.disagreciI I I I I Iagreestrongly quite slightly neither slightly quite stronglyN-25 Overall, I believe that a CAS would be easy to use.diaagreeiI I I I I I Irestrongly quite slightly neither slightly quite stronglyN-26 Using a CAS would improve my job performance.disagreeI I I I I I Iagreestrongly quite slightly neither slightly quite stronglyN-27 CAS are not very visible in my organization.disagreeI I I I I I Iagreestrongly quite slightly neither slightly quite stronglyN-28 Overall, I would find using a CAS to be advantageous in myjob.disagreciI I I I I Istrongly quite slightly neither slightly quite stronglyN-29 Before deciding whether to use any CAS applications, I wouldbe able to properly try them out.disagreeI I I I I Iagreestrongly quite slightly neither slightly quite stronglyN-30 Learning to operate a CAS would be easy for me.disagreeI I I I I Iagreestrongly quite slightly neither slightly quite stronglyN-31 Using a CAS would enhance my effectiveness on the job.disagree[I I I I I Ilagrecstrongly quite slightly neither slightly quite strongly145N-32 Using a CAS would fit into my workstyle.disagreelI I I Istrongly quits slightly neither slightly quite stronglyN-33 If I were to use a CAS, I would have difficulty explainingwhy using a CAS may or may not bebeneficial.disagreeI I I I Iagreestrongly quits slightly neither slightly quite stronglyN-34 I would be permitted to use a CAS on a trial basis long enough tosee what it could do.disacI I I I Istrongly quite slightly neither slightly quite stronglyN-35 Using a CAS would give me greater controlover my work.disagreelI I Istrongly quite slightly neither slightly quite stronglyN-36 Using a CAS would increase my productivity.disagreciI I I Istrongly quite slightly neither slightly quite stronglyN-37 Having a CAS is a status symbol in my organization.disagreelI I I Iquite stronglyI I Istrongly quite slightly neither slightlyN-38 It is easy for me to observe others using aCAS in my firm.disagreciI — I I Istrongly quits slightly neither slightlyN-39 I have had plenty of opportunity to see theCAS being used.disagreeI I I I I IagreeIjagreejagreejagreeagreeagreeagreequite stronglystrongly quite slightly neither slightly quite strongly146FINALLY, INTHIS SECTIONWE WOULD LIKETO ASK AFEW GENERALQUESTIONS.C-I Overall, myusing a CAS inmy job wouldbe (place an X onall four scales):I II II II ‘‘extremely quiteslightly neitherslightly quiteextremelyhannfulfI III IIextremely quiteslightly neitherslightly quiteextremelywmeI II IIfoolishextremely quiteslightly neitherslightly quiteextremelynegativeL__.._. I II IIIpositiveextremely quiteslightly neitherslightly quiteextremelyC-2 Assumingthat any decisionto use the CAS istotally up to you,how wouldyou rate yourpotential use ofthe CAS inthe next six months?likelyunlikelyextremely quiteslightly neitherslightly quiteextremelyimprobableprobableextremely quiteslightly neitherslightly quiteextremely147C-3 Approximatelyhow 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 Dayper 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? (Placean ‘X’ underthe appropriate column for each applicablesource):Less About Aboutthan 1-3 2-4 MoreOnce Times Once Times About than Did notNotat per per per per Once Once UseA]] Month Month WeekWeek per Day per Day CASOther personnel frommy companyPersonal friend (non-employee)Public accounting firmNon-Accountantcomputer consultantOther(please specify)148C-.5 Identify yourSUPPORT GROUP, whose official function it would be to support youin the CAS(if more than one choose the primary source ofhelp); (place an ‘X’ under the appropriate columnfor each applicable source):Other personnelNon-accountantfrom my Professionalcomputercompany Personal friend accountingfirm consultant NoneTHANK YOU FOR YOUR PERSEVERANCE ANDCOOPERATION SO FAR. NOW,PLEASE GO ON TO THE NEXT PAGE.149In this last section,we would liketo ask you somequestions about yourself.Remember, allanswers areconfidential, andno respondentcan be identified,so please give ascandid a responseas possible.FIRST, WE WOULDLIKE YOU TOONCE AGAIN INDICATEAGREEMENTORDISAGREEMENTWITH A NUMBEROF STATEMENTS;THIS TIME ABOUTYOURSELF.PLEASE PLACEAN ‘X’ IN THEAPPROPRIATESPACE.I-i I am generallycautious aboutaccepting newideas.disagrecI III I1agstrongly quiteslightly neitherslightly quitestrongly1-2 I rarelytrust new ideasuntil I cansee whether thevast majorityof people aroundme acceptthem.disagreeiI II I.1 Istrongly quiteslightly neitherslightly quitestrongly1-3 1 am awarethat I am usually oneof the last peoplein my groupto accept somethingnew.disagreeI III Iagreestrongly quiteslightly neitherslightly quitestrongly1-4 1 amreluctant aboutadopting newways of doingthings until I seethem workingfor peoplearound me.disagreeI II II Iagreestrongly quiteslightly neitherslightly quitestrongly1-5 1 find itstimulating to be originalin my thinking andbehaviour.disagreelI II II Iagreestrongly quiteslightly neitherslightly quitestrongly1-6 I tendto feel that the oldway of living anddoing thingsis the best way.disagreeI II II IJagreestrongly quiteslightly neitherslightly quitestrongly1501-7 I am challenged by ambiguities andunsolved problems.disagreefI I I II I fagecstrongly quite slightly neither slightlyquite strongly1-8 I must see other people using new innovations before I will considerthem.aisagrecfI I I II I(agreestrongly quite slightly neitherslightly quite strongly1-9 1 am challenged by unanswered questions.disagreeI I I II I Iagreestrongly quite slightly neitherslightly quite strongly1-10 1 often find myself sceptical of new ideas.disagreeI I I II I Iagreestrongly quite slightly neitherslightly quite stronglyNEXT, WOULD YOU PLEASE INDICATEHOW LIKELY OR UNLIKELY EACH OFTHE FOLLOWING STATEMENTS AREBY ONCE AGAIN PLACING AN ‘X’ IN THEAIPROPRIATE SPACE.S-i Most people who are important tome think I should use the CAS in my job.likelyunlikelyextremely quite slightly neither slightly quiteextremelyS-2 My close friends think that I should use theCAS in my job.likelyI I I I I Iunlikelyextremely quite slightly neither slightly quiteextremelyS-3 My co-workers (peers) think that I should use the CAS in myjob.likelyI I I Iunlikelyextremely quite slightly neither slightly quite extremely151S-4 My immediate superiors think that I should use the CAS inmy job.likelylI I I I I Iunlikelyextremely quite slightly neither slightly quiteextremelyS-5 Senior management thinks that I should use the CAS inmy job.lik.eIyII I I I I I Iunlikelyextremely quite slightly neither slightly quite extremelyS-6 My subordinates think I should use the CAS in my job.likelylI I I I II Iuldlyextremely quite slightly neither slightly quiteextremelyS-7 Generally speaking, I want to do what most people who are importantto me think I should do.likelyunlikelyextremely quite slightly neither slightly quite extremelyS-8 Generally speaking, I want to do what my close friends thinkI should do.likelyI I I I I Iunlikelyextremely quite slightly neither slightly quite extremelyS-9 Generally speaking, I want to do what my co-workers think I shoulddo.1iiclyI I I I 1 I Iunlilcelyextremely quite slightly neither slightly quite extremelyS-lO Generally speaking, I want to do what my immediatesupervisors think I should do.likelyI I I I I IJunlikelyextremely quite slightly neither slightly quite extremelyS-li Generally speaking, I want to do what senior management thinksI should do.likelyI I I Iunlikelyextremely quite slightly neither slightly quite extremelyS-12 Generally speaking, I want to do what my subordinates think I should do.152likelyI I I I I I Iunlikelyextremely quite 8lghtly neither 8lightly quite extremelyFINALLY, WE WOULD LIKE TO ASK A FEW QUESTIONS ABOUTYOURSELF[OR STATISTICAL PURPOSES. COULD YOU PLEASEINDICATE: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? Placean ‘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 theappropriatecolumn):___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 thespace 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 ofthe accounting function. Thekey aspect of a CAS is that it is computer technology that you would use directly, as opposedto 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 questionsas best as you can.6. Move rapidly through the questionnaire. We are interested in your firstimpressions, 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.likelyI I I I I I Iunlikelyextremely 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 XI I I Iunlikelyextremely 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.likelyf I I IXI I Iunlikelyextremely quite slightly neither slightly quite extremelyIn addition to the “likely-unlikely” pairs, other pairs suchas “disagree-agree” will also be used. Theyshould be answered in the same fashion. In making your ratings, please remember thefollowing points:1. Place your marks in the middle of spaces, NOT ON TIlE BOUNDARIES.likelyIXunlikelyextremely quite slightly neither slightly quite extremelyTillS NOT TillS2. Never put more than one ‘X’ ona single answer line.One other question format will be used. In this case, you will be asked to circlea number or lettercorresponding to a particular answer for a question. Pleasebe careful to see that your circle goes aroundonly the letter or number which corresponds to your desired response.158SECTION ATO BEGIN, WE WOULDLIKE TO ASKYOU ABOUT YOUREXPERIENCE WITHCOMPUTERS ANDOTHER HIGH-TECHNOLOGYPRODUCTS ANDSERVICES.A-i Have you everused a multi-functiontelephone (includingsuch functionsas call forward, speeddialing, call waiting,etc.). (Place an‘X’ beside the appropriateanswer):NO)LysIf you use a multi-functionphone, whichfunctions do you use?(Place an ‘X’beside theappropriate functions):CALL TRANSFER(CONSULTATIONS)HOLDTHREE-WAYCONFERENCECALL FORWARDINGCALL PARKING___CALL PICKUPCALL WAiTINGRING AGAIN/AUTOMATICCALL BACK%SPEED CALLINGX LAST NUMBERDIALLEDSAVE NUMBERAND REPEAT159A-2 How often do you use the products listed below? (Place an ‘X’ underthe 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-relatedactivities listed below; on paper, via electronicmail, on floppy disk, etc...? (Place an ‘X’ under the appropriate column foreach applicablearea):About 1-Less than 3 Times About 2- AboutMoreNot at Once per per Once per 4 Times Once perthan Onceall Month Month Week per WeekDay per DayReceive computer outputx(reports/documents)Submit documents, etc. toXothers 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>II I IOwpm 1-15 16-30 31-45 46-6061-75 >75wpm wpm wpmwpm wpm wpmA-5 How many educational courses (at any level) haveyou had about computers, but which didnotinclude your personal hands-on use?COURSESA-6 How many educational courses haveyou had which required your personalhands-on use ofcomputers?COURSESA-7 My firm receives non-CASsupport for the following areas(place an ‘X’ under the appropriatecolumn for each applicable area):noneconstant1 2 3jsome 5 67AccountingAuditBusiness AdviceFinancial PlanningGov’t Compliance —MarketingTaxXOther(please specify)161A-8 My firm receives non-CAS support fromthe following sources external to the firm (placean ‘X’under the appropriate column for each applicablesource):noneconstant1 2 31some 5 6Personal friend (nonemployee)Public accounting finnNon-Accountant computerconsultantNoneOther(please specify)A-9 I am satisfied with the current levelof support for non-CAS areas I receivefrom the followingsources external to the firm (place an ‘X’ underthe appropriate column for each applicablesource):satisfiedunsatisfiedextremely quite slightly neitherslightly quite extremelyPersonal friend (noncmploy)Public accounting finnNon-Accountantcomputer consultantNoneOther(please specify)A-1O How muchaccess to the use of a CAS do you feel you currentlyhave?unlimitedI I I II Ihuntedextremely quite slightly neitherslightly quite extremely162A-i 1 How knowledgeable do you feel you are of the uses of theCAS?unlimitedI<I I I I Ilimitedextremely quite slightly neither slightly quiteextremelyA-12 Have you ever used a CAS? (Place an ‘X’ beside theappropriate column):)(CURRENTLY USE A CAS_______________PLEASE SKIP TO SECTION BHAVE NEVER USED A CASPLEASE SKIP TO SECTIONCUSED TO USE A CAS BUT NO LONGER DOSOPlease answer A-13 to A-is only if youused to usea CAS but no longer do.A-13 Could you please indicate approximatelywhen you first began to use a CAS, and when youstopped using it.STARTED______ ______MONTH YEARSTOPPEDMONTH YEARA-i4 Please indicate which ofthe CAS functions below you used by indicatingthe number of monthsyou used them.Accounting Graphics Information Report SpreadsheetStatistical Textlword OtherSoftware Generation Retrieval GenerationAnalysis Processing (pleasespecify)MONTHS[A-i5 Couldyou please indicate very briefly whyyou no longer use the CAS.PLEASE SKIPTO SECTION CSECTION B163Please answer questions in this section only if youcurrently use the CAS.FIRST WE WOULD LIKE TO GET YOURIMPRESSIONS OF THE CAS. IN THEFOLLOWING, WE WILL PRESENT YOU WITHA NUMBER OF STATEMENTSEXPRESSING PARTICULAR VIEWPOINTSABOUT THE CAS. WE WOULD LIKE YOU TOINDICATE HOW MUCH EACH STATEMENT REFLECTSYOUR PERSONAL VIEWPOINTBY PLACING AN ‘X’ IN THE APPROPRIATE PLACEON THE DISAGREE-AGREE SCALESPROVIDED. ALTHOUGH THERE MAY APPEAR TOBE A NUMBER OF SIMILARSTATEMENTS, PLEASE PROVIDE A RESPONSE TOEACH ONE.U-i Using a CAS enables me to accomplishtasks more quickly.disagreciI I I II Istrongly quite slightly neitherslightly quite stronglyU-2 Using a CAS is completely compatible withmy current situation.disagreejI I I II ><‘ Iiagstrongly quite slightly neitherslightly quite stronglyU-3 Using a CAS is compatible with all aspectsof my work.disagreeI I I IIX agreestrongly quitc slightly neitherslightly quite stronglyU-4 My superiors expect me touse a CAS.1disagrceI I I II Iagreestrongly quite slightly neitherslightly quite stronglyU-5 I believe that a CAS is cumbersometo use.IdisagreeXI I I IIagreestrongly quite slightly neitherslightly quite stronglyU-6 Using a CAS improvesmy image within the organization.disagreeI IIagreestrongly quite slightly neitherslightly quite strongly164dagreeI I IIstrongly quite slightlyneither slightlyU-9 I think that using a CAS fitswell with the way I liketo work.disagrecI II I Istrongly quite slightlyneither slightly quiteU-lO My use of a CAS is voluntary(as opposed to required by mysuperiorsdisagreelI I <II I Istrongly quite slightlyU-il I have seen whatothers do using their CAS.disagreciIU-12U-7 Using a CAS improves the quality of workI do.disagrecI I Istrongly quite slightlyU-8 Using a CAS makes it easier to do myjob.I-neither slightly quitejagreeagreestronglyxquite strongly>( agrecstronglyor job description).agreeneither slightly quitestronglyU-13.lxistrongly quite slightlyneither slightly quite stronglyI’ve had a great dealof opportunity to try various CASapplications.disagrecI II Istrongly quiteslightly neither slightlyquite stronglyIn my organization, one seesCAS on many desks.disagreeI Iagreeagreeagreeagreestrongly quite slightlyneither slightly quite stronglyU-14 My bossdoes not require me to use a CAS.stronglydisagree)<}I I II I Iquite slightlyneither slightly quitestronglyIIU-17I would have no difficulty tellingothers about the resultsof using a CAS.disagrecfI II II Istrongly quiteslightly neitherslightly quitestronglyI know where I can go to satisfactorilytry out various usesof a CAS.disagreestrongly quiteslightly neitherslightly quitestronglyPeople in my organizationwho use a CAS havemore prestige thanthose who do not.disagrccI II II Istrongly quiteslightly neitherslightly quitestronglyU-18 Although it mightbe helpful, usinga CAS is certainly notcompulsory in myjob.disagree‘< II II II Iagreestrongly quiteslightly neitherslightly quitestronglyU-19 My using a CASrequires a lot of mentaleffort.‘xdisagreestrongly quiteslightly neitherslightly quitestronglyUsing a CAS is oftenfrustrating.IIdisagreeI III Istrongly quiteslightly neitherslightly quitestronglyPeople in my organizationwho use a CAS havea high profile.disagreestrongly quite slightlyneither slightlyquite stronglyA CAS was availableto me to adequatelytest run variousapplications.disagreesII Istrongly quiteslightly neitherslightly quitestronglyagreeU-15U-16-1165agreeagreeagreeU-20U-21U-22agreeagreeagree166U-23 I believe I could communicate to othersthe consequences of using a CAS.xdisagreestrongly quite slightly neitherslightly 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 III>cI Iagreestrongly quite slightly neither slightly quitestronglyU-26 Using a CAS improves my job performance.disagreciI I Istrongly quite slightly neitherU-27 CAS are not very visible in my organization.disagreei‘<‘ I I I Istrongly quite slightly neither slightlyquite stronglyU-28 Overall, I find using a CAS tobe advantageous to my job.disagreefI I I I IU-29Iagreeagreequite stronglyslightly quite stronglyJagrccjagreagreeU-30strongly quite slightly neither slightly quitestronglyBefore deciding whether to use any CAS applications,I was able to properly try them out.disagreef I > I I II Iagreestrongly quite slightly neither slightlyquite stronglyLearning to operate a CAS iseasy for me.disagreeI I I I I I Iagreestrongly quite slightly neither slightlyquite strongly167U-31 Using a CAS enhancesmy effectivenesson the job.disagreeI L Istronglyquite slightlyneither slightlyquite 8troflglyU-32 Using a CAS fits intomy work style.disagreeI II II IIagreestrongly quiteslightly neitherslightly quitestronglyU-33 I would havedifficulty explainingwhy a CAS mayor may not bebeneficial.disagreeIXI IIagreestrongly quiteslightly neitherslightly quitestronglyU-34 I was permittedto use a CAS ona trail basis longenough to see whatit could do.dsagree_______________strongly quiteslightly neitherslightlyU-35 Using a CASgives me greatercontrol overmy work.I II II II IIquite stronglyneither slightlyquiteI Istrongly quiteslightly neitherslightly quite3tronglyU-37 Havinga CAS is a status symbolin my organization.disagiestrongly quiteslightly neitherslightly quitestronglyU-38 It iseasy for me to observeothers usingCAS in my firm.disagreeI IIagreedisagreestrongly quiteslightlyU-36 Usinga CAS increasesmy productivity.disagrcejIstrongly>cagreeagreeagreeagreeagree-II jstrongly quiteslightly neitherslightly quitestrongly168[NALLY,IN THIS SECTIONWE WOULDLIKE TOASK YOUA FEWQUESONSABOUT YOUR USEOF THE CAS.B-i Overall,my using a CAS inmy job is (placean X on allfour scales):goodIII III ‘‘extremelyquite slightlyneither slightlyquiteextremelyharmfulf II II IXbeneficialextremelyquite slightlyneither slightlyquite extremelywiscjI II II Ifoolishextremely quiteslightlyneither slightlyquite extremelynegativeI II IIpotiveextremely quiteslightly neitherslightly quiteextremelyB-2 Assumingthat any decisionto use the CASis totally upto you, how wouldyou rate yourpotential use ofthe CAS inthe next six months?likelyf ‘ II III IIunlikelyextremely quiteslightly neitherslightly quiteextremely1improbableprobableextremelyquite slightlyneither slightlyquiteextremelyB-3 Approximatelywhen (monthand year) didyou first start usinga CAS beyondany trial of ityoumay have carriedout?10MOWTHYEAR169B-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):increasediI I I I IIdsigmfi- some- marginally same marginally some- significandy what what cantlyB-lO Overall, how hasyour usage of CAS changed in the last six months? (Place an ‘X’ in theappropriate column):increasedfI I I Ifdecreasedsignifi- some- marginally same marginallysome- significandy what what candyB-li I have been using a CASfor (place an ‘X’ under the appropriate column):Less than About 1-3 AboutAbout More thanonce per times per once per 2-4 timesonce per once perNot at all month monthweek 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 10times I received continuing support from a source external to my firm using my CAS(Place an ‘X’ under the appropriatecolumn for each applicable source, up to a maximum of 10times in total. Totalmay be less than 10.):‘‘1h123 415167.819110Personal friend (nonemployee)Public Accounting finn)(Non-Accountantcomputer consultantOther(please specify)B-15 Currently, if I need help with my CAS, I knowI can get support from the following sources(place an ‘X’ under the appropriate column for each source):none constant1 2 3 some5 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 formy CAS that I receive from thefollowing sources (place an ‘X’ under the appropriate column for each applicablesource):satisfiedunsatisfiedextremely quite slightly neitherjslightly quite extremelyOther personiel frommy companyPersonal friend (non-employee)Public accounting firmNon-Accountantcomputer consultantNoneOther(please specify)175B-18 Howeffective 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 applicablesource):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 yourcurrent 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 finnXNon-Accountantcomputer consultantOther(please specify)THANK YOU FOR YOUR PERSEVERANCE AND COOPERATION SO FAR. NOW,PLEASE SKIP TO THE FINAL SECTION, SECTION D.SECTION D176In this lastsection, we wouldlike to ask you some questions about yourself. Remember, all answers aieconfidential,and no respondent can beidentified, so please give as candid a response as possible.FIRST, WE WOULDLIKE YOU TO ONCE AGAIN INDICATEAGREEMENT ORDISAGREEMENT WITH ANUMBER OF STATEMENTS; THISTIME ABOUTYOURSELF. PLEASEPLACE AN ‘X’ IN THE APPROPRIATE SPACE.1-1 I am generally cautious aboutaccepting new ideas.Adisagreeagreestrongly quite slightly neither slightlyquite strongly1-2 I rarely trust newideas until I can see whether the vastmajority of people around me acceptthem.I I I Istrongly quite slightly neither slightlyquite strongly1-3 1 am aware that I am usually oneof the last people in my group to accept something new.disagreeI I I IIagreestrongly quite slightly neither slightlyquite strongly1-4 I am reluctant about adopting newways of doing things until I see them working for peoplearound me.disagreesI IxlI I I Istrongly quite slightly neither slightlyquite strongly1-5 I find it stimulating to be originalin my thinking and behaviour.disagreeI I I I IJagreestrongly quite slightly neither slightly quitestrongly1-6 I tend to feel that the old way ofliving and doing things is the best way.disagreeI I I I IIagreestrongly quite slightly neitherslightly quite strongly1771-7 I am challengedby ambiguities and unsolved problems.disagree_____________________________________strongly quite sightlyneither slightly1-8 I must seeother people using new innovationsbefore I willdisagreefI I Istrongly quite slightly1-9 I am challengedby unanswered questions.disagreeI I Istrongly quite slightly1-10 I often find myselfsceptical of new ideas.disagreciI I Istrongly quite slightlyneitherNEXT, WOULD YOUPLEASE INDICATE HOWLIKELY OR UNLIKELY EACHOFTHE FOLLOWING STATEMENTSARE BY ONCE AGAIN PLACINGAN ‘X’ IN THEAPPROPRIATE SPACE.S-i Most people who are importantto me think I should use the CASin my job.likelyI I I II I Iunlikelyextremely quite slightlyneither slightly quiteextremelyS-2 My close friends think thatI should use the CAS inmy job.likelyI I ‘‘ I IIunlikelyextremely quite slightlyneither slightly quiteextremelyS-3 My co-workers (peers) think thatI should use the CAS in my job.likelyI II I IIunlikelyI I Iagreequite stronglyconsider them.agreeneither slightlyquite stronglyneitherI I_iIslightly quite stronglyagreeagreeI I IIslightly quite stronglyextremely quite slightlyneither slightly quiteextremely178S-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.likelyI I I < I I I Junlikelyextremely 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.likelyI I I I I Iunlikelyextremely 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.likelyI I I‘>I I I Iunlikelyextremely quite slightly neither slightly quite extremely179FINALLY, WE WOULD LIKETO ASK A FEW QUESTIONSABOUT 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 workedin your current department.1yearsP-6 Years you have worked inthis coeçany.I“yearsP-7 What is the hi9hest LeveLof education that you coapLeted? (pLacean ‘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 organizationaLLevel is (place an ‘X’ beside the appropriatecoluT)_________EXECUTIVE/TOP MANAGEMENT________MIDDLE MANAGEMENTSUPERVISORYPROFESSIONAL/EXEMPTIc TECHNICAL/NON-PROFITCLERICALOTHER (pLease specify)GENERAL BUSINESS INFORMATION180F-i Nuiter of: EirçLoyeesAccounting staffFull time______-3Part timeF-2 Annual Sales last year(in thousands of dollars. k=1,000).< $250kI$250k-$SOOkI$500k-$1,000kF-3 Type of organization(eg. profit,non-profit, CO-OP,etc...)\Jj tF-4 Does your firm planto iirçilement or expanda CAS in the nexttwo years?(Yes or No)Yes NoDon’t Know‘7If Yes, approximatelyhow nich do youexpect to spend on theCAS in this time?$F-5 THANK YOU VERYMUCH FOR YOURPARTICIPATION!If you wish to add anycoments or furtherobservations,please use the spacebelow or simpLyattach themtthis page.r$1,000k-$10,000k> $10,000kIIIAPPENDIX 11-B181182AppendixII- BiClientLetter183YOUR FIRM’SLEtTERHEADDear Client:The University of British Columbia has contacted our firm aboutparticipating in a study on InformationTechnology (iT). They have also requested permission to contact our clientsin order to ask you toparticipate in the study.Our firm has met with the researchers fromUBC to find out more about the nature of the study. Webelieve that the results from this studywould be important to both our finn and to our clients’ helpingus to manage the new forms of IT that will be introducedinto firms like yours over the next few years(and beyond).We would like to encourage you to participate in this studyand fill out the enclosed questionnaire(s).You may find more than one questionnaire with the enclosed material. Please distribute aquestionnaireto the owner/manager, the chief accountant,and to any other accounting staff members interested inparticipating. Also, please use the enclosed return envelope tomail the completed questionnaires.Confidentiality is assured and will be maintained in two ways:1. Your responses cannot be traced back to your firm as theUBC researchers do not and will nothave access to your name or addrçss (unless you specifically include this informationon thequestionnaire). All mailings are handled by our firm.2. Since you will be mailing the completed questionnaire back to theUBC researchers, no personnelfrom our accounting firm will have access to your responses.If you have not been provided with enough questionnaires, pleasecall our office or photocopy sufficientadditional questionnaires.If you have any questions about this study please contact the UBC researchers at thephone number onthe attached letter.Our firm is not sponsoring or otherwise associated with either the research study or theUBCresearchers.184AppendixII- B2Partner LetterFaculty of CommerceAlbert S. Dexter185and Business AdministrationAssociate Professor2053 MainMallManagement Information Systems_______Vancouver, B.C.Canada V6T 1Z2Telephone: (604) 822-8380Fax: (604)822-8489September 13, 1991Dear Sir/Madame:We are conductinga study at the Universityof British Columbia onInformation Technology(iT). Wewould like to determinehow IT is affecting SmallBusiness firms.Many firms have installedcomputer systems, whichare a type of IT. Someof these systems have beeninstalled successfullywhile others have not beenvery successful. The purposeof our research is todetermine whatthe difference is betweenfirms that have successfullyinstalled computer systemsandthose that were not sosuccessfully installed. Wewill obtain this informationfrom a questionnaire thatasks respondentstheir opinions aboutusing computers.We hope to use the resultsfrom this study to help ownersand managers make soundbusiness decisionsabout acquiring otheriT in the future. It is undeniablethat firms will be purchasingother IT in thefuture. Technologysuch as Teleconferencing,Networking, Image Processing,Desktop Publishing,Multimedia, etc.,are currently becoming establishedas the newest forms ofIT that many businesses arelooking at to improvetheir competitive position.Over the next five toten years therewill be other if’sthat we can scarcelyconceive as yet (couldyou have imaginedour current if ten years ago?).We would like to get youropinions about using computersby filling out a questionnaire.This will takeapproximately 20 to25 minutes. Your opinion isimportant, whether ornot you currently use a computer,and we would like tohear from you. Please notethat your answerswill be completely confidential,and that anonymity isassured.Once again, the resultsof this research shouldhelp us to better understandwhat people think aboutpersonally usingcomputers. Other studieshave shown that there is a linkbetween what employees thinkand how an organization performs.Thus our results should enableorganizations to bettermanage thespread of computersand other IT. As a tokenof our appreciation, once the studis completed, wewould be pleased to send you acopy of our findings,conclusions and recommendationsif you send usa card indicating your nameand address. We hope to receiveyour completed questionnaireby the endof the week. Please mail it in the envelopeprovided.If you have any questions about this questionnaire,please call Rick Laktinat (604)-270-8953.Thank you for your assistance.Sincerely,Rick LaktinAlbert S. Dexter

Cite

Citation Scheme:

    

Usage Statistics

Country Views Downloads
United States 11 2
India 3 0
China 2 2
Japan 2 0
Portugal 1 0
Pakistan 1 0
Tanzania 1 0
Sweden 1 0
Somalia 1 0
Lebanon 1 0
Indonesia 1 0
Morocco 1 0
City Views Downloads
Unknown 7 2
Ashburn 7 0
Mohali 2 0
Tokyo 2 0
Phoenix 2 0
Lahore 1 0
Mountain View 1 1
Stockholm 1 0
Sunnyvale 1 1
Guangzhou 1 0
Beijing 1 0

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

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:
http://iiif.library.ubc.ca/presentation/dsp.831.1-0086714/manifest

Comment

Related Items

Admin Tools

To re-ingest this item use button below, on average re-ingesting will take 5 minutes per item.

Reingest

To clear this item from the cache, please use the button below;

Clear Item cache