xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.loc.govstandardsmarcxmlschemaMARC21slim.xsd
leader nam Ka
controlfield tag 001 001787917
007 cr mnu|||uuuuu
008 070514s2005 flu sbm 000 0 eng d
datafield ind1 8 ind2 024
subfield code a E14-SFE0001432
Klaus, Timothy Paul.
An examination of user resistance in mandatory adoption of Enterprise Systems
h [electronic resource] /
by Timothy Paul Klaus
[Tampa, Fla] :
b University of South Florida,
ABSTRACT: User resistance is an important issue in the implementation of an Enterprise System (ES). However, despite the prevalence of user adoption literature, user resistance literature is scarce. Although some studies have conceptualized user resistance as the opposite of user adoption, a mandatory, role-transforming system such as an ES clearly shows that users may use a system while resisting it. Although this area is highly relevant, it is theoretically underdeveloped. This study examines user resistance at the individual level of analysis to determine the underlying reasons for user resistance, the types of resistant behaviors, and the management strategies to minimize resistance. It also seeks to understand the types of users that exist during an implementation and in particular, the groups of resisters. This dissertation identifies four categories of reasons for user resistance, which comprise a total of twelve reasons for user resistance. Resistant behaviors are also identified and classified. Three categories of management strategies are also identified, comprising a total of eight management strategies that are useful in minimizing user resistance. Groups of ES users are also described and examined.
Dissertation (Ph.D.)--University of South Florida, 2005.
Includes bibliographical references.
Text (Electronic dissertation) in PDF format.
System requirements: World Wide Web browser and PDF reader.
Mode of access: World Wide Web.
Title from PDF of title page.
Document formatted into pages; contains 199 pages.
Adviser: James Ellis Blanton, Ph.D.
x Business Administration
t USF Electronic Theses and Dissertations.
4 0 856
An Examination of User Resistance in Mandatory Adoption of Enterprise Systems by Timothy Paul Klaus A dissertation submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Information Sy stems and Decision Sciences College of Business Administration University of South Florida Major Professor: James Ellis Blanton, Ph.D. Richard Will, Ph.D. Anol Bhattacherjee, Ph.D. Robert Fuller, Ph.D. Date of Approval: December 8, 2005 Keywords: User Acceptance, ERP Systems, Resistant Behaviors, Mandatory Adoption, User Types Copyright 2006, Timothy Klaus
ACKNOWLEDGEMENTS I would like to first of all thank and dedicate this work to my wife Susan for enduring this dissertation with me. You have been a great encouragement and support and I look forward to the rest of this j ourney of life with you by my side I would also like to thank God for changing my life and giving me a new start, and for providing strength and health throughout this PhD process. There are many others who also have s upported me through this process. In particular, Ellis Blanton, Rick Will, Anol Bh attacherjee, and Robert Fuller Â– thanks for your valuable advice, your support, your quick feedback, and working with me longdistance to finish the PhD. Furthermore, Michael Harris Â– thanks for imparting your knowledge and wisdom to me in our many c onversations throughout the program. Last of all, I would like to tha nk the ISDS faculty and my Ph D comrades for all the support and our many conversations.
i TABLE OF CONTENTS LIST OF TABLES.............................................................................................................iii LIST OF FIGURE..............................................................................................................vi ABSTRACT......................................................................................................................v ii CHAPTER I. INTRODUCTION........................................................................................1 Motivation..................................................................................................................... ..1 Nature of ES Implementations....................................................................................5 Overview of ES Studies..............................................................................................7 Research Questions.........................................................................................................9 Conceptual Model.........................................................................................................14 CHAPTER II. REVIEW OF LITERATURE...................................................................17 User Resistance.............................................................................................................17 User Acceptance vs. User Resistance.......................................................................18 Three Approaches that Explain User Resistance......................................................24 Reasons for User Resistance.....................................................................................30 Reasons for Resistance to Organizational Change...................................................33 User Resistance Behaviors............................................................................................36 Management Strategies to Minimize User Resistance..................................................39 General Management Strategies to Minimize Resistance........................................40 The Motors of Change..............................................................................................44 CHAPTER III. STUDY 1.................................................................................................54 Epistemology................................................................................................................54 Methodology.................................................................................................................56 Step 1: Expert Panel..................................................................................................56 Step 2: In-depth Case Study......................................................................................58 Step 3: Semi-structured Interv iews in Two Organizations.......................................66 Reliability/Validity.......................................................................................................68 Results........................................................................................................................ ...72 Types of Users..............................................................................................................77 CHAPTER IV. STUDY 2.................................................................................................80 Methodology.................................................................................................................80 Overview of Q-Methodology....................................................................................81 Application of Q-Methodology.................................................................................87 Overview of Q-Methodology Steps..........................................................................88 Step 1: Questionnaire Development.........................................................................90 Step 2: Pilot Data Collection.....................................................................................91
ii Step 3: Full data collection.......................................................................................91 Step 4: Analysis of Pilot/Full Data Collection..........................................................92 Reliability/Validity.......................................................................................................93 Results of Pilot Study....................................................................................................94 Results of Full Data Collection.....................................................................................96 Preliminary Tests......................................................................................................96 Research Question 4a: User Groups.......................................................................101 Research Question 4b: Resisting Groups................................................................105 Research Question 4c: Management Strategies......................................................106 CHAPTER V. CONCLUSIONS A ND FUTURE DIRECTION....................................109 Discussion of Study 1.................................................................................................109 Comparing the Results to Other Studies.................................................................110 Managing the Reasons for User Resistance............................................................121 Discussion of Study 2.................................................................................................128 Contributions...............................................................................................................131 Limitations..................................................................................................................13 3 Future Research..........................................................................................................135 REFERENCES...............................................................................................................137 APPENDIX A: SAMPLE QUOTES FO R REASONS FOR RESISTANCE................154 APPENDIX B: SAMPLE QUOTES FOR RESISTANT BEHAVIORS.......................160 APPENDIX C: SAMPLE QUOTES FO R MANAGEMENT STRATEGIES...............163 APPENDIX D: CODING SCHEME..............................................................................170 APPENDIX E: INTERVIEW SCRIPT..........................................................................172 APPENDIX F: Q-METHOD OLOGY QUESTIONNAIRE...........................................174 APPENDIX G: SAMPLE OF CONS ISTENT/INCONSISTENT CODING.................179 APPENDIX H: CONCOURSE STATEMENTS............................................................185 APPENDIX I: QUOTES FROM QUEST IONNAIRE FOR EACH GROUP................188 APPENDIX J: DEMOGRAPHICS OF QUESTIONNAIRE RESPONDENTS............192 ABOUT THE AUTHOR...................................................................................END PAGE
iii LIST OF TABLES Table 1: Studies identifying ES Critical Success Factors..................................................11 Table 2: A Sample of Studies Examining System Implementations.................................23 Table 3: Reasons for User Resistance................................................................................31 Table 4: Non-IT Studies Examini ng Resistance to Change Reasons................................35 Table 5: A Classification of Types of User Resistance.....................................................36 Table 6: Change Leadership..............................................................................................42 Table 7: Managing User Resistance..................................................................................43 Table 8: Six Basic Building Bl ocks in Explaining Change...............................................47 Table 9: Applicability of the Motors of Change................................................................52 Table 10: Comparison of the Positivist and Interpretivist Epistemologies........................55 Table 11: Identification of Reasons for Resistance...........................................................63 Table 12: Identification of Resistant Behaviors.................................................................64 Table 13: Identification of Management Strategies to Minimize Resistance....................65 Table 14: CohenÂ’s Kappa for Coding of Step 3 Interviews...............................................69 Table 15: Reasons for User Resistance..............................................................................74 Table 16: User Resistance Behaviors................................................................................75 Table 17: Management Strategi es to Minimize Resistance...............................................76 Table 18: Quotes regarding different types of users..........................................................78 Table 19: Description of Steps...........................................................................................88 Table 20: T-test for Equality of Means among Two Questionnaire Versions...................98 Table 21: T-test for Equality of M eans among Respondents who filled out the Questionnaire fully versus those who did not.............................................99
iv Table 22: Principal Components Factor Analysis...........................................................100 Table 23: Factors of User Groups (Normalized Factor Scores and Statement Rankings).......................................................................................102 Table 24: Analysis of Vari ance based on Factor Grouping.............................................105 Table 25: Rank Ordering of Management Strategies......................................................107 Table 26: Rank Ordering of Reasons for User Resistance and Resistant Behaviors........................................................................................................108 Table 27: Comparison with Hirschheim and Newman (1988) Â– Reasons.......................111 Table 28: Comparison with Markus (1983) Â– Reasons ..................................................116 Table 29: Comparison with Kotter and Schlesinger (1979) Â– Management Strategies.................................................................................................... .....118 Table 30: Comparison with Venkatesh et al. (2003).......................................................120 Table A-1: Sample quotes fo r reasons for resistance.......................................................155 Table B-1: Sample Quotes for Resistant Behaviors........................................................161 Table C-1: Sample Quotes for Mana gement Strategies to Minimize Resistance.................................................................................................. ...164 Table G-1: Examples of Consistent Coding Â– Reasons for Resistance...........................180 Table G-2: Example of Inconsistent Coding Â– Reasons for Resistance..........................181 Table G-3: Example of Consiste nt Coding Â–Resistance Behavior..................................182 Table G-4: Example of Consistent Coding Â– Management Strategy to Minimize Resistance.....................................................................................183 Table G-5: Example of Inconsistent Coding Â– Management Strategy to Minimize Resistance.....................................................................................184 Table H-1: Concourse Statements for Reasons for User Resistance...............................186 Table H-2: Concourse Statemen ts for Resistance Behaviors..........................................187 Table H-3: Concourse Statements for Management Strategies.......................................187 Table I-1: Quotes from Questionnaire Respondents.......................................................189 Table J-1: Gender.............................................................................................................1 93
v Table J-2: Education........................................................................................................193 Table J-3: Age................................................................................................................. .193 Table J-4: Position...........................................................................................................1 94 Table J-5: Employees in Organization.............................................................................194 Table J-6: Industry of Employer......................................................................................195 Table J-7: Scope of OrganizationÂ’s System.....................................................................195 Table J-8: System Vendor................................................................................................196 Table J-9: Statistics fo r Numeric Demographics.............................................................196 Table J-10: ES Modules Used by Respondents...............................................................197
vi LIST OF FIGURES Figure 1: Conceptual Model..............................................................................................15 Figure 2: System Conversion Contexts..............................................................................22 Figure 3: Aspects of User Resistance................................................................................28 Figure 4: Resisting and Supporting Behaviors..................................................................38 Figure 5: Methodology......................................................................................................89 Figure 6: Categorization of Re sistance by Factor Number................................................96 Figure 7: Resistant Beha viors by Group Number............................................................106
vii An Examination of User Resistance in Mandatory Adoption of Enterprise Systems Timothy Paul Klaus ABSTRACT User resistance is an important issue in the implementation of an Enterprise System (ES). However, despite the prevalence of user adoption literature, user resistance literature is scarce. Although some studies have conceptualized user resistance as the opposite of user adoption, a manda tory, role-transforming system such as an ES clearly shows that users may use a system while re sisting it. Although this area is highly relevant, it is theoretically underdeveloped. This study examines user resistance at the individual level of analysis to determine the underlying reas ons for user resistance, the types of resistant behaviors, and the manageme nt strategies to mini mize resistance. It also seeks to understand the t ypes of users that exist during an implementation and in particular, the groups of resisters. This disse rtation identifies four categories of reasons for user resistance, which comprise a total of twelve reasons for user resistance. Resistant behaviors are also identified and classified. Th ree categories of management strategies are also identified, comprising a total of eight mana gement strategies that are useful in minimizing user resistance. Gr oups of ES users are also described and examined.
1 CHAPTER I. INTRODUCTION User resistance is an important, yet relatively understudied domain in system implementations. In particular, user resist ance generally is exhibited during and after large system installations that affect the way users perform their jobs. This chapter first describes the motivation behind examining this area, which includes a description of the context and overview of the area. Next, the research questions for this dissertation are identified and described. Third, the expect ed contributions of this dissertation are identified. Motivation Enterprise Systems (ESs) are software packages used for integrating and managing business processes across organizatio nal activities and are widely deployed in organizations from numerous industries. ESs refer to commercial so ftware packages that enable the integration of bus iness processes and transacti on-oriented data throughout an organization (Markus, Axline, Petrie and Tani s 2003). They include organizational-wide software such as Enterprise Resource Pla nning (ERP) systems, scheduling, customer relationship management, product configurati on, and sales force automation (Markus and Tanis 2000). Not only are increasing numbers of organizations installing full ESs or ES modules, but also organizations currently using ESs are expanding their use. AMR
2 Research estimates that the ERP market alon e will grow to 31.4 billion by 2006, at a rate of 10 percent annually (Surmacz 2002). ESs have evolved from produc tion planning and control to integrating a ll parts of an organization with suppliers and customers. Typically, an ES has a suite of software modules available for business functions, su ch as inventory management, accounting, scheduling, and forecasting. Smaller firms tend to implement several modules or components of modules while larger organizatio ns more often install a larger number of available modules (Chalmers 1999; Ferman 199 9). The integration of these functions provides management with tools to better monitor and plan for changing business conditions. A clear benefit of ESs is the level of interoperability that allows for improved management decision-making and monitoring that is expensive or difficult to attain with custom-built systems. Another benefit is th at ES vendors often mode l their software after Â“best practicesÂ” and thus an organizationÂ’s business proce sses can be improved through alignment with these practices Markus et al. (2000, p. 180) describes 23 technical and business reasons as to why organizations choos e to adopt ESs. Interestingly, the reasons for adoption that are listed benefit the orga nization, such as provi ding a greater business profit, but do not directly benefit end-users. ESs have gained credibility as their wide spread implementations have led to the creation of more stable and adaptable syst ems and improved management tools. Through removing inefficiencies in business proce sses, ESs have led many organizations to greater profitability. In fact Hitt, Wu, and Zhou (2002) found that financial markets have
3 consistently rewarded adopters through an increased market valuation. ES implementations are important as they are pervasive, ongoing and require a fair amount of compliance as well as job transformation. These large-scale proj ects require changes that upset the status quo of individuals in the organization. Successful implementations remain a daunting issue as numerous articl es report on implementation catastrophes as well as implementations that have failed to pr ovide projected benefits (i.e., Bingi, Sharma and Godla 1999; Robey, Ross and Boudreau 2002). Many projects cover spans of multiple years and incur millions of dollars yet yield poor results (Stein 1999; Dryden July 27, 1998). One important reason for th is is user resistance (Jiang, Muhanna and Klein 2000). User resistance is an important issue in ES implementations and has been said to be Â“at the root of many enterprise softwa re project failuresÂ” (Hill March 26, 2003, p. 1). For example, Callahan (2002) found a significa nt amount of user re sistance even after nine months of ERP integration testing, part ly due to the many in terfaces with existing systems. Maurer (2002) finds that the reason for low ES re turn on investments is user resistance. Hines (2002) notes th at since end user resistance of ten is cited as an important cause of organizations failing to achieve projected benefits, PeopleSoft, an ES vendor, purposely made user-related improvements in version 8.8. Furthermore, a report on 186 companies that implemented the SAP ES f ound that resistance is the second most important contributor to time and budget ove rruns and is the fourth most important barrier to SAP implementation (Cooke a nd Peterson 1998). Additional studies also
4 reveal how usersÂ’ resistance causes ES implementation failures (Krasner 2000; Wah 2000; Robey et al. 2002; Umble and Umbl e 2002; Barker and Frolick 2003). Although user resistance is an important i ssue, especially in ES implementations, Marakas and Hornik (1996) point s out that Â“few theoretical foundations currently exist in the literature for explaining user resistanceÂ” (p. 209). Although studies in other fields have examined resistance to change, the concep t of user resistance st ill lacks a theoretical underpinning as to its cause. Yet, it is important for management to understand user resistance since it indicates an underlying problem with an implementation. Although there are some IT studies which describe us er resistance (i.e., Jia ng et al. 2000; Shang and Su 2004), IT studies have focused much more on user acceptance rather than user resistance. This is understandable as many types of systems or technologies have voluntary acceptance and thus user resistance is not an issue. Unfortunately for ES research, user acceptance models fail to ac count for the mandator y and job transforming nature of ES implementations. Although there is a lack of theoretical foundations, user resistance remains an important and relevant issue faced by numerous organizations. User resistance must be reduced in order to reap efficiency benef its, particularly for systems that transform business processes such as ESs. As an ES is used to transform an organization by fundamentally changing business processes, us er resistance can greatly affect an ES implementation. A model of user resistan ce could lead to improved implementation strategies and results.
5 Nature of ES Implementations One of the major benefits of an ES im plementation is efficiency achievements through process reengineering. However, the pr ocess reengineering also can be a catalyst for user resistance. An ES is only a tool, yet as a master craftsman uses a chisel to carve a piece of wood, management can use an ES to chip off the inefficiencies from the organizational processes. While other syst ems may only automate existing processes, effective ES use not only changes organiza tional technologies, but through redesign, fundamental business processes are transfor med. Cooper (2000) found that IT can be used as an effective reengi neering tool, although the approp riate creative organizational climate is required. There are several key differences between an ES implementation and other types of system implementations. First, ESs requi re mandatory usage throughout all affected levels of the organization. Mandatory usage is necessary for the system to integrate the data and produce organizational snapshot a nd trend analysis reports. Second, an ES implementation generally results in the reengi neering of jobs, often requiring changes in job tasks and reward structures. A clear bene fit of and reason for ES implementation is the efficiency gains through process reengi neering and thus these changes are made during the system implementation. Third, in or der to minimize cost and time of future upgrades, standardized modules are only par tially customized for employees as opposed to a full customization that may be perf ormed for software produced in-house. Customization is only minimally performed si nce every upgrade that an ES software
6 vendor delivers also needs to be customized, t hus increasing both initial and future costs. Due to these increased costs, managers are discouraged from making modifications unless they are absolutely necessary. Because of the three contextual differen ces noted above, the end-userÂ’s perceived usefulness and perceived ease of use is not a priority; rather, the goal of implementation is to achieve efficiencies through reen gineered processes and provide better organizational reports to managers for impr oved decision-making. This inherent nature of ES revolves around the business processe s, not the user, and can both breed and proliferate resistance. A business process has been defined as Â“a set of logically related tasks performed to achieve a defined business outcomeÂ” (Wu 2003, p. 2). Business process reengineering received much a ttention around 1990 (Hammer 1990; Hammer and Champy 1993) by both systems and business pe ople and has been defined as Â“The fundamental rethinking and radical design of business processes to achieve dramatic improvement in critical, contemporary measur es of performance such as cost, quality, service, and speedÂ” (Maurer 2002, p. 2). R eengineering can entail eliminating or transforming organizational processes and ch ange the way transactions are performed with suppliers and customers. ESs are not ne eded for reengineering, but one main benefit of an ES is the process reengineering that occurs as the technology is implemented. Employees can be greatly affected by the job transformation caused by the ES implementation. This transformation is ofte n difficult, as found in Alvarez and Urla (2002), which suggests that users have values, work habits, and dilemmas that carry over
7 and challenge the new system. This readjust ment usually causes a temporary reduction in performance (Hitt et al. 2002), and unresolv ed resistance can cause a much greater problem (Jiang et al. 2000). Because of this transformation, along with the other previously identified characteristics and features of mandatory, role-transforming systems, studies examining other types of sy stems may not be applicable in explaining the response of users in an ES context. Ye t it is vital for management to not only have employees use the system, as resistance can devastate the implementation, but also embrace the system in order to reap the full be nefits. In regards to resistance, Ross and Vitale (2000) describes how resistance took pl ace in many forms since some usersÂ’ jobs significantly changed, some lost power, and mo st had to unlearn as well as relearn. Essentially, an ES implementation requires or ganizational change, which often alters the tools, skills, rewards, tasks of the job, or ganizational structures, and even beliefs and values. Overview of ES Studies As described in the previous paragraphs ES implementations necessitate some degree of organizational change. As these sy stems often are vital to an organizationÂ’s long-term success, understanding the nature of us er resistance is important. A research stream on ESs has developed in the last se veral years because of their importance to organizations. These systems are important to study not only because of their contextual differences but also because of the following: 1) ES implementations are very costly; 2) there have been many ES failures; 3) an ES is a long-term investment made to increase
8 efficiencies and provide be tter management tools necessary for many organizations operating today. Trade publications have featured many ES-related articles, but despite the importance of ESs, Esteves and Pastor (2001) notes that academic research publications on the topic have only started appearing recen tly. In regards to academic research, Robey et al. (2002) identifies two streams of ES variance research: studies focusing on antecedents to success and studies examini ng a succession of ES-related events. For example, variance research generally include s an antecedent research stream of critical success factors for ES (Esteves and Pastor 2001). Studies examining a succession of ESrelated events include process model rese arch, such as 3-stag e (Bhattacherjee 2000; Gosain 2004; Gattiker and Goodhue 2005), 4-stage (Dong 2001; Gosain 2004), and 5stage (Martinsons and Chong 1999; Ross and Vita le 2000) models. Publications that fit in this stream are often cas e studies and interviews to understand the processes through which an organization traverses. There are also other issues related to ES implementations that are highlighted in the liter ature, such as the transition of power that occurs through the process changes (Sia, Tang, Soh and Boh 2002) and the potential problem of misalignment between organizationa l structures and an ES package (Soh, Sia, Boh and Tang 2003). The number of ES-related research publica tions has increased in the last several years. However, in spite of the recent in crease in these publications, few studies have examined user resistance in the ES context. This is of particular interest not only because
9 of the wide-spread uses of ESs, but because studies show that the type of technology affects the type of resistance (Kendall 1997; Jian g et al. 2000). Thus, the reasons for user resistance to an ES implementation likely differ from user resistance to other types of systems. Also, the ES literature lacks st udies focused on the individual and the change that is faced by employees through the implemen tation of an ES. Furthermore, there is a lack of studies that examine management st rategies in minimizing user resistance. Research Questions Due to reports that address the failure of ES implementations and the importance of minimizing user resistance (i.e., Kras ner 2000; Callahan 2002; Maurer 2002; Hill March 26, 2003), a better understanding of us er resistance is needed. In previous paragraphs, the context of ES implementations has been described. As user resistance is an issue affecting most ES implementations, it is important to examine this area. All resistance does not hurt an organization and th ere are valid reasons as to why users both passively and actively resist system/software implementa tions (Keen 1981); in fact, Fiorelli and Margolis (1993) argue s that some level of resistan ce can be beneficial to the organization as it may draw attention to problems in the change and to address unresolved system issues. Th ere is a difference between re sisting a system that an employee believes will make the organization worse off and resisting due to selfish ambitions; however, in either case, the em ployee is resisting a nd thus hindering the implementation of the system. Whether or not re sistance is beneficial in specific cases, it must be addressed so that proposed changes can either be effectively implemented or modified. Understanding the reasons as to why users resi st can help in identifying
10 important underlying issues that will ultimate ly help to bring a greater degree of longterm success for the company. Following are the first two research questions: 1) Why do users resist an ES implementation? 2) How does user resistance manifest itself in an ES implementation? From a managerÂ’s perspective, it is important to understa nd user resistance so that strategies can be implemented to minimize user resistance. For example, through understanding the underlying reasons for user re sistance, managers can make appropriate modifications to a rollout plan. These stra tegies are important, yet few studies examine management strategies in minimizing user resistance. However, managementÂ’s perspective has been shown to be important, e xhibited through Table 1 of ES studies that have identified critical success factors. Th ese studies identify tangible critical success factors upon which management can bu ild ES implementation strategies.
11 Table 1: Studies identifying ES Critical Success Factors Source Type Critical Success Factors Rao (2000) Case Study Infrastructure resources planning, local area network, servers, PCs, training facilities, human resources planning, education about ERP, commitment to release the right people, top managementÂ’s commitment, commitment to implement Â“vanilla versionÂ”, wellworking manual systems, strategic decision on centralized versus decen tralized implementation. Gupta (2000) Survey of ERP companies Securing top management commitment, forming crossfunctional task forces assessing hardware requirements, deploy the system step-by-step rather than all at once, early planning for user training and support, streamlining decision making to move implementation quickly, and being patient as ERP implementation takes time. Cissna (1998) Interview Top management support, heavy involvement of users, assignment of best people to implementation teams Stratman and Roth (2002) Survey of ERP users Strategic IT planning, executive commitment, project management, IT skills, business process skills, ERP training, learning, change readiness, and improved business performance Nah and Lau (2001) Literature Review ERP teamwork and composition, top management support, business plan and vision, effective communication, project management, project champion, appropriate business and legacy systems, change management program and culture, business process reengineering an d minimum customization, software development, testing and troubleshooting, monitoring and evalua tion of performance Akkermans and Van Helden (2002) Surveyed managers to identify the top 10 CSFs Top management support, pr oject team competence, interdepartmental coope ration, clear goals and objectives, project management, interdepartmental communication, management of expectations, project champion, vendor support, and careful project selection. Willcocks and Sykes (2000) Multiple Case Studies Business themes, new business model and reengineering drives tech nology choice, senior-level sponsorship, championship, support and participation, "dolphin" multifunctional teams, time box philosophy, regular business benefits, CIO as strategic business partner, nine core IT ca pabilities retained/being developed in-house, inhouse and insourcing of technical expertise prefe rred, supplier partnering-strong relationships and part of team, ERP perceived as
12 business investment in R&D and business innovation rather than primarily as a cost-efficiency issue. Although the critical success factors revolve around issu es that management can control, it is interesting to note from this table that each study has identified different critical success factors. This may be becau se of the different contexts in which these systems are studied as well as different resear ch participants that identified the critical success factors. The identification of success f actors is not new. In fact, general systems implementation literature has identified a number of factors that management can manipulate which can ultimately affect the su ccess of an implementation. Some of these factors include politics (M arkus 1983), user involvem ent (Blake and Olson 1984; Baronas and Louis 1988; Barki and Hart wick 1989), communications between developers and end users (De Brabander a nd Thiers 1984), end-usersÂ’ expectations (Ginzberg 1981), and end-user attitude (Robey 1979). Larsen (2003) identified several hundred antecedents of information system success. Many of these antecedents, however, are based only specific contexts a nd are not relevant to the ES context. In spite of the numerous f actors identified that management can influence or control, many systems still fail completely or fail to provide the anticipated benefits. Since there are different technol ogies that are used and thes e systems are implanted into different organizational cultures and structur es, there are many reasons why a system may fit well into one organization yet fail in another organization. Since this paper is focused on user resistance of an ES, the management strategies and factors that
13 management influence are examined regarding their effectiveness in minimizing the level of user resistance. Management can increase the probabili ty of a successful system through proactively enacting management strategies to minimize user resistance. If there are unresolved issues, resistance will remain and thus management should strive to identify those issues and respond appropriately. Through enacting effective strategies, management could decrease the level of resist ance. Thus, the third research question is: 3) In the ES context, what management st rategies are effective in minimizing user resistance? A number of studies have identified user groups. Fo r example, Jurison (2000) found that perceptions of technology and adop tion rates varies among types of users. Zhang and Han (2005) also examines different types of users and found that there are differences among stereotyped groups. Ranc hhod and Zhou (2001) identifies sets of user patterns among Internet users. Furthermore, Chen and Chen (2005) derive profiles of types of users in a recommendation system. It is very likely that in the implementation of an ES, types of users also exist. In regards to user resistance, it is also very likely that types of resisters exist. To further understand user re sistance, it is important to understand these groups.
14 There may be some groups that are s upportive of the system while others are resistant. Through understanding the types of users that exist, management will be more prepared to set forth effective strategies to minimize resistance. Furthermore, an understanding of the types of users also le ads to an understanding of why users are supportive of the system. A fourth research question is proposed to understand the types of users that have similar resisting viewpoints. 4) What types of users exist in an ES implementation? a) What are the characteristics of these naturally occurring types of users? b) What types of resisting us ers exist in an ES implementation? c) What are the management strategies identified by these groups that will be most effective in minimizing th e level of resistance? Conceptual Model This paper revolves around user resistance in an ES context. As mentioned in the beginning of this chapter, user resistance to an ES is an important, yet minimally studied area. The four research questions all revolve around user resistance in an ES implementation and are identified in Figure 1 below:
15 Figure 1: Conceptual Model This chapter has introduced user resistan ce, the nature of an ES implementation, an overview of ES studies, and research questio ns. Chapter II reviews the literature that can be used to address the four research questions. In rega rds to each question, a classification of the literature is provided. Second, approaches are provided for various research streams that can be used as lenses to address the research questions. Since previous studies have not focused on user resistance to an ES, this is an exploratory study to understand th e reasons for user resistance user resistance behaviors, and the management strategies to minimize user resistance. Due to the exploratory nature of this study, a qualitative approach serves best in answering the first three research questions and is described in ch apter three. In chapter four a primarily quantitative study is described which addresses th e fourth research question. This research encompasses a User Resistance Reasons for User Resistance (Why?) User Resistance Behaviors (How?) Managements Strategies to Minimize User Resistance (What to do?) Types of Users (Who?)
16 multi-method approach, similar to the qualitative data collection followed by the quantitative data collection that is u tilized by Koh, Ang, and Straub (2004).
17 CHAPTER II. REVIEW OF LITERATURE ES implementations require a high level of support from people throughout the organization. The lack of support is shown th rough user resistance, which is an important issue faced by top management in the implemen tation of an ES. This literature review first examines the literature on user resistance, including user resistan ce behaviors. Next, studies that address the management stra tegies in minimizing user resistance are examined. User Resistance This paper defines user resistance as Â“usersÂ’ opposition to system implementation.Â” User resistance often re sults from a mismatch between management goals and employee preferences. Studies genera lly have considered resistance to be the flip side of acceptance. However, appare nt acceptance may be masked by passive types of resistance (Marakas and Hornik 1996). For example, non-use of mandated systems, such as an ES system, would only suggest bl atant disregard for management policies and would likely result in sanctions. More likel y, systems would be resisted through covert actions, such as procrastination, Â“forgettingÂ” certain tasks, or slow performance (Marakas and Hornik 1996).
18 In the following paragraphs, the IS liter ature related to system acceptance and resistance is addressed to clarify the di stinction between user acceptance and user resistance and the theoretical roots of these con cepts. In order to gain further insight into user resistance, the literature regarding the vol untariness of a system is also examined and its applicability is accessed. Several well-us ed models and theories are identified and described as to their relevance to user resi stance in an ES implem entation setting, which mandates use. Second, three approaches to us er resistance are desc ribed and evaluated. Third, studies that have examined reasons for user resistance are examined. Fourth, nonIS literature related to resistan ce to change is also brought in to add to the limited studies available in the IS literature. Finally, this section analyzes the studies identifying user resistance behaviors. User Acceptance vs. User Resistance There is an extensive body of research that has focused on system acceptance in voluntary settings. For example, the Theory of Reasoned Action (Ajzen and Fishbein 1980), the Technology Acceptance Model (TAM) (Davis 1989; Mathieson 1991; Taylor and Todd 1995), and the more recent Unifie d Theory of Acceptance and Use of Technology Model (Venkatesh, Mo rris, Davis and Davis 2003) have their roots in the context of voluntary adoption. These studi es have consistently found relationships between beliefs, attitudes, be havioral intentions, and usag e behavior and focused on the initial decision on whether to use a system. An important distincti on between the context of these studies and the contex t of an ES implementation is that voluntary adoption is not an option for ES users. Sin ce the theories noted above were developed in the context of
19 voluntary adoption to explain th e acceptance of an innovation, they are not advantageous to use in studying user resistance in a manda tory context. Furthermore, the focus of studies using these theories revolved more around behavioral intention, and thus the cognitive processes, rath er than actual behavior. Another theory is Innovation Diffusi on Theory (Rogers 1976), which has its foundation in the communication literature and revolves around the spread of an innovation. This theory also ha s its roots in volunt ary adoption and is not applicable to the ES context because of the mandatory nature of an ES implementation. There are several studies that have applied some of the user acceptance models to mandatory settings and have found mixed resu lts. For example, Hartwick and Barki (1994) examined voluntariness as a moderating construct and f ound that the level of user participation and involvement depends on th e level of voluntariness. Bagchi, Kanungo, and Dasgupta (2003) expanded on the Hartwick et al. (1994) model and evaluated user involvement, concluding that there are a number of sources that influence a userÂ’s view. Venkatesh and Davis (2000) also considers ma ndatory adoption, and extends TAM to test its usage in both voluntary and mandatory se ttings, finding significant relationships to support TAM (Venkatesh and Davis 2000). However, more recently, Brown, Massey, Montoya-Weiss, and Burkman (2002) found non-significant results for some of the relationships in the TAM and that attitude is not related to behavioral intention in a mandatory use environment. Although both Br own et al. (2002) and Venkatesh et al. (2000) examine mandatory usage, one difference between the studies is that Venkatesh et
20 al. (2000) does not account for the degree of system integration or the level of accountability. The preceding paragraphs indicate that the level of voluntariness, based on the type of system, affects the le vel of user acceptance or resi stance. In regards to user resistance, it is also likely that the context of the system affects the level of resistance. The previously mentioned studies examine mandatory adoption, but none of the studies revolve around ESs, that by nature are mandato ry and transform jobs. On the surface it appears that the Technology A cceptance Model applies to both voluntary and mandatory contexts. However, in situations such as an ES implementation, not only is the use mandatory, but it often radically transforms the job description/re sponsibilities of the user. Theories such as the Theory of Reasoned Action, the Technology Acceptance Model, the Unified Theory of Acceptance and Use of Technology Model, and Innovation Diffusion Theory are not designed for the ES context. Moreover, these theories and studies are not designed to focus on user resi stance. Consequently, their applicability is limited in considering the issue of user resist ance to ESs. In the following paragraphs, the literature rela ted to user response to system implementations is examined for its applicability to user resistan ce in the ES environment. More specifically, an examination is made regarding how users respond to syst em implementations across system types. The system type may affect the way people respond to a system, yet few studies have examined how the type of system af fects user response. Jiang et al. (2000) investigates resistance across system types and finds that the managers who responded to
21 the survey identified different types of resistance, depending on the system type. Fichman (1992) also notes that there are different types of sy stems, and classifies systems based on locus of adoption (Individual vs. Organization) and cla ss of technology (high knowledge burden or high user interdependenc ies vs. low knowledge burden or low user interdependencies). This classification is useful in identifying different types of systems. However, studies have not shown how the voluntariness as well as the transforming nature of a system affects user resistance. Based on Zuboff (1988) and Schein (1992) there are three general uses for technology: automate, informate, and transfor m the organization. For example, a system can automate existing processes, which can make jobs easier and reduce the costs of operation. It can also be used to informate, which includes providing information to enable a job to be performed, such as improving the collection, processing, and dissemination of information that essen tially improves the way by which a job is performed. Lastly, it can be used to transform which includes redefining the firm and/or supply chain and transforming the tasks performed, the type of job, and the reward structure. Figure 2 suggests six categories of system implementation research based on the necessity of adoption and th e level of process change.
22 Figure 2: System Conversion Contexts Most of the user accepta nce studies have examined system implementations fitting into category 1 and 2, with some studies that fit into category 4 or 5. A category 3 implementation is rare as most organizatio ns require adoption if they reengineer processes. Category 6 implementations occur frequently as ES implementations generally entail reengineered processes. Ho wever, most user acceptance research has not examined the effects of requiring users to use a system that radically transforms their job. Table 2 expands on the categories of Figure 2 by providing a sample of studies that fit into the categories. The sample was selected from a wide variety of studies examining different types of systems and different contexts to identify the types of systems that fit into the six categories. The lack of research involving category 6 system implementations along with few studies that ha ve examined user resistance presents the opportunity for theoretic al development. Voluntary 1 Electronic Brainstorming System 2 Internet Technologies to Support Teaching 3 Mandatory 4 Payroll System CAD system Check Clearing System 5 Inventory Management System Sales Analysis System 6 ERP System Customer Relationship Management System Automate Informate Transform Process Change Necessity of Adoption
23 Table 2: A Sample of Studies Examining System Implementations Study Type of System Variables Method of Measurement Results Comments Category Todd and Benbasat (1999) DSS IV-Cognitive Effort and Incentives DV-Strategy DSS kept track of what user commands Cognitive Effort affects strategy selection Voluntary adoption, students in lab setting 1 Dennis and Reinicke (2004) Electronic Brainstorming IV-Type of Brainstorming DV-Effectiveness Survey, Likerttype scale Some significant differences in outcomes depending on the type of brainstorming used Voluntary, Lab setting 1 Lewis, Agarwal and Sambamurthy (2003) Internet Technologies to support teaching IV Â– institutional factors, social factors, individual factors DV-Ease of Use, Perceived Usefulness Survey, Likerttype scale Institutional and Individual factors affect ease of use and perceived usefulness, but not social factors Voluntary adoption 2 Joshi and Lauer (1998) Computer-Aided Design (CAD) Case Study Â– examines the impact of CAD implementation Qualitative Data Some factors affect user evaluation and acceptance Mandatory adoption 4 Yoon, Guimaraes and OÂ’Neal (1995) Expert System IV-Developer skill, End-user characteristics, Shell characteristics, User involvement, Problem Difficulty, Domain Expert Quality, Management Support DV-User Satisfaction Survey, Likerttype scale All relationships are supported Assumes it is a mandatory adoption since project managers working with expert systems are surveyed 5 Brown et al. (2002) Computer Banking System IV-Perceived Usefulness, Perceived Ease of Use, Perceived Behavioral Control, Subjective Norm DV-Attitude, Behavioral Intention Survey, Likerttype scale Perceived Usefulness affects Attitude, and both Perceived Behavioral Control and Subjective Norm affects Behavioral Intention Mandatory adoption (BPI) 5 Somers, Nelson and Karimi (2003) ERP Performed CFA of End-User Computing Satisfaction: examined Content, Accuracy, Format, Ease of Use, and Timeliness Survey, Likerttype scale The End-User Computing Satisfaction instruments maintains psychometric properties in ERP domain Sampled users of ERP systems (not examining the implementation) 6
24 In the following section, three approaches are described from the IS-literature that have been used to examine user resistance. These paragraphs describe the three approaches. Three Approaches that Ex plain User Resistance Few theoretical perspectives have been o ffered to explain the phenomenon of user resistance. However, for the st udies that have examined user resistance, there have been three general approaches that have been ta ken. These categories were first articulated and developed by Markus (1983) although they have since been expanded. Markus (1983) identifies these three perspectives: 1) system-oriented; 2) pe ople-oriented; and 3) interaction-oriented. Similar to these three approaches is the technological, organizational, and emergent perspectives addressed by Markus and Robey (1988). These perspectives are essent ially lenses through which researchers can investigate issues. For example, Jasperson, Carte and Saunders (2002) used these three lenses to examine the structure between t echnology and organizational power. The first lens, or approach in examining user resistance, is the system-oriented approach which suggests that resistance o ccurs because of technology-related factors such as the user interface, performance, secu rity, ease of use, and degree of centralization (Markus 1983; Jiang et al. 2000). This pers pective is similar to the technological perspective described by Markus and Robey (1988) and is based on the forces over which a user has little control. For example, the technology impacts the way work is done and thus this technological perspective would sugge st that the technology is the cause of the
25 resultant behavior of employees. The problem with approaching user resistance from only this perspective is that the technology affects employees in different ways. For example, studies have found that tec hnology may both centralize and decentralize authority (Klatzky 1970; Brown and Magill 1998), both increase and de crease the level of power (Markus 1983; Dawson and McLaughlin 1986), and fail to produce change even when expected (Robey 1981; Bjorn-Andersen, Eason and Robey 1986; Franz, Robey and Koeblitz 1986). The change that results fr om the technology likely affects the level and type of user resistan ce. However, this approach by its elf is not likely to explain user resistance well. The people-oriented approach suggests user resistance occurs because of individual or group factors such as bac kgrounds, traits, and attitude towards the technology (Markus 1983; Jiang et al. 2000). This is similar to the Â“Organizational ImperativeÂ” described by Markus and Robe y (1988), which proposes that technology is put into place to meet organizational needs a nd thus technology is the dependent variable. This perspective implies that the IT is able to meet both the social and technical needs of the organization. People with this view see IT as a tool used to address organizational problems. Thus, if there is user resistance, it is because of people-related issues, such as the lack of skills and motivation of the employees, or organization-related issues, such as communication and job structures, not because of the technology aspects (Markus 1983; Markus and Robey 1988). This approach is helpful to identify certain types of user resistance, but this approach, by itself, also is insufficient in explaining user resistance in a complex situation such as an ES implementation.
26 The interaction-oriented approach suggest s that perceived social losses caused by interaction between people a nd the technology affects resistan ce, such as changing power relationships, social structure, and job structure (Markus 1983; Jiang et al. 2000). This is similar to the Â“Emergent PerspectiveÂ” o ffered by Markus and Robey (1988). This approach suggests that there are complex intera ctions that affect bot h the uses and results of IT. Furthermore, it sugge sts that identical technologi es can be implemented in different contexts and result in different outcomes from the employees because of the different settings. For example, Silver, Ma rkus, and Beath (1995, p. 367), identifies how the external environment, firm strategies, or ganizational structure, organizational culture, business processes, IT infrastructure, information system features, and the implementation process interact to essentially affect the level of us er resistance. This perspective thus does not lead to simple m odels, but rather because of the complexity, research using this emergent perspective usually requires a rich description of the organizational processes, technology f eatures, and user intentions. The interaction approach is not limited to sociotechnical and po litical issues, but Markus (1983) focuses on thes e two variants of the intera ction approach. The sociotechnical variant suggests that when there is a poor fit between a syst em and the Â“division of laborÂ”, such as the system requiring diffe rent roles and responsibilities or different communication structures, people resist the system. The political variant addresses the interaction between the system and the or ganizational power, and thus systems that control data centrally will be resisted in or ganizations that have decentralized authority
27 structures. Moreover, people who lose power are likely to resist. The implications of this approach is that system implementers, if they believe the resistance is due to the interaction between the organizational cont ext and the system, will try to resolve organizational problems or misfits prior to in stalling the system and implement strategies only after a thorough analysis of the organiza tion. For example, in an IT study, Dawson et al. (1986) found that as an information system was implemented, foremen lost power as the assistants who work with the system gained power. Thus, the attitudes toward the system are affected depending on the relative ga in or loss in power as well as the tasks that are changed in the jobs. Figure 3 below shows a depiction of various aspects of and potential sources for user resistance. As there are different as pects, understanding the underlying reasons for user resistance necessitates the examination of three areas.
28 Figure 3: Aspects of User Resistance Because of the complexity of the ES context, there are technology-determined factors, people-determined factors, and inter action-determined factors that are likely to influence users. In regards to examining the first two research que stions, understanding these three approaches to user resistance help s to guide the directi on in investigating why user resistance occurs in an ES implementati on and how user resistance manifests itself. This study takes all three perspectives into account for the following reasons: 1) An ES implementation affects the issues addressed by all three perspectives since it changes the User Resistance People-oriented approach -Backgrounds -Traits -Attitude towards technology -Communication -Job Structures System-oriented approach -User Interface -Performance -Security -Ease of Use -De g ree of Centralization Interaction-oriented approach -Perceived social losses, such as power -Changing social structure and job structure
29 tool people use (technology-a pproach), affects the employees (people-approach), and upsets power relationships, social structures, and job structures (interaction-approach); 2) ES implementations have been shown to be successful in some organizations while disastrous in others. The technology, people, and the interaction between the technology and the organizational structures are an important reason for the success or failure; 3) The nature of an ES implementation is complex. A technology that is used as a tool to alter organizational structures is not just resisted because of a simple issue such as a lack of ease of use; rather, there are underlying, co mplex issues that may even change the paradigms of employees that need to be reso lved in order to redu ce user resistance. Hirschheim et al. (1988) states that the inte raction of various causes of user resistance intertwine to produce a particular instance of resistance, which makes it very difficult or impossible to develop a simple causal relati onship. Furthermore, Hirschheim et al. (1988) describes four sources of resisting att itudes: the individual, the system, the change strategy, and the perceived outcomes of the chan ge. Resistance may not mean that users will resist all systems, but rather the system that is proposed is being resisted. When considering user resistance, it is im portant to note that the users who resist may also be managers. LaNuez and Jermier ( 1994) states that managerial sabotage is increasing and that Â“sabotage with catast rophic potential is becoming an increasing concernÂ” (p. 223). In an ES setting, (Sia et al. 2002) found that even management resisted empowerment as they sought to regain the power lost from the ERP implementation. Managers can be territorial and resist a system due to a number of issues such as losing power or prestige and must be take n into account in the study of
30 user resistance. For example, Ross et al. (2000) points out that managers felt that the computer was controlling them (changing the wa y they run the organization) rather than being a tool. Reasons for User Resistance Because of the mandatory, role-transforming nature of an ES implementation, it is likely that users resist an ES for different reas ons than resisting other types of systems. It is important to understand the underlying reasons for why user s resist an ES implementation. Table 3 below shows reasons for user resistance, as identified in the IT literature. All of thes e articles suggest different reasons for user resistance, which may be because these reasons are suggested based on diffe rent systems within different contexts.
31 Table 3: Reasons for User Resistance Reason(s) for User Resistance Methodology Source Job insecurity None-Opinion Piece De Jager (1994) Users have values, work habits, and dilemmas that usually carries over and challenges the new system Narrative Analysis Alvarez and Urla (2002) Loss of power None-Opinion Piece Keen (1981) No communication channel to address fears or frustration because of some form of penalty for disagreeing with superiors None-Opinion Piece Marakas and Hornik (1996) Loss of status, economic insecurity, interpersonal relationships altered, change in job content, change in decision making approach, loss of power, and uncertainty/unfamiliarity/misinformation None-Reasons are identified, but no explanation is provided regarding their source Hussain and Hussain (1984, p. 391) Misalignment of the ES with the organization, or in other words, an inap propriate level of fit. None-Conceptual Development Gosain (2004) Parochial self-interest (resisting due to losing something of value), Misunderstanding and lack of trust (misconceptions of implications and not understanding the benefits), Different Assessments (Employees see greater costs than benefits while management sees the reverse), Low Tolerance for Change (Employees fear the development of new skills and behaviors), and Increased Efforts (Additional abilities or efforts are required with the change). None-This is a noncomprehensive list based on several Management and IS journals Shang and Su (2004, p. 150) Current habits (level of structure in existing practices) and perceive d risk of adoption (performance uncertainty as well the social, economic, or physical, consequences). None-Mentions that these two constructs seem the most useful in understanding resistance Sheth (1981) Interface can be confusing and difficult, Process changes None-Identifies several articles that mention these reasons OÂ’ Leary (2000) Innate conservatism, lack of felt need, Uncertainty, Lack of involvement in the change, redistribution of resources, organizational invalidity, lack of management support, poor technical quality, personal characteristics of the designer, level of training and education, cognitive style of user None-literature review of resistance to change Hirschheim and Newman (1988) Job security, lack of understanding, human nature None-opinion piece Ainsworth (1977)
32 It is interesting to note from this table th at most of the reasons for user resistance are not related to specific system character istics (i.e., interface); rather, many of the reasons for user resistance are due to the job changes resultin g from the system (i.e., loss of power). Also, only one of these articles bases the identified reas ons on part of their study. The other articles mention reasons for user resistance but di d not actually identify the reasons themselves. Although user resi stance has been mentioned in numerous studies, there has not been a ny study found that actually sought to comprehensively understand why users resist. The previous paragraphs have classifi ed system types and identified user resistance as an important, yet relatively unstudi ed concept. In order to gain more insight into this concept, organizational change litera ture related to resistance is examined in the following paragraphs. Far more literature has focused on the concept of resistance to change, rather than user re sistance. Although the resistan ce to change literature can partly explain user resistance to an ES, the ES implementation is a type of change that requires users to adapt to new processes a nd use a standardized system to enter and retrieve information. Some of the principl es addressed in this research stream are applicable to user resistance to an ES impl ementation. Users may partly be affected by the technology, but there are a number of issues not related to the technology characteristics that also aff ect user resistance. For exam ple, Martinsons et al. (1999) found that a number of nontec hnical factors are associated with smoother organizational change. Orlikowski and Barley (2001) discusses how organization and technology
33 studies have epistemological differences and ma kes the argument that there is much value in having a greater interaction between orga nization and technology st udies. Essentially the interaction of these fiel ds of study can lead to a better understanding of the phenomenon involved in an ES implementation. Thus, the following paragraphs bring in the literature on organizati onal change to shed light on how employees respond to change. Furthermore, this se ction identifies organizational is sues that management needs to address. Reasons for Resistance to Organizational Change In regards to resistance, the management l iterature has defined resistance as Â“the forces against change in work organiza tionsÂ” (Mullins 1999, p. 824). Employees often respond to change with resistance and thus resi stance to change is a well-studied area in the organizational change lite rature; one source describes it as the natural reaction employees have to anything that upsets the status quo (Conner 1993). The change management literature is filled with exampl es of employees resisting change (i.e., Mainiero and DeMichiell 1986; Knights a nd Vurdubakis 1994; Folger and Skarlicki 1999). However, studies have focused more on organizational factors; few studies have examined employeesÂ’ resistance to change at an individual level (Jermias 2001). Resistance to change is important to c onsider; minor resistance can reduce the speed of change while major resistance can ultimately cause management to abandon its plans (Davidson 1994). Doppler (2004) notes that resistance is a normal phenomenon and that ignoring resistance can cause many fu ture problems; alternatively, recognizing
34 resistance and dealing with it appropriatel y can reduce enduring pr oblems. Gravenhorst and in Â‘t Veld (2004) points out that cha nge and resistance go hand in hand; accordingly, change suggests resistance and resistance implies change. Although the conceptualizat ions of resistance identified in the previous paragraphs are useful in identifying the notion of resistance, studies that have identified reasons for why employees resist change are not as useful in unde rstanding resistance because of the inconsistent results. It is no ticeable from Table 4 that these studies, which focus on reasons for user resistance, differ on both the number of reasons as well as the actual reasons for resistance. Perhaps this is due to the differing environments and the types of changes faced by the employees. For example, in an organizational merger, it seems logical that the reasons for resistance to change would be diffe rent than the reasons for user resistance to an ES. Also, just b ecause a reason is identified does not imply that it is the driving force for resistance ev en though it may contribute to resistance.
35 Table 4: Non-IT Studies Examini ng Resistance to Change Reasons Study Reasons for Resistance to Change Kotter and Schlesinger (1979) 1) employees think they will lose something they value, such as a position, power, or relationships; 2) em ployees have a lack of trust in the person or people implementing the change or misunderstandings occur; 3) employees see a greater co st than benefit from the change; 4) employees have a low tolerance for change because of a lack of skills or because it makes them feel uneasy performing new behaviors and working with different relationships Kegan and Lahey (2001) psychological dynamics that occur in employees because of what is called Â“competing commitmentsÂ” Â– thus employees, even though they may want to change, have hidden commitments that compete with the commitment they have toward the change. Ford, Ford and McNamara (2002) resistance may not be in response to a current issue; rather, there are ongoing background conversations that create a context for the level of change initiative as well as the responses to the change. Pardo del Val and Fuentes (2003) myopia, denial, perpetuation of ideas, implicit assumptions, communication barriers, organizatio nal silence, direct costs of change, cannibalization costs, cross subsidy comforts, past failures, different interests among management and employees, environmental changes, resignation, inadequate strategic vision, implementation climate, departmental politics, in commensurable beliefs, deep rooted values, social issues, leadersh ip inaction, embedded routines, collective inaction, lack of capabilities, and cynicism Trader-Leigh (2002) Self-interest (employeeÂ’s interests are not met), Psychological impact (i.e., job security, social impact), Tyranny of custom (organizational culture was too rigid), The Redist ributive Factor (redistributing resources and changing policies), The Destabilization Effect (job role change leads to untrained/ine xperienced employees), Culture Compatibility (incompatibility of ch ange with organizational culture), and the Political Effect (constraint s based on organizational politics) The previous paragraphs have examined the reasons for resistance to change, from the management literature. Although ther e are a plethora of reasons, studies have not focused on the underlying reasons for user resistance in an ES implementation. Moreover, studies have not focused on how user resistance manifests itself throughout the ES implementation. The following paragrap hs describe conceptualizations of user resistance behaviors.
36 User Resistance Behaviors For this study, user resistance behaviors is defined as Â“outward manifestations of opposition to the system implementation.Â” Shang et al. (2004) offers one conceptualization of user resistance behaviors and organizes resistance into three types. This classification and descrip tion is based on several studies identified in their literature review. As shown in Table 5 below, the th ree categories are non-de structive, passivelydestructive, and proa ctively-destructive. Depending on the circumstances surrounding an ES, any three of these behaviors may be manifested through the users, causing implementation problems. Although Kling (1980 ) found that users ofte n resist rationally, it is interesting to note the sp ectrum of behaviors that may ar ise when users resist the ES. Table 5: A Classification of Types of User Resistance Resistance Type Resistance Behaviors Non-destructive Request job transf er or withdraw from the job Increased absenteeism or tardiness Communicate negative feelings to coworkers Passively-destructive Refuse to cooperate with other employees Neglect work assignments Waste time and make minimal effort to improve knowledge or skills Inferior quality performance Dissonance with consultants Proactively-destructive Deliber ately sabotage a work process Make careless mistakes Adapted from Shang et al. (2004, p. 151) One other IT study that clas sifies resistant behavior s is Lapoint and Rivard (2005), based on three cases. The behaviors ar e classified on the following scale: 1)
37 Adoption; 2) Neutrality; 3) Ap athy; 4) Passive Resistance; 5) Active Resistance; and 6) Aggressive Resistance. Othe r than Lapoint and Rivard (2005) and Shang (2004), few IT studies have focused on the types of behavior s manifested through th e user resistance. However, it is important to better understa nd conceptualizations of resistance and types of behaviors that may be exhibited. Thus the following paragraphs draw from the management and psychology literature that revolves around resistance to change. Prior to discussing the behaviors of user resistance, it is important to further characterize user resistance behaviors. It is unclear from some studies what exactly is meant by user resistance behaviors. For exam ple, Piderit (2000) is one conceptualization of resistance to change, a nd suggests that the resistance literature has focused on three somewhat overlapping conceptualizations of re sistance: attitude (i.e., beliefs about the object), emotion (i.e., frustration, anxiety, aggression, feelings in response to the attitude), and behavior (i .e., intentional acts of co mmission or omission). Although Piderit (2000) focused the be havior on intentional acts, fo r this study, it is irrelevant whether or not the act is intentional; rather, this study focuses on any outward manifestations of opposition to the system implementation. Bovey and Hede (2001), a psychology paper, conceptualizes t ypes of resistance behaviors, framing resistance on a continuum with active resistance on one side and passive resistance on the other. Active resistance may enta il expressing opposition to the system through a voicing of an opinion, or a more extreme opposition would be leaving the organization. Passive forms of resistance ar e much harder to detect and deal with and
38 may entail withdrawing from conversations, av oiding the required tr aining, and delaying an implementation. Resistance behaviors ma y also exist on a continuum between overt and covert behaviors. Overt behaviors coul d include making a sta nd against the system and openly obstructing the implementation. On the other hand, covert behaviors are when employees grudgingly use a system, find ways to work around the intended purpose of a system, or sabotage the system to ensure its failure, purposefully misenter data, not use the system for its intended purposes, or complain about the system to coworkers. Figure 4 shows one way that resistan ce behaviors have been classified: Figure 4: Resisting and Supporting Behaviors Overt (openly expressed behavior) Covert (concealed behavior) Active (Originates action) Resistance Oppose Argue Obstruct Support Initiate Embrace Resistance Stall Dismantle Undermine Support Support Cooperate Passive (not acting, inert) Resistance Observe Refrain Wait Support Agree Accept Resistance Ignore Withdraw Avoid Support Give in Comply Adapted from Bovey et al. (2001a, p. 375) and Bovey et al. (2001b, p. 534) A person who resists may exhibit one or more of these behaviors; the value of this classification lies in the identification of the potential types of behaviors.
39 Falbe and Yukl (1992) presents a simila r concept as Bovey and Hede (2001), yet has a different terminology as it different iates resistance from commitment and compliance. When an employee receives a re quest, commitment would result in agreeing and enthusiastically exercising initiative to take positive action on the request. Compliance would result when the employee is apathetic, initiates minimal effort, and does not exert initiative. Re sistance would result when th e employee refuses, argues, delays the response, or seeks to cancel the request. The way Falbe et al. (1992) defines compliance is similar to how Bovey et al. (2001) describes passive resistance behaviors and the way Falbe et al. (1992) describes resistance is simila r to how Bovey et al. (2001) describes active resistance. The previous paragraphs discuss reasons for user resistance and user resistance behaviors. Resistance is a complex phenomeno n, and the context and type of change are likely to influence the type of resistance. The next secti on focuses on the third research question, which deals with the management stra tegies used to minimize the level of user resistance. Management strate gies to minimize user resistance are also a complex issue because of the complexity of the implementation and the underlying causes of user resistance. Management Strategies to Minimize User Resistance As stated previously, an ES implementa tion necessitates change Whether or not the employees are aware of the effects of th eir user resistance, from a management and system implementerÂ’s perspective, it is an is sue that needs to be dealt with in a manner
40 that produces favorable results. Aladwani ( 2001) discusses the need for management to proactively deal with user resi stance rather than reacting when it arises. However, this requires management to understand the nature of user resistance and take appropriate steps, such as appropriately marketing th e ES to employees (Aladwani 2001). Following are the results of studies that have examined management strategies to minimizing user resistance. General Management Strategies to Minimize Resistance An important factor to successful implem entation is minimizing usersÂ’ resistance to change (Marakas and Hornik 1996; Josh i and Lauer 1998). It is important for management to have strategies in place to minimize user resistance. Without adequate strategies, it is quite possible for management to errantly search for the resisters, punish the compliers, and promote the uninvolved. Th e other extreme would be for management to take no action against the resisters, which would also lead to problems. Few IT studies have examined manage ment strategies to minimize user resistance. Jiang et al. (2000) examines ma nagement strategies and identifies twenty general strategies, although thes e strategies are based on resi stance literature. Managers can use this list of gene ral strategies as a checklist for va rious types of systems: Involve employees, Open communication, Provide chan ge info, Initiate moral boosts, Pace conversion, Redevelop modularly, Reward idea s, Document standard s, Clear authority, Upgrade environment, Pilot study, Alter job titles, Show sympathy, Orientation, Job transfers, Separation pay, Hiring freeze, Job counseling, Group therapy, and Retrain
41 employees. However, as with most checklists, all the items are not applicable to every environment; this is demonstrated by the conc lusion of Jiang et al. (2000) that the system type affects the management strategies employed. In the following paragraphs, the non-IT literature that discusses management strategies to minimize resistance is exam ined. Although there are countless potential strategies, Ross et al. (2000) discusses how firms could deal with resistance by providing stock options as incentives. Since there ar e numerous potential management strategies, rather than focusing on individual strategi es related to organizational change, the following paragraphs identify literature th at examines categories of management strategies. There appears to be four general ma nagement approaches that encompass strategies that deal with organizational change. Hersey and Blanchard (1988 p. 340-341) points out two general strategies that manageme nt implement: participative and directive. Dunphy and Stace (1993) includes participative and directive as well as addresses two other general strategies: consul tative and coercive. Participative change strategies are more of a bottom-up approach that involves groups in selecting and formalizing new methods to reach the goals. In a system implementation setting, this may include obtaining user input in the design stage (Fl oyd 1993), user training and testing (Hu, Clark and Ma 2003), and providing a vehicle fo r employees to participate in process improvement (Edosomwan 1996). Directive st rategies are management-directed and include power redistribution (Legare 1995; Go ltz and Hietapelto 2002), financial rewards
42 for learning the system (Lawler 2000), human resource involvement during the adoption process (Martinsons and Chong 1999), and el iminating jobs for users who fail to adequately learn the new system (Mainier o and DeMichiell 1986). Consultative strategies involves providing employees with support and information and only involves employees minimally in goal-setting. Coerci ve strategies involve forcing change on employees, often imposing a threat to non-co mpliers. The four styles of change leadership are described fu rther in the table below: Table 6: Change Leadership Type Styles of Change Leadership Participative This involves widespread participation by employees in important decisions about the organization's future, and about the means of bringing about organizational change. Directive This style of leadership invol ves the use of managerial authority and direction as the main form of d ecision-making about the organization's future, and about the means of bringing about organizational change. Consultative This style of leadership i nvolves consultation with employees, primarily about the means of bringing about organizational change, with their possible limited involvement in goal se tting relevant to their area of expertise or responsibility. Coercive This style of leadership involve s managers/executives or outside parties forcing or imposing change on key groups in the organization. Adapted from Dunphy et al. (1993, p. 920) Shang and Su (2004) is an IT study wh ich identified potential management strategies based on the four styles of leadership identif ied above. The table below describes a number of strate gies that have been used to manage user resistance:
43 Table 7: Managing User Resistance Management Style Management Strategies Directive (Use of managerial authority to effect change) Pace conversion to allow for reasonable readjustment period Document standards so new pro cedures are easy to learn and reference Retrain employees to be effec tive users of the new systems Reward ideas that will improve throughput Clarify job definition before the changeover Alter job titles to reflect increased responsibility Arrange for voluntary job transf ers to avoid users with no interest in new procedures Call a hiring freeze until all disp laced personnel are reassigned Give unions higher wage rates in return for a work rule change Give one of its leaders, or some one it respects, a key role in the design or implementation of a change Participative (Widespread participation by employees on direction and process of change) Involve employees in development of new systems to encourage a feeling of ownership Provide employees with information regarding system changes to preserve ownership Open lines of communication between employees and management Initiate morale boosting activ ities: company parties and newsletters to promote community Consultative (Provide employees with information and moral support) Provide job counseling and orga nize group therapy to help employees adjust Listen and provide emotional support Conduct orientation sessions to prepare for change Be receptive to complaints following conversion to maintain employee contact and trust Provide one-on-one discussions Coercive (Forcing or imposing change on key groups) Implicitly and/or explicitly threaten loss of job and promotion possibilities Fire or transfer peopl e who resist change Shang et al. (2004, p.152). The previous paragraphs discuss general categories of management strategies for dealing with change. The next section deals with the build ing blocks of organizational change. Although the unit of analysis for th is study is the indivi dual, these building
44 blocks are helpful in understanding the pro cess of transition, which is important in determining management strategies to minimi ze user resistance. As there have been studies that have examined the underlying building blocks for change theories, the following section identifies, categories, a nd builds upon these basic building blocks. The Motors of Change Â“Motors of changeÂ” explain why and how ch ange unfolds and refers to the building blocks from which change theories are de rived (Van de Ven and Poole 1995). This section identifies six building blocks that have been discus sed in several publications. Principles are extracted from the six change processes in regards to how the type of change affects resistance and the management st rategies that would be most successful in the ES context. Using these motors in examining the pattern of an ES-facilitated organizational change is valuable for four r easons: 1) they are the roots from which many change theories are based; 2) they can be used in building theory that can be used to explain the pattern of changes in an ES impl ementation; 3) they focus research towards certain aspects of the change that are key building blocks in expl aining change; and 4) they also address multiple perspectives to change Â– for example, Robey and Boudreau (1999) points out that multiple interpretati ons are useful in id entifying patterns of influence and change. The non-IS literature that focuses on issu es related to change in organizations should help to identify perspectives and issues that management and users would encounter in an ES implementation. These th eoretical perspectives are brought in from
45 the organizational change literature to help in addressing the issue of user resistance. There are a number of articl es that examine the change processes in organizations; however, two articles reduce the change proces ses into simple Â“motors of changeÂ” which explain organizational cha nge (Ford and Ford 1994; Van de Ven and Poole 1995). Change theories in disciplines from biologi cal science to organizat ional behavior often use one or a combination of these Â“motorsÂ” to explain the change. Van de Ven and Poole (1995) addresse s four motors of change: teleology, dialectics, life cycle, and evolution. Ford and Ford (1994) discusses two other motors: trialectics and formal logic. For the teleological motor, there is one discrete entity that shares a common goal. This entity may accept th is goal either implicitly or explicitly but the social construction process is clearly visible. Also, constraints and requirements exist in order for that entity to a ttain the goal. For the dialecti cal motor, two or more entities exist that oppose each other. These opposing entities engage in some form of conflict between them, which leads to either a new entit y, the defeat of one of these entities, or a stalemate between thes e entities. Conflict is necessary between two opposing entities leading to some form of synthe sis that results from this conflict. For the lifecycle motor, change causes an entity to progress through di stinguishable stages. There is some form of logic, rule, code, or a routine that de termines the stages and the progression that occurs. For the evolutionary motor, multiple entities exist and there are mechanisms that lead to some form of selection, variat ion, and retention of the entities or the characteristics of these entities. Ford et al (1994) describes the tr ialectics motor as an entity that is attracted to one of multiple Â“material manifestation pointsÂ”, which are places
46 of equilibrium until there is a stronger attr action to another Â“material manifestation pointÂ”. Ford et al. (1994) describes the Formal Logic motor as the examination of something that occurs, the resulting eff ect, and the relationship between these two occurrences. Table 8 is a framework that draws from the work of Va n de Ven et al. (1995) and Ford et al. (1994) as well as other publicati ons that have addressed or used these motors of change in order to build upon these concep tions. The differing attributes of the six motors are pointed out and their applicabilit y and usefulness to the ES environment is described.
47 Table 8: Six Basic Building Bl ocks in Explaining Change Lifecycle Formal logic Dialect ic Teleology Trialectic Evolution Metaphor Reoccurring Set Processes Replacement of old ideas/entities Conflict Between Entities Goal-oriented cooperation (continuous improvement) Employees attracted to best option Best option eventually succeeds Progression A linear sequence exists that guides the change Removal of old process and replacement with new Recurring conflict between entities with eventual synthesis Iterative process of goal setting, implementation, reassessment Entity attracted to best option and remains until a better option exists With multiple options, there is recurring conflict until the best option remains Contributing Forces Previous life cycles New process is substituted Opposing entities and the level of conflict Goals and the success of the implementation Level of attraction of options Level of conflict Assumptions about resistance The type of resistance that occurred in a previous lifecycle will occur again The old and the new cannot coexist, so resistance does not occur All conflict is because of resistance Those who do not support the goal are resistors Resistance does not exist; an entity does not embrace a change because of a lack of attraction towards it Resistance is immaterial because the best option eventually succeeds over a long time period IT-related Example Software Development Waterfall Model Direct cutover to new system Subordinates are forced to use a system Incremental System development Programmers attracted to most suitable programming language for the task Multiple word processing packages in use until one option Â“winsÂ”
48 Diagram Van de Ven et al. (1995, p. 520). Ford et al. (1994, p. 759) Van de Ven et al. (1995, p. 520). Van de Ven et al. (1995, p. 520). Ford et al. (1994, p. 765) Van de Ven et al. (1995, p. 520). View on Change Predictable, based on past change Throws out old and replaces it with new (change through replacement) Changes emerges from a synthesis of the conflict Change occurs because of the goals that are set Entities are attracted to change Used in describing long periods of growth with no major upheavals Usefulness in Identifying Resistance in an ES Implementtation There are some processes that are consistent across organizations in an ES implementation and to some degree resistance can be predicted This motor does not focus on resistance Â– rather it is focused on the old being thrown out in order for the new to exist. Competing structures are destroyed prior to enacting new structures This is a useful lens in which to examine the conflict between management and users in an ES implementation Management definitely sets goals in an ES implementation, however, this motor does not address the conflict or resistance between the goal-setters and those who must comply To some degree, if users are attracted to a change, they will be more supportive and less resistance will exist. However, a dilemma for management is how to make it attractive This does not apply to ES change, because the system is mandated and implemented quickly rather than a longer time period where the best system is selected
49 Ford et al. (1994) describes how motors of change affect the level of resistance in more detail than Van de Ven (1995). In rega rds to formal logic, opposition is viewed as two mutually exclusive entities, one of whic h needs to be displaced. Thus, resistance does not occur as the old and the new cannot coexist. From a dialectics standpoint, resistance occurs because of the opposition between the entities. Since two opposing groups exist, one opposition group is failing to go along with the change and thus that entity is considered to be resistant. Based on this view, the way to minimize resistance is to enact mechanisms that reduce the level of resistance. Thus, in an ES setting, that would entail management strate gies that make it more painful not to comply than to comply or easing the transiti on through strategies such as providing more detailed explanations of the change. From a trialectics standpoint, it is assumed that resistance does not occur as there is no opposition that people need to overcome; rather, if employees do not seek after the proposed change it is due to a failure to appropriately attract employees. Thus, an appropriate mana gement strategy from this standpoint would be to make the proposed change more attrac tive to employees. If the ES implementation appears attractive to employees, they embrace the system and the organizational changes that are to occur. Although all the motors of change may be present in an orga nizational change, it is likely that one or two may explain most of the resistance that occurs. The following paragraphs describe several studies that have focused on either one or multiple motors in explaining organizational change.
50 Soh et al. (2003) uses a dialec tical perspective to explain the misalignment that occurs between an organizationÂ’s struct ures and the structure that is embedded in the ES. Soh et al. (2003) finds that one set of forces arose from the structures embedded in an ES and another set of forces developed from the or ganization that had its set structures. The structures in an ES may include decisionmaking, reports, processes, and organizational controls. On the other hand, organizational structures include shared norms, current processes, values and expectations, a ll of which have developed through the organizationÂ’s history. These two different stru ctures are often at odds with each other, leading to the dialectical nature that tends to be present in an ES implementation. Soh et al. (2003) found that there is a misalignme nt between the ESÂ’s structures and the organizationÂ’s structures in ar eas such as data ownership, da ta entry, job scope, reports, workflow changes, and revenue processing. Greiner (1972) discusses the evolutionary and revolutionary approaches in describing the nature of change. Â“Historical forces [organizational age, organizational size, stages of evolution, stag es of revolution, and the growth rate of industry] do indeed shape the future growth of organizationsÂ” (G reiner 1972, p. 38). Grei ner (1972) refers to evolution as the periods of time that no major upheaval occurs as opposed to the revolution which is the periods of time th at organizations experience considerable turmoil. Cule and Robey (2004) develops an or ganization change theory based on the dialectic and teleological motors. The teleol ogical perspective is taken into account as
51 this goal-oriented approach appears to be imp licit to managers. The dialectical approach is taken into account as employees do not necessarily support the goals, and thus interplay exists between these opposing forces. Furthermore, Cu le et al. (2004) uses both an individual (teleological) and organization (dialectical includes multiple individuals) level of analysis in order to increase expl anatory power while maintaining consistency between the two levels. Cule et al. (2004) found that a teleological motor, among senior level management, essentially constructed th e new organization, but that the goals were resisted as employees did not support the new goal. The six motors are used as lenses by whic h to examine management strategies to minimize resistance. First of all, the ES implementation may have conflict between entities, which seems to be inherent in ES implementations. The use of the dialectics approach is used to examine the ES change and may lead to a furt her understanding that helps to identify the contributing forces to the struggle between management and users and the resulting synthesis that occurs. Second, the trialectics motor is useful as there are incentives used to attract users to change and thus reduce user resist ance and lead to an improved understanding of management strategi es. Third, the teleology motor is useful as management sets goals for the ES; this leads to a better understanding of the development of management strategies and goa l setting. Fourth, th e lifecycle motor is used in examining management strategies as there may be some process cycles of ES implementations that are likely to carry over fr om one organization to another. Fifth, the evolution motor helps to identify incrementa l changes that occur in the organization. Although the ES change tends to be more of a revolution to the organization, some
52 structures receive only gra dual change through the implem entation. Sixth, the formal logic motor is useful since there are structur es that are discarded with the ES that may help in the understanding of management strategies. Therefore, since motors may contribute to a better understanding of management strategies in an ES implemen tation, six principles derived fr om these motors are used in this study, and described in Table 9 below: Table 9: Applicability of the Motors of Change Motor Principle Areas to Examine Dialectic There is a struggle between management and users that eventually leads to some form of synthesis The nature of the struggle as well as what leads to the synthesis of ideas Trialectic There are attr active attributes of a change that draw users towards the change What attractive attributes exist in an ES implementation that can guide managementÂ’s decisions Teleology There is some form of goal setting that occurs in the organization and potential conflict w ith those who are not supportive of the goals The nature of goal setting and the resulting conflict that occurs Lifecycle There are repeated processes that occur from one ES to another The nature of implementation processes and how they vary from one implementation to another Evolution There are evolutionary aspects of the change which may affect the management strategies Evolutionary aspects of the change Formal Logic There are structures that are discarded that may affect the effectiveness of the management strategies The removal of organizational structures
53 As there are no clear theories available by wh ich to examine the research questions, this chapter has addressed the related literature and provided lenses by which to examine both user resistance and management strategies. The nature of this study is exploratory, and thus the following chapter lays out a methodol ogy to both explore answers to the research questions, as well as validate quantitatively answers to two of the four research questions.
54 CHAPTER III. STUDY 1 This study is examining an area where th eory is lacking. Thus, the research questions revolve around identifying issues pe rtinent to user resistance. The underlying reasons for user resistance, th e resistant behaviors that are exhibited, and the management strategies to minimize resistan ce all need to be explored. In order to answer the first three research questions, a qualitative study was conducted, which allows for systematically gathering data that may not be subject to quantification. This study encompassed interviewing people who have been involved as managers, IT personnel, or users in an ES implementation. These inte rviews were tape-recorded, transcribed, and analyzed to understand the underlying reasons for user resistance, the user resistance behaviors, and the management strategi es that affect user resistance. Epistemology A qualitative research method is best suited and has been used to answer the first three research questions. Qualitative resear ch methods enable researchers to examine social and cultural issues through the use of interviews, observation, questionnaires, manuscripts, and researcherÂ’s impressions (Myers and Avison 2002). Qualitative research can be interpretive positivist, or critical, de pending on the researcherÂ’s philosophical assumptions (Myers and Avison 2002). Positivist research assumes that reality is objective and has meas urable properties and generally attempts to test theory.
55 Interpretive research assumes th at access to reality is thr ough social constructions, such as consciousness, language, and shared mean ings, and attempts to understand the context and processes. Critical research assumes that social reality is consti tuted historically, and attempts to perform a social critique (Myers and Avison 2002). Study 1 works within the interpretive epistemology as it seeks to understand reality through social constructions and understand the context and processes. Ta ble 10 below further co ntrasts the positivist and interpretive epistemologi es in relation to this study. Table 10: Comparison of the Positivist and Interpretivist Epistemologies Positivist Interpretive Unit of Analysis ES user ES user in the context of the system and organization Goal Identify reasons for user resistance, resistant behaviors, and management strategies to minimize resistance Understand the meanings behind the reasons for user resistance, resistant behaviors, and management strategies to minimize resistance Coding Test the hypothesized categories or categories identified in previous research Use of grounded theory to derive categories not identified previously Viewpoint on the transcripts The meaning is static and can be derived from the text The meaning is based on contextual issues and can only be understood by understanding the context For interpretive research, an impor tant feature, stemming from the anthropological tradition, is the Â“thick descriptionÂ”, due to the intertwined and complex conceptual structures (Walsham 2002). This detailed description is necessary to understand the complex interactions among empl oyees that ultimately affect outcomes. For the use of theory in interp retive case studies, there are th ree major uses: Â“as an initial
56 guide to design and data collection; as part of an iterative process of data collection and analysis; and as a final product of the researchÂ” (Walsham 2002, p. 104). Methodology There were three steps taken in data collec tion. The first was the use of an expert panel. The second step was an in-depth case study of an implementation at a large university. The third step, which was used to validate the findings, was interviews with multiple employees in an Asian airline and a ce llular company. For all of these steps, the level of analysis was the individual. Step 1: Expert Panel The goal of the first step is to unders tand the major issues related to user resistance that arises in an ES implemen tation. The interview script is shown in Appendix E and was developed with genera l questions revolving around the first three research questions. The discussion with th e expert panel was semi-structured as many follow up questions were added to further prob e into the comments made by participants. There were two parts to this expert panel: a focus group with seven IT professionals that have been involved with Enterprise Systems a nd an interview with an expert that has led the rollout of several ESs as a CIO or a Fortune 500 firm. The focus group ranged from heavy involvement in an implementation to o ccasional usage. This focus group was used to extract perspectives on the reasons for resistance, the resistant behaviors and the management strategies to minimize resistan ce. The session lasted over an hour and all conversation was recorded and transcribed. Mo st of the members of the focus group told
57 multiple stories about the types of resistance they saw during the ES rollout, their perceived reasons for the resistance, the resistant behaviors, and the strategies management used to minimize the resistance. The separate interview with the individual expert followed a similar format to the focu s group. This interview lasted approximately one hour and also was reco rded and transcribed. The transcripts from the interviews of both the focus group and the individual expert were analyzed to extract the major pr inciples and concepts. All comments related to reasons for resistance, resistant behaviors, and management strategies to minimize user resistance were highlighted and then analyze d. Upon completion of analysis of the data collected in the case study (d escribed in step 2 below), a further analysis of these transcripts was performed in order to integrat e comments from the expert panel with the in-depth case study. No claim is made regarding the representa tiveness of these experts; however, their level of involvement with the rollout of an ES was useful in developing an initial understanding of the user resistance. The sa mple quotes in Appendices A, B, and C that are labeled Â“F1Â” are from members of the focus group and provide information on user resistance that was experienced by the IT pe rsonnel that comprised the focus group. The purpose of this first step was to gain an ini tial understanding of user resistance in an ES implementation and to highlight some of the key issues, not to make any claim about the representativeness of their comments. The inert bias in the fact that all experts were IT
58 professionals was taken into account and thus both step 2 and step 3 focused on users from various backgrounds rather than focusing on IT professionals. Step 2: In-depth Case Study The second step was an in-depth case study. If no a priori theory is posited, a grounded theoretical approach can be used with case studies (Eis enhardt 1989). Case methodology is useful when a natu ral setting is required and in particular, a rich natural setting may be useful for generating theori es (Benbasat, Goldstein and Mead 2002). Most case studies are explorat ory as they seek to explor e and describe a phenomenon (Benbasat et al. 2002). Some case studies desc ribe the events and then present one or multiple theories to explain events (Mar kus 1983; Franz and Robey 1984; Kling and Iacono 1984) while other case studies test theories (Keen 1981; White 1984; Bonoma 1985). Eisenhardt (1989) suggests that case study research may produce concepts, propositions, or a conceptual framework. Th e results of this study can be generalized through these outputs, which are similar to Â“g rounded theoryÂ” (Glase r and Strauss 1967). Strauss (1990) points out that grounded theo ry builds theory yet does not begin with theory; rather, it focuses on an area of study that is relevant. The area of study, or related literature, stimulates sensitivity to theory th rough the identification of relationships and concepts. Moreover, the literature is usef ul because of the descriptions provided of reality. Strauss (1990) differentiates theory and description by pointing out that theory uses concepts, which are interpretations on da ta, and relationships be tween the concepts.
59 Description does not include forming data into a conceptual theme and even though it may include organizing data into themes or c oncepts, it tends to have summaries of the data rather than interpretations. There are four purposes for the procedures of grounded theory: 1) build theory; 2) incorporate the n ecessary rigor into the process to enable the theory to be Â“goodÂ” science; 3) help the analyst break biases and assumptions; and 4) provide grounding needed to develop an explanat ory theory that closel y represents reality (Strauss 1990, p. 57) For this case study, a large public university was selected that is located in the southeastern United States. This location was ideal for several reasons: 1) The rollout of the ES was a major change, affecting many employees; 2) This university faced resistance in many departments; and 3) Becau se the university is a state institution with stable employment, employees would likely be more forthright with their resistance experiences. With approximately 40,000 st udents, and close to 10,000 employees, The selected system contained nine modules: Purchasing/Procure to Pay/Order to Cash, Grants, Accounts Payable, Asset Manageme nt, Accounts Receivable/Billing, Budgets, General Ledger, Project Costing, and Record to Report. Users of the system were sought out to be interviewees for this case study. Employees were selected based on three criteria: 1) represent different departments; 2) represent di fferent positions; 3) they use (or used) the system regularly. There were 22 people interviewed from all levels of the organization: 5 clerical staff, 2 IT professionals, 3 trainers 2 top management, 4 middle management, 4 office managers, 1 accountant, and 1 purchaser. Seven of these users were superusers, which is an employee that undertakes an either partor full-time role
60 with the ES implementation that requires a grea ter degree of commitment to the project. This employee tends to be more knowledgeable and skilled with the system and business processes and often is the first person a group of people go to for support. This study used semi-structured interviews for the primary data collection. These interviews obtain the interpretations of the interviewees in regard to the processes and events of system implementation, reflecting an external reality (K irk and Miller 1986; Cooper 2000). Although some questions are directly related to the intervieweesÂ’ response to the system, questions were al so asked that require d the intervieweesÂ’ interpretation of events. For example, the intervieweesÂ’ interpretation of the reasons for user resistance and resistant behaviors of others was sought out along with the intervieweesÂ’ own reasons for user resistan ce and resistant behaviors. Because these users experienced the implemen tation of the system and know and talk with other users who experienced the implementation, the e xperiences of the in terviewee and the intervieweeÂ’s interpretation of others was sought. The literature review served as the ba sis for developing the primary questions noted in the interview guide (Appendix E). The data collection at this organization continued until a point of theoretical saturation; in other words, the value of an additional interview was considered negligible (Eis enhardt 1989). The interview length ranged from 25 to 77 minutes, averaging 47 minutes The interviews were recorded and transcribed in order to acquire all of the intervieweeÂ’s comments, yielding 242 pages of single-spaced transcripts (135,200 words).
61 Although one interviewee who worked in th e legal department of an organization appeared suspicious and cautious of what was said, the rest of the interviewees appeared candid in their responses and did not mind be ing recorded. When questions were asked during the interviews, the researcher tr ied to listen well while conveying a nonjudgmental attitude. Walsham (2002) was taken into account as it warns that data may lose its richness if the interviewing style of the interviewer is over-directing the interview through tight controls. On the other hand, if th e interviewing style is excessively passive, the interviewees may conclude that the research er is not interested in their views or have no views of their own, which may lead to th e doubting of the professional competence of the researcher (Walsham 2002). Techniques were used from grounded theo ry (Glaser and Strauss 1967) in an attempt to derive basic concepts and stru ctures among the concepts. Each interview transcript was analyzed in depth. All the sent ences from the transcripts were first marked whether or not they had any direct relevance to the areas under inves tigation (there were a number of paragraphs that provided extra information such as backgrounds on the individual or system, but did not relate directly to any of the research questions). Next, all statements related to reasons for user resistance, resistan ce behaviors, and management strategies to mini mize resistance were extracted for further analysis. These extracted statements included statements from the expert panel as well as the interviewees. Each of these extracted statements (and contex t, if useful in understanding the sentence) was put into one of three separa te documents either reasons for resistance,
62 resistant behaviors, or management strate gies to minimize resistance. These three documents were then analyzed to identify themes. As the researcher progressed through the tran scripts, there was a need to refine the emerging themes. Strauss (1990) recommends several steps in coding Â– the phenomena under investigation needs to be labeled, categor ies need to be discovered, categories need to be named, and the categories need to be developed based on their dimensions and properties. Multiple themes were identified in the areas of reasons for user resistance, resistant behaviors, and management strategi es to minimize user resistance. After the initial identification of themes, there were multiple iterative rounds of analyzing the themes that emerged and reclassifying stat ements according to what emerging themes improved the classification. This essentially followed the hermeneutic process laid out by Klein and Myers (1999) which suggests an iterative process of reflecting on the interdependent meanings of the parts (indi vidual statements) and the whole (evolving themes or conceptual framework). For the reasons for resistance, Table 11 below identifies the four rounds in the iterative process to uncover th e underlying reasons for resistance. In round 1, which is the first time the statements were read, 26 themes emerged. After rereading all the statements related to each theme, the stat ements were either kept in the same group, merged with another group with similar undert ones, or renamed to better describe the theme that emerged.
63 Table 11: Identification of Reasons for Resistance Round Themes Reasons for Resistance 1 26 Lose Freedom/become more accountable; Culture/Environment; Computer Self-Efficacy & Computer Skills; Lose Expertise; Communication; Job Change; Mgmt vs. End-User or Dept.; New skills/Lack of skills; Uncertainty; Lack of Incentives; Changed terminology/structure; Lack of Fit; Process Problem/Change; Complexity; Workload (extra work, more work to get same info, extra time); Tech issue; Shadow System; Training; Lack of Input; Lack of Knowledge; Lack of per ceived value; Stressful; Loss of power; Learning style; Users w ho donÂ’t use it much; Comfort. 2 14 Communication; Complexity ; Computer Self-Efficacy; Culture/Environment/Mgmt. vs. e nd-user; Lack of Input; Lose Expertise/Power; Lose Freedom /Become more accountable; New Skills/Skillset/Lack of skills/New way of thinking; Psychological Contract Change; Process Problem /Change; Tech Issue; Training; Uncertainty; Workload 3 13 Additional Workload; Uncerta inty; Lack of Input; Loss of Autonomy; Loss of Expertise/Power; Facilitating Environment; Changed Expectations; Process Change Problem; New Skillset; Technical Problems; Complexity ; Poor Communication; Poor Training 4 12 Uncertainty; Input; Control/P ower; Self-Efficacy; Technical Problems; Complexity; Facilita ting Environment; Communication; Training; Job/Job Skills Change; Workload; Lack of Fit; The resistant behaviors also went through an iterative process. However, since the second round produced distinct ly different behaviors, the choice was made to classify the behaviors according to types of behavior rather than themes of behaviors. The classification of resistant behaviors, shown in Table 12 round 3, is based on a classification of behaviors proposed by Bovey et al. (2001a p. 375) and Bovey et al. (2001b, p. 534). These studies classify beha viors based on an overt-covert continuum and an active-passive continuum.
64 Table 12: Identification of Resistant Behaviors Round Themes Resistant Behaviors 1 21 Animosity; Upset/Cry; Quitting Job/Turnover Intention/Job Change; Refusal/avoided when possible; Result of non-thorough training; Trying to use old system; Procra stinate; Not paying attention; Negative Attitude; Morale; Less Productive; Less Motivation; Hack to try to get system to do something; Impatience; Use Shadow System; Enter in Info just to get something done; Do Something their way; DidnÂ’t want to learn; Did not follow process then blame system; Complaints; Challenged. 2 19 Refusal to use system; Challenge system/plan; Hack at system; DonÂ’t follow process; Quit job/job change; Use shadow system; Try to use old system; Avoid system us e; Enter in info inappropriately; Complaints; Lower morale; Defens ive; Turnover Intention; Not Motivated; Less Productive; Impa tient; Not paying attention; Procrastinate; DonÂ’t want to learn 3 4 Overt-Active; Overt-Passive; Covert-Active; Covert-Passive Last of all, Table 13 below shows the f our rounds of iterative theme development among statements leading to the eight distinct management strategies identified in round 4.
65 Table 13: Identification of Management Strategies to Minimize Resistance Round Themes Management Strategies 1 47 Process Change Meetings; In terface with Existing Systems; Training; Communication; Listen to Feedback; Vanilla; Process Change; Change in Management Strategy; Upper Management not understanding lower stuff; Shadow System; Visit other locations; ES selection process; Timeframe; Volunteers; Incentive; Selection; Alternatives; Full time rollout pe ople; Mgmt Inconsistency; Two sided view; Centralizing; SDL (sol ution design lab); Backfill Jobs; More Resources; Focus on Business Processes; Planning; Clear Vision; Help/Support; Empathy; Ga in Support; Documentation; Pay Structure; Plan; Upgrades; Invol vement; Structure; SuperusersÂ’ Plan; Standardize; Capture Non-compliers; Initial Session; Individual Stepping Up; Internati onal Issue; Lack of Enforcement; Questionnaire; Consultants; Reassi gn People; Managers donÂ’t use system 2 21 After the Rollout; Change in Management Strategy; Communication; Customizations; Documentation; ES Selection Process; Help/Support; Implementa tion Team; Incentive; Lack of Enforcement; Listen to Feedback; Mgmt Consistency/Inconsistency; Non-Management Strategies; Non-re sistance related; Planning; SDL (solution design lab); Superusers plan; Training; Upgrades; Upper Management not understanding; Visit other location. 3 11 After the Rollout; Communication; Customizations vs. Reengineering; ES Selection; Help and Support; Implementation Team Structure; Incentive; Listen to Feedback; Non-Resistance Related; Training; Upper Management not understanding. 4 8 Top-down communication; Li sten to Feedback; Provide Help/Support; Training; Incentives ; Clear Consistent Plan; Management Expertise; System Customizations Sample quotes are provided in Appendix A regarding the reasons for resistance, Appendix B for resistant beha viors, and Appendix C for ma nagement strategies to minimize resistance. Employees within the same organization judge management strategies very differently and it is interesting to note the ex istence of multiple realities within the quotes. For example, within th e same organization some employees think the top-down communication is excellent while others find fault with it.
66 As this research is exploratory in und erstanding the reasons for resistance, the behaviors that are manifested, and the management strategies to minimize resistance, an a priori list was not available. Thus, the th emes that emerged through the iterative process of analyzing the data and restructuring the themes each were assigned a code. The coding is described in the Reliability/Va lidity section. Step 3: Semi-structured Inte rviews in Two Organizations The third step was the use of semi-structured interviews with employees heavily involved in ES implementations at two organizations to valid ate the findings of the first two steps. The use of multiple organizations is useful in order to make more controlled observations and controlled deductions and in crease the level of generalizability. Although case studies tend to collect data th rough multiple means (Benbasat et al. 2002), the use of multiple interviews in multiple organizations can be useful when the focus of the research is on theory build ing, description, or theory test ing. In a case study, building theory is an iterative process as a research er may compare cases, redefine the research question, then add another case (Eisenhardt 1989) For this researc h, multiple interviews were conducted at multiple organizations to better understand the nature of ES implementations and the coinciding user resistance. For this third step, multipl e interviews were conducted at two organizations: One organization is an airline located in Asia and the second organizati on is a cellular phone company located in the U.S. Although the seco nd step encompassed users from all levels
67 of the organization, this third step only included people heavil y involved in the rollout of an ES. Because of their widespread experi ences, these interview ees were useful in validating the findings based on the first two steps. Phone interviews were conducted with 7 employees from the airline compa ny and with 4 employees from the cellular phone company. Although these employees represent a number of different areas within each organization, all of them were heavily involved with the implementation of the system. The roles of these employees were Accounting Operations Manager, Project Manager, Finance Manager, IT director, HR System Manager, Purchasing System Manager, Financial Systems Manager, R ecruitment Manager, IT for Corporate Accounting, Accounts Receivable Manager, a nd Procurement Manager. All of these interviews were recorded, lasting an av erage of 40 minutes, ranging from 25 to 51 minutes. The recordings were all transc ribed, yielding 106 singlespaced pages (47872 words). Two research assistants coded these transcri pts as well. They were instructed to use the same coding scheme developed from th e first two steps of this study, shown in Appendix D. Also, they were instructed to identify any other reason for user resistance, user resistance behavior or ma nagement strategy to minimize the resistance that was not on the coding scheme. This is discussed fu rther in the reliabil ity/validity section. Examples of the ratersÂ’ coding is provided in Appendix G. This Appendix provides examples of coding that was consistent am ong the raters as well as coding that was inconsistent.
68 Reliability/Validity A study cannot be valid without first being reliable. Firs t of all, reliability is shown through the suitable use of and adhe rence to the case study protocol (Yin 2003). The interviews were semi-structured with the general categories of questions shown below and a more detailed interv iew script shown in Appendix E: 1) What is your level of in volvement in the project? 2) What resistance or opposition to the system did you observe? 3) Why do you think this resistance occurred? 4) What management strategies that you observed were useful in minimizing resistance? Follow up questions were asked from the interv iew script in order to further probe into the underlying issues. Reliability is also shown through the codi ng. For step 2, after the categories were discovered and named, the codes/themes were checked for reliability and definitional clarity (Miles and Huberman 1994). Two graduate research assistants, taking part only in the coding and unfamiliar with the research, were used to read and code the transcripts from which the researcher had derived the ca tegories. Both research assistants were provided with a one-page coding scheme th at identified each code/theme and its operational definition. Each paragraph in the transcripts could be assigned zero, one, or multiple codes. The research assistants firs t examined one interview transcript, discussed discrepancies, and then continued to code a sample of the remaining transcripts. The CohenÂ’s Kappa statistic was used to anal yze the level of correspondence between the
69 coders, which is a measure of the strength of agreement between coders adjusted for chance agreement. CohenÂ’s Kappa for this coding was 88.7%, well above the 61% level that is suggested to have Â“substantial strength of agreementÂ” (Landis and Koch 1977, p. 165). The coders used the coding scheme devel oped from the first two steps to code the interviews conducted during step 3. The c oding of the 11 interviews resulted in a CohenÂ’s Kappa of 83.1%. This adds support for the reliability of the constructs since the CohenÂ’s Kappa statistic for the coding of th is third step also was well above the 61% threshold. The actual coding values for the i ndividual transcripts are identified in Table 14 below. Interviewee7 had the least amount of experience with the implementation and thus did not contribute as much information. The coders were in complete agreement with all of Interviewee7Â’s statements related to user resistance, re sistant behaviors, and management strategies. Table 14: CohenÂ’s Kappa for Coding of Step 3 Interviews Organization Interviewee CohenÂ’s Kappa 1 0.806 2 0.750 3 0.785 4 0.828 5 0.803 6 0.890 Airline Company 7 1.000 1 0.769 2 0.825 3 0.843 Cellular Company 4 0.841
70 Also for step 3, besides coding, the code rs were also asked to identify any concepts/constructs they saw in the intervie w transcripts that we re not in the coding scheme. There were two con cepts/constructs that were ma rked as not in the coding scheme, both of which were identified as poten tial reasons for resistance. One was the aggressive time frame of the implementation and the other was a lack of trust in the system. While both of these are not include d in the coding scheme, it was decided not to include them because the context of these tw o issues suggested that they both tie into reasons for resistance already on the coding sche me. For the first issue, the aggressive time frame, either the workload increases because of the quick implementation, or additional problems are created such as tech nical problems and lack of fit problems. Since these issues were already addressed, it was decided not to include aggressive timeframe as a reason for user resistance. Fo r the second issue, the lack of trust in the system was only mentioned in one transcri pt; the context implied that the underlying reason was either a lack of self-efficacy or unc ertainty, or both. Thus, no new constructs were added to the coding scheme. This lack of identification of new constructs by the coders adds support for the validity of the cons tructs originally identified from steps 1 and 2. For all three steps, external validity was an important consideration as it essentially is the generalizability of the studyÂ’s findings (Yin 2003) External validity was established in several ways: 1) through the use of an expert panel which is comprised of experts who have been involved with ES s within various industries; 2) detailed examination of user resistance through inte rviews with employees from organizations
71 representing three different industries; and 3) followi ng the theoretical sampling techniques suggested by Glaser and Strauss ( 1967), the research sites have been chosen because of the similar yet varied conditions. All the organizations were involved with ES implementations, but are from different industr ies. The ESs were not identical, since the software packages were from different ve ndors as well as the or ganizations installing different modules; however, the ES implement ations were mandatory in all cases and they changed the workflow processes and altered jobs. For step 2, triangulation of the data also contributed to validity. It has been suggested that Â“every organiza tional situation is likely to be filled with multiple and frequently conflicting interp retations and meaningsÂ” (Prasad 1993, p. 1404). Thus, in a case study, it is important to establish constr uct validity. Construct validity is supported through the use of multiple sources and multip le data collection methods (Benbasat, Goldstein and Mead 1987; Benbasat et al. 2002; Yin 2003). In regards to the multiple sources, statements made from one intervie wee were compared and contrasted with statements made with other inte rviewees in order to triangula te the ideas suggested by the interviewees. Multiple data collection methods were also used, since the interviewer was given access to training manuals, emails, memos, and other written documentation concerning the project. There was also an overview of the system provided for the researcher, which provided a better understanding of the process through which users traverse. Besides the diverse and differi ng opinions among the users, there were no discrepancies found among the various data sources.
72 Results Based on the iterative construct formati on process described in the previous chapter, there were reasons for user resist ance, resistant behaviors, and management strategies to minimize resi stance that emerged. Tables 15, 16, and 17 identify the emergent constructs. As noted in these tables the constructs have also been placed into categories, such as Table 15 categorizing the constructs into individual, system, organizational, and process issues. The use of categories was added to provide a better understanding of the types of reasons, be haviors, and management strategies. In order to address research question one, Â“What are the underlying reasons for why users resist an ES implementation?Â”, Table 15 below addresses the reasons for resistance. There were four constructs th at best fit under the category of Â“Individual IssueÂ”: Uncertainty, Input, Control/Power, a nd Self-Efficacy. These constructs best fit under this category because they all are individual psycholo gical constructs that are intrinsic. Each employee has a level of desire towards thes e constructs. For example, one employee may be satisfied with uncertainty as long as his job is not on the line while another employee is satisfied only if the daily tasks are predictable. There is a greater chance that employees not satisfied with these Â“Individual IssuesÂ” will cause an unfavorable outcome to the organization. The constructs Technical Problems and Complexity were both put into the category of Â“System IssueÂ” because they were pr imarily related to system usage. In an
73 organizational change that is not technology-enabled, thes e constructs would not be contributing factors. However, the impleme ntation of a large system requiring usage often leads to users experienci ng technical difficulties due to bugs or the complexity of the system. The constructs Facilitating Environment, Communication, and Tr aining were all put into the category of Â“Organizational I ssueÂ” since they revolve around organizational aspects necessary to meet the needs of users. Whether or not an organizational change requires technology, employeesÂ’ at titudes are affected by thes e constructs because they revolve around organizational issues. For example, one organization may embrace new technologies in spite of poor communication while another organization has always been relatively stable, and no t conducive to embracing new technologies, although communication may flow well between employees. Finally, the constructs Job/ Job Skills Change, Workload, and Lack of Fit were placed into the category Â“Process IssueÂ” becau se they all are problems faced by users resulting from the changed processes synonymous with ES implementations. Technology-enabled change requires new proces ses that change the jobs of employees and often requires new skills. New processes usually demand a greater workload in the short-term and sometimes for the long-term. Furthermore, problems may arise because the new processes do not fit well with in the organizational structure.
74 Table 15: Reasons for User Resistance Construct Definition Examples Uncertainty User is unclear of the future Unknown future, potential threat, lack of clarity Input UserÂ’s opinions are not considered The thoughts and opinions of users were not sought out Control/Power User loses control or loss of recognition as the expert Leveled playing field, not the expert anymore Individual Issue Self-Efficacy Perceived lack of capability Lack of confidence, lack of computer skills/abilities Technical Problems Problems with the system Bugs in system, f eatures that donÂ’t work right System Issue Complexity System is complicated to use Difficult to access, Poor user interface that lacks logic or is not intuitive Facilitating Environment Organizational culture is not conducive to the change Lack of technology usage in organization, bureaucracy that is slow to change Communication Communi cation to users is problematic Lack of communication, users not hearing benefits of system, lack of coordination, users not understanding why Organizational Issue Training Training does not meet organizational needs Lack of training, training seems to be a waste of time, incompetent trainers, timing of training, sufficiency of training Job/Job Skills Change UserÂ’s job or job skill requirements changes Revised job description, different job tasks, new skills, new way of thinking Workload User is required to put forth additional effort Extra work, more work to get same info, extra time Process Issue Lack of Fit Process problem between the system and organizational structure Problematic changes to processes, new processes not working as planned To address research question two, Â“Through what behaviors does user resistance manifest itself in an ES implementation?Â”, Table 16 shown below addresses the user resistant behaviors that were found. The resistant behaviors that were described by employees involved with ESs were classified by the scheme laid out by Bovey et al.
75 (2001a, p. 375) and Bovey et al. (2001b, p. 534). These articles classified resistant and supportive behaviors based on whether they were overt (clearly expressed) or covert (minimally expressed) and on whether they we re active (person take s action) or passive (person is inert). Although there are likely other behavior s that may be exhibited by users, these behaviors are the ones that were mentioned during the interviews. The 2 x 2 matrix has been used to classify the types of behaviors that were mentioned. Table 16: User Resistance Behaviors Overt (clearly expressed) Covert (minimally expressed) Active (takes action) Refusal to use system Challenge system/plan Hack at system DonÂ’t follow process Quit job/job change Use shadow system Try to use old system Avoid system use Enter in info inappropriately Passive (inert) Complaints Lower morale Defensive Turnover Intention Not Motivated Less Productive Impatient Not paying attention Procrastinate DonÂ’t want to learn To address research question three, Â“In the ES context, what management strategies are effective in minimizing user resistance?Â”, Table 17 shown below categorizes and describes the management st rategies to minimize resistance that were identified in the interviews. A discussion on how each of these strategies is effective in minimizing user resistance is provided in Chapter V.
76 Table 17: Management Strate gies to Minimize Resistance Construct Definition Examples Top-down communication Top management/implementati on team communicating to users Communicating the types of changes to occur, the benefits of the system, the goals and vision, the Â“whysÂ”, managers sharing information with subordinates Effective Communication Listen to Feedback Management listening and responding to the input of users Distribute/collect questionnaires, address complaints Provide Help/Support Management offering assistance to users Availability of consultants or helpline, providing a support system to interface with the system Training Train the users at an appropriate time in a way that is suitable for their needs Trainers with knowledge/communication skills, address the needs of trainees, appropriate time frame Effective Education/Su pp ort Incentives Suitable motivators for users to learn and use the system Incentives to take training and to do extra work Clear Consistent Plan Straightforward consistent strategies Clear direction, consistent management strategies, following through with plans || opposite: confusion, failure to carry out plans Management Expertise Management understanding of processes and system Decision makers understand system and processes, Decision Makers understand the details Effective Direction/Plannin g System Customizations Customize the system to the processes in place Tailor the system to fit the usersÂ’ preferences/needs There are three categories that were iden tified, as shown in Table 17 above. The first category is Â“Effective CommunicationÂ”. This consists of communication from either top management or the ES implementation t eam to the users, which is the Top-down Communication strategy. It also consists of communication from the users to either top management or to the ES implementation team, which is the Listen to Feedback strategy. The second category is Â“Effective Education/ SupportÂ” and include s strategies that management can set in place to educate and support the user. This is done through the
77 Provide Help/Support strategy, wh ich involves assist ing the user, the Training strategy, which involves training the user effectively, and the Incen tives strategy, which involves supporting the user with suitable motivators. Table 17 essentially demonstrates the implications of this study since management strategies are identified that have emerged from the qualitative data as useful in minimizing resistance. Thus, the th ree categories of strategies that managers should strive for are to effectively comm unicate, effectively provide education and support, and effectively provi de plans and direction. Types of Users The results of Study 1 are also useful in setting the groundwork for Study 2, which deals with research question 4 and revol ves around types of users. Since the first part of this research question deals with th e existence of groups, Table 18 below offers a few quotes out of the many comments made by us ers that demonstrates the existence of different types of users. Comments are ma de on each of these quotes regarding how the quotes suggest that there are different types of users.
78 Table 18: Quotes regarding different types of users Quotes Comments on Quotes U3Using the old system, having it for 20 years they were experts in their field all of the sudden, you have leveled the playing field. And the new person coming off the street knows just as much about the system as you do, so you are no longer an expertÂ… Its 18 months later I don't have reports that I used to have. The interviewee points out that for some, the loss of control/expertise is the issue, but for others, it is not having what is needed U2they werenÂ’t able to access their budgets for six monthsÂ…it can be very frustrati ng for people, especially when they are not computer savvy or ha ve some sense that it is not really you, itÂ’s the systemÂ… [the system changes my job since itÂ’s a] different way and it mean s that I have to spend more time helping peopleÂ… I had to be available to answer questions. I had to be availa ble to help people solve their problems with the system. The interviewee points out that some people were not able to access budgets, others were not computer savvy, and personally, the process change was the driving issue U7I witnessed some people get ting just exasperated because the people who were training them were not that knowledgeable in the subject matter and uh, you know itÂ’s hard to say whose fault that isÂ… It became much more time consuming [Interviewer: Would you say approximately about double the time?] Yes, I would sa y. Of course part of it had to do with our inexperience with the systemÂ… we were reading the paper at the time about the takeover. And that didnÂ’t help morale either. They thought the software vendor has a terrible program, but theyÂ’re going to be around for a while. And wait a minute I read in the paper this morning that there may be this hostile takeover. That wasnÂ’t good either. Some people had poor training, while the interviewee experienced a more time-consuming job and faced uncertainty U10[the system] increased my workload in the sense in that I donÂ’t sit and wait for the depart mental ledgers to get to me so I can look at them. I can go in and run them myself or go in and look at them myself or I can run reports that are more specific to what I want, which before we always had to ask somebody else to run the reports for usÂ… IÂ’m not intimidated by computers or systems or things like that because I know that nothing I can do on this si de is going to hurt anything thatÂ’s in that systemÂ… I think some of it is intimidation with this system that theyÂ’re, they now have to go in and do a lot of things that they never had to do before and theyÂ’re they just donÂ’t feel comfortable with it and they donÂ’t feel comfortable going beyond theyÂ’re comfort zone. For the interviewee who is not intimidated by computers, it is the increased workload and process changes that matter. The interviewee points out that others are intimidated by the system and are not comfortable U9the training just gets le ss and lessÂ… new employees or employees who didnÂ’t have these roles before, but are taking them over, thereÂ’s just far too little training U9it takes much longer to get invoices paid. To me, I mean, The interviewee points out that some people face a lack of training while the interviewee
79 honestly, I canÂ’t think of any benefits [of the system]. faces a lack of communication of the benefits Based on the examples provided above, it appe ars that there are different types of users. Study 2, which is outlined in the next chapter, strives to identify and understand these groups. Although evidence is suggested from Study 1 regarding the existence of multiple groups, Study 2 does not include any hypotheses since Research Question 4 is exploratory.
80 CHAPTER IV. STUDY 2 In order to answer the fourth research que stion (What types of us ers exist in an ES implementation?), a second study was conducted based on the findings of the first study. Study 2 revolves around types of users that have common characteris tics and resistance patterns. The goal of this st udy is to further understand user resistance through seeking to answer the fourth research question. In accomplishing this goal, types of users are identified, the characteristics of resist ant groups are identified, and management strategies that are effective in minimizing the resistance of these groups are identified. Methodology Study 2 encompasses the development of a primarily quantitativ e questionnaire, a pilot test of the questionnaire, and a collection of a full data set of the questionnaire. The questionnaire sample is ES users. To be st answer the fourth research question, Qmethodology is used. Q-methodology encompasses the use of the Q-sort, which can be useful for understanding pockets of resistan ce as well as user pe rceptions of training issues and management strategies in order to mitigate resistance. For example, Brown (2004) notes that Q-methodology can comp lement a project managerÂ’s set of methodologies for understanding the perceptions of stakeholders. Furthermore, Thomas and Watson (2002) points out that Q-sort is pa rticularly suited for either exploring or validating both positivist and interpretivist conceptions within IS research. For this
81 study, Q-methodology is used within the interp retive epistemology to explore groups of users. Overview of Q-Methodology In Q-methodology, the goal is to uncover patterns of thought, not discover what percentages of people think certain ways (V alenta and Wigger 1997). The variables are the respondents, not the Q-statements (McKeown and Thomas 1988). Q-methodology deals with states of minds, which is why some publications have compared it to quantum theory that is concerned with states of matter (Brown 1986). The purpose of Qmethodology is to understand individuals and groups, not to generalize to populations, although to some extent generalization is possible (Thomas and Watson 2002). Brown (1993) points out that just as significant rese arch can be conducted through a single case study, the focus of Q-methodology is on the qua lity, not the quantity of the data. The researcher studies the individual to examine if responses revolve around one or multiple themes. The statistical analysis of the scale scores does not necessarily lead to predictability, but rather to an understanding of the nature of the factors that emerge and underlying thought patterns (B rown 1980; Brown 1986). Q-methodology was initially proposed by Stephenson (1935) and further developed in Stephenson (1953). The Q-me thodology requires the development of a concourse, which is a representative sample of statements about a domain of interest. The concourse is not limited to words, but ma y also include photogr aphs, collections of paintings and musical selections (Brown 1993) A concourse is t ypically derived through
82 interviewing people and record ing what they say or pulling out clips from essays or newspapers (Brown 1993). Respondents operate on the concourse by means of a Q-sort, which is a sorting of all the items, based on the criteria specified by the researcher. Qmethodology is distinct from R-methodology; R-methodology studies the relationship among variables (Steelman and Maguire 1999), us ing a technique to correlate variables such as regression or stru ctural equation modeling. One of the differences between Rmethodology and Q-me thodology is that samples in R-methodology are based on a se t of persons in a population; in Qmethodology, samples are based on statements drawn from a population. This is described in Stephenson (1935, p. 297), which states that R-methodology refers to Â“a selected population of n individuals each of whom has been measured in m testsÂ” while Q-methodology refers to Â“a population of n different tests (or essays pictures, traits or other measurable material), each of which is measured or scaled by m individuals.Â” Steelman and Maguire (1999) also addresses this, pointing out that while R-methodology is focused on patterns across variables, Q-methodology is focused on patterns of respondentsÂ’ perspectives. Brown (1980) points out that the letter Q is used to represent person correlations, as opposed to trait co rrelations used in R-methodology. In other words, R-methodology deals with the correl ation and factoring of traits while Qmethodology revolves around the correlation and f actoring of people. Because of this difference, Brown (1993) points out that Q-methodology interprets the factors by examining the factor scores rather than the factor loadings (which is done in Rmethodology). A factor score is Â“the score fo r a statement as a kind of average of the
83 scores given that statement by all of the Qsorts associated with the factorÂ” (Brown 1993, p. 177). A second difference is that Q-methodology revolves around subj ectivity. This concept was proposed by Stephenson ( 1935), and focuses on respondents measuring rather than being measured. Subjectivity is Â“a personÂ’s communication of his or her point of viewÂ” (McKeown and Thomas 1988, p. 12) is always anchored in self-reference, and is a key foundation of the Q-methodol ogy (Brown 1993). An objective methodology measures a person based on tests while a s ubjective methodology requires the person to actively measure the tests. In other words, while an objective methodology strives to measure certain dimensions, a subjective me thodology strives to understand the relative values of the dimensions. Q-methodology Â“co mbines the strengths of both qualitative and quantitative methodsÂ” (Brown 1996, p. 561) as it provides insight into the philosophic structures of subjective phenomen a, measures patterns within indi viduals, permits the structuring of hypotheses, and is a comprehensive approach for studying subjectivity (Brown 1993). Q-methodology is concerned with Â“operant subjectivityÂ”, which is the naturally occu rring subjectivity of the respondent (Brown 1980; McKeown and Thomas 1988). Although there are power ful statistics underlying Q-methodology, the method revolves around a science of subjectiv ity (Brown 1993). If a researcher were concerned about how objective traits are clustered toge ther, R-methodology would be used. On the other hand, Q-methodology is usef ul if a researcher is focused on clustering like-minded perceptions that are su bjective rather than objective.
84 Third, Q-methodology had different goals in that it seeks to capture a wide array of perceptions rather than make any claims regarding the population. On the other hand, R-methodology statistics seek to generali ze. Q-methodology does not require a high response rate since it is focused on unders tanding a types of re spondents (Stephenson 1953, p. 5; Brown 1980; Brown Under Review). For R-methodology, rigor is often associated with identifying a representative sample of the population of people; for Qmethodology, rigor is placed on the identifica tion of the items in the concourse as representative of the population of statements in a domain. Another difference is the sample si ze. For Q-methodology, Stephenson (1953, p. 5) points out Â“It is widely belie ved that it is essential to work with large numbers of cases in psychology, so that valid generalizations may be reached. We are to work, instead, with a single person, at the cal l of a theory.Â” Typically, there are around 50 respondents, such as Gottschalk (2001) which had 58 re spondents and Steelman and Maguire (1999) which had 68 respondents. There are also se veral studies such as Shields and Cragan (1981), which used 400 respondents and stated that the large sample st abilized the factor structures and permitted a discriminant analysis to identify respondent characteristics. Also, Brown (Under Review) identifies st udies that used larger samples. A fifth difference is the task of the re spondents. R-methodology entails rating and includes an assumption that the variables and their associated errors are independent of one another; Q-methodology is a ranking techni que in which each individual ranking is dependent on all other rankings in any gi ven Q-sort. Thus, with Q-methodology, the Q-
85 sort represents the respondentÂ’s coherent point of view on the concourse. The main problem with ratings is the lack of scal e use and indifference among the topics. For example, in a study of 20 issues on a scale of 1 to 10, the mean response rate ranged between 5.4 and 9.1 (Brancheau, Janz and Weth erbe 1996). The Q-sort is Â“a modified rank-ordering procedure in which stimuli are pl aced in an order that is significant from the standpoint of a person operating under specified conditionsÂ” (Brown 1980, p. 195). Since the Q-sort requires the respondentsÂ’ op inions and involves the task of ranking the items in the concourse, it is Â“an individualÂ’s conception of th e way things stand. As such, it is subjective and self-re ferentÂ” (Brown 1980, p. 6). An opinion necessitates an opinion-maker, implying some degree of se lf-reference. The Q-methodology preserves the respondentsÂ’ self-refere nce (Stephenson 1953). A sixth difference is the forced-distri bution feature of the Q-methodology. The shortcomings associated with traditional questionnaires and surveys are avoided with the forced-distribution feature that requires that participants sort statements into a quasinormal distribution (Nunnally 1978). Brown ( 1980) notes that the forced-distribution feature violates the independe nce assumption of statistical tests such as for ANOVA, yet points out that violating the forced-distrib ution requirement inva lidates choice and psychological significance that underlies self-reference. Although controversy remains between freevs. forced-distribution, Brow n (1971) concludes that mathematically, the distribution does not matter sin ce the factors are influenced far more by the ordering of the concourse than the type of distribution. Furthermore, Nu nnally (1978) points out that the Â“criticisms are not well justifiedÂ” (p. 615) for several reasons, such as Â“the exact
86 distribution form has little effect on the kinds of analyses which are made of the dataÂ” (p. 616). Despite the inherent differences be tween Q-methodology and R-methodology, the analysis of Q-methodology does bear some re semblance to cluster analysis. The Qsorting allows research ers to identify, categorize, and understand individual opinions and perceptions and to cluster groups based on perceptions (McKeown and Thomas 1988). Although groups are clustered, cluster analysis differs from Q-methodology as it does not revolve around the subjectivity that is part of the Qmethodology (Brown 1993). Also, the assumptions about the data is different since cluster analysis assumes independent responses between variables, whereas Q-me thodology assumes that all responses are dependent on all others. Thomas and Watson (2002) also differentiates cluster analysis from Q-sort, based on two reasons: 1) Cluster analysis strives to ach ieve representation through a large sample and random sampling Â– Qsort preserves the se lf-reference rather than achieving representation; and 2) Cl uster analysis strives for groups of objects with broad categorizations with the researcher assuming that group members are homogeneous and that within a margin of error, their res ponses are identical Â– Qsort creates groupings of people based on self-referent resp onses rather than on rese archer grouping criteria. As with any methodology, there are disadva ntages. Respondents may feel limited because of the assigned grid. Furthermor e, as in other types of questionnaire methodology, there may be some form of social desirability to sort the items in some way. Also, there is the potential of res pondentsÂ’ viewpoints changing over time. One
87 other drawback is that typical ly Q-methodology is not used to make generalizations about the population at-large, although this may be achieved under certain restrictions. Application of Q-Methodology Q-methodology has been used in numerous studies, as Brown (1986) notes that over 1,500 publications have used this me thodology and Brown (1993) notes that over 2,000 publications have used this methodology. Anderson (2003) notes that researchers have used Q-methodology in Â“communicati on, conflict resolution, counseling and intervention services, environmental resear ch, feminism, gender issues, information systems management, leadership skill, opera tions management, organizational culture and person-organization fit, personality, political psychology, political systems, psychology, public policy, risk training and quali ty assurance, strategic decision making, and even violence in relationshipsÂ” (p. 10). Although Q-methodology has been used much more in non-IS research, some IS research has used this methodology. Q-sort ha s been used as the main methodology in IS articles to understand key IS issues (Go ttschalk 2001), examine the competencies of software engineers (Turley and Bieman 1995), compare academicsÂ’ and practitionersÂ’ views on key IS issues (Pimchangthong, Plai sent and Bernard 2003), examine project managersÂ’ viewpoints (Tractinsky and Jarv enpaa 1995), study att itudes of analysts towards the development of information sy stems (Dos Santos and Hawk 1988), identify and examine groups of IT personnel (W ingreen, Blanton, Newton and Domino 2005), and identify and understand the importance of IS activities in organizations (Dos Santos
88 1989). Furthermore, Q-methodology has b een used in conjunction with other methodologies (Kaplan and Duchon 1988; Kendall and Kendall 1993). Thomas and Watson (2002) provides an in-depth explanation regarding Qmethodology and its use in IS literature a nd gives an example of its application. Furthermore, Thomas and Watson (2002) poi nts out that Q-sorting can help MIS interpretive researchers by minimizing the infl uence of the researcher on the subjects, allowing readers to check the re searcherÂ’s interpretative bias through examining the data themselves, and providing a subjective understanding of groups. Overview of Q-Methodology Steps Brown (2004) suggests three major step s for Q-methodology Â– 1) Establish the concourse; 2) Administer the Q-so rt; and 3) Factor Analyze the Q-sort. This is similar to the three steps and their components suggest ed by Thomas and Watson (2002) in Table 19 below: Table 19: Description of Steps Step Components of Step Questionnaire Development Represent the topic with Q-samples Decide the distribution Pilot/Full Data Collection Ensure self-reference Force the distribution Randomize Q-sample initial ordering Use a standardized format for Q-samples Pilot/Full Data Analysis Factor analyze to produce groupings Apply induction or abduc tion to produce insights Adapted from Thomas and Watson (2002, p. 154).
89 In order to further clarify and expand on these three steps, Figure 5 was created, similar to the methodology of Anderson (2003) Figure 5 highlight s three steps from Study 1. Study 2, which addresses research question #4, begins w ith the qualitative findings of Study 1. Based on the findings of Study 1, the concourse was established to be used in Study 2 as a basis for questionnaire development. As shown in Figure 5, questionnaire development is followed by pilo t data collection and analysis, and then primary data collecti on and analysis. Figure 5: Methodology Adapted from the methodology used by (Anderson 2003, p. 11). Q-methodology Study 2-Quantitative Research Study 1-Qualitative Research Qualitative Data Collection Qualitative Data Analysis Qualitative Findings Questionnaire Development Pilot Data Collection Pilot Data Anal y sis Full Data Collection Full Data Anal y sis Quantitative Findin g s
90 Step 1: Questionnaire Development Generally, for a concourse, there are 30-60 statements that are used with scales such as -4 to +4 or -5 to +5 (Brown 1980) although scales may be used only ranging from -2 to +2, such as in Thomas and Watson (2002). Study participants then sort these statements in a quasi-normal pattern. Brow n (1980) points out that the distribution may be more flat if topics are addre ssed that elicit st rong, opposite opinions. For the questionnaire development, the statements identifying reasons for resistance and the statements identifying beha viors were combined. This is in order to understand what statements were most represen tative of the userÂ’s experience during the implementation. Combining the reasons and beha viors led to a total of 29 statements that were in the concourse. It was determined that the 29 statements would be sorted from -3 to +3, as seen in Appendix F. A separate concourse was created to examine the desirability of various types of management strategies in the system implementation. This concourse had a total of 8 statements, w ith a scale of -2 to +2. Appendix H shows the various items and their corresponding concourse statement. Following the recommendation of Brown ( 1993) that a Q-sort should be followed where possible with an elabor ation of the respondentsÂ’ point of view, qualitative questions followed both of the Q-sorts. Th ese qualitative questions asked respondents why they chose the statements that were mo st extreme. The res pondentsÂ’ elaboration on the ranking of concourse items helped to further understand the respondentsÂ’ points of view.
91 Step 2: Pilot Data Collection The second step is the pilot test of the questionnaire utilizing a small convenience sample of ES users. The ai m of the pilot test is to te st the questionnaire and obtain feedback from the respondents regarding the content, length, and structure of the questionnaire. For the pilot data collection, 110 ES users in one or ganization were sent questionnaires. There were 35 questionnaires that were returned (32% response rate). Four of these questionnaires were not us ed because they were incomplete. Step 3: Full data collection The third step was the colle ction of the questionnaire da ta. As noted previously, most of the Q-methodology studies have sa mple sizes under 100. For the full data collection, a larger sample size was sought out for the purposes of understanding types of users in multiple organizations. A convenience sample was used for the data collection. A total of 317 ES user groups were emailed and an email was sent to three user group listserves. The emails sought out a person w ho would be willing to participate in the study by agreeing to distribute the questionnaire to 15-20 users within their organization. Only organizations that had rolled out a system less than three years ago were included in the data collection. There were a total of 24 members from these user groups that agreed to distribute questionnaires. Each of thes e members who agreed to participate received a packet of 20 questionnaires and business reply envelopes along with instructions on distributing the questionnaires (Except for tw o members, who received the questionnaire via email). Several weeks after sending out the questionnaire, a follow-up email was sent
92 to each of these members. It was found that a total of 354 questionnaires were actually distributed to ES users. 128 of these ques tionnaires were returned, which shows a 36.2% response rate from ES users who actually r eceived the questionnaire. However, since 480 questionnaires were sent out to user group members to distribute the questionnaires, there was a 26.7% response rate to questionnaires that were sent out. Step 4: Analysis of Pilot/Full Data Collection For the analysis of the data, the Q-met hodology uses factor analysis that accounts for variance shared among respondents. Genera lly the number of factors are selected if they have an eigenvalue greater than one, although Brown (1986) notes that it is not the absolute cut off value in the selection of f actors. Brown (1980) di scusses theoretical vs. statistical significance of factors and states that Â“statistical criteria may yield a factor that is not statistically significant, or Â… may fail to extract a factor that is highly important theoretically. The general prin ciple would therefore seem to be that theory and judgment must be relied upon in the absence of othe r criteriaÂ” (Brown 1980, p. 43). The factors that are derived are groups of study re spondents that have similar Q-sorts. In the analysis of the full data co llection, Thomas and Watson (2002) was followed, which recommends that the analysis of the Q-sort should contain: 1) Factor loading arrays; 2) normalized factor scores; 3) the statement(s) on which arrays load. This can help the reader to both check and reinterpret the researcherÂ’s logic, thus minimizing any errant effects of the resear cherÂ’s judgment on th e interpretation of factors. Furthermore, Thomas and Watson (2002) recommends that the researcher should
93 use eigenvalues and a detailed factor analysis procedure to limit data manipulation. The Varimax Rotation seems to be the most co mmon procedure used in Q-sort studies (McKeown and Thomas 1988). However, judgment al rotation is widely used if there are good reasons to abandon Â“simple structureÂ” (McKeown and Thomas 1988, p. 52). Researchers using the Q-methodology often us e Varimax or Quartimax rotation, although Q-methodology allows the judgmental rotation as l ong as it is in step with theory, which is termed Â“theoretical rotationÂ” (Bro wn 1980, p. 39). McKeown and Thomas (1988) further points out that sometimes the centr oid method is employed since it frees the researcher to approach the problem with Â“abductive logicÂ” (McKeown and Thomas 1988, p. 53). Reliability/Validity Brown (1980) points out that individualsÂ’ responses are at issue, not the operational definition, and thus Â“The concept of validity has very little status [in Qmethodology] since there is no outside crit erion for a personÂ’s own point of viewÂ” (Brown 1980, p. 174-175). Although it has been s uggested that a comparable Q-analysis and R-analysis suggests some degree of validity (Brouwer 1992-1993), Q-methodology research overall has treated validity as ir relevant since the me thodology is striving to understand the relative opinions of respondent s. Dennis (1988) points out that the reliability and validity of Q-methodology lies in the data rather than the measure and that Â“ascertaining construct or pr edictive validity are inappropr iate and irrelevantÂ” (Dennis 1988, p. 413). Q-methodology is related more to qualitative research rather than
94 quantitative research in its approach to validity, since there is no substitute to a respondentÂ’s point of view (Dennis 1988). Despite the minimal importance Q-me thodology researchers have placed on validity and reliability, validity has been es tablished in two ways Validity has been established in the development of the concourse in that it wa s drawn from the literature and interviews (Dennis 1988). Also, content va lidity has been established in that the sample statements were reviewed by domain experts and tested in a pilot study. In regards to content validity, Dennis (1988) was followed as it recommends that domain experts should be used to ensu re: Â“(1) items included in the Q-set constitute an adequate representation of the domain, (2) one cell is not overrepresented to the underrepresentation of another, and (3) the items are relevant to the domain studiedÂ” (Dennis 1988, p. 414). Results of Pilot Study The purpose of the pilot data collection is to perform a preliminary check on the data and examine if types of users emerge fr om the data. For the limited amount of data (n=31) that was in the pilot test, 8 factors we re selected for further analysis, based on the eigenvalues and the percentage of variance explained. Th ese eight factors explained a total of 72% of the variance. A Varimax rotation, commonly used in Q-methodology, was used to extract the user types. Because of the small sample size, one of these groups only had one person in it, which leads to uncerta inty regarding whether this is a type of
95 user or merely an individual. Thus, for the full-data collection, a larger sample was collected so that groups cont ain multiple respondents. As for the factors that emerged, the two factors that explained the most variance (one factor explained 14% and the other factor explained 11 %) had no resistance behaviors that were representa tive of their ES experiences. Three other factors had only one resistant behavior as repr esentative of their experience (either system avoidance or challenged the system plan). The other thr ee factors had multiple resistant behaviors, such as decreasing productivity, complaini ng, not wanting to learn the system, and avoiding the system. Based on how respondent s indicated the resist ant behaviors were representative of their ES e xperience, Figure 6 was derived. Figure 6 categorizes these 8 factors based on the degree of resistance (the degree to which they indicated resistance behaviors were representative of their ES experience) and the type of resistance (how active or passive the resistant behaviors we re that were representative of their ES experience):
96 Figure 6: Categorization of Re sistance by Factor Number Degree of Resistance Least Greatest Active Type of Resistance Passive An analysis of preferred management st rategies was also conducted. From the sample of respondents in the pilot study, it was clearly exhibited that management expertise was the most preferred management strategy. There are two strategies that are tied for second Â– training and manageme nt listening/responding to users. Results of Full Data Collection Preliminary Tests In order to check for bias based on the orde ring of the concourse statements, one of two potential questionnaires were randomly distributed to the respondents. The concourse statements in the second questionnaire we re randomly changed around, so the ordering was different. A t-test was then performed to examine if there was any difference in the respondentsÂ’ ranking of the concourse items based on the questionnaire version. As 1 5 4 2 3 6 7 8
97 shown in Table 20 below, only two of the 37 concourse statements showed a significant difference at alpha=0.05. However, to hold the experiment-wise error ra te at an alpha of 0.05, significance is shown at a value below 0.05/37, or 0.0014. There was no statistically significant difference found at alpha = 0.0014. Thus, the ordering of the concourse statements likely made no difference to respondents.
98 Table 20: T-test for Equality of Means among Two Questionnaire Versions Concourse Item tdfSig. (2-tailed) CON1 1.1311250.260 CON2 1.2431250.216 CON3 2.1291250.035 CON4 -0.7011250.485 CON5 -0.8361250.405 CON6 -0.8041250.423 CON7 0.9841250.327 CON8 -2.5981250.011 CON9 -1.1911250.236 CON10 1.3061250.194 CON11 0.4441250.658 CON12 -0.6941250.489 CON13 1.9391250.055 CON14 -1.4851250.140 CON15 -0.0961250.923 CON16 -1.1321250.260 CON17 0.5881250.558 CON18 1.2541250.212 CON19 0.6481250.518 CON20 -1.7431250.084 CON21 -0.2961250.768 CON22 0.3791250.706 CON23 -0.7781250.438 CON24 -0.6841250.495 CON25 1.0901250.278 CON26 0.0301250.976 CON27 -1.0651250.289 CON28 -1.3241250.188 CON29 1.2411250.217 CON30 -1.2731250.205 CON31 0.2391250.811 CON32 0.7791250.437 CON33 0.5571250.578 CON34 -1.4781250.142 CON35 0.1711250.864 CON36 1.3061250.194 CON37 0.0781250.938 A second t-test was performed in order to check for non-respondent bias. There were some respondents who only filled out the dem ographic portion of the questionnaire rather
99 than filling out the whole quest ionnaire. In order to check that the demographics of the responders who completely filled out the que stionnaire were no diffe rent than those who did not completely fill out the questionnaire, a t-test was conducted, as shown in Table 21 below. None of the demographic variables were shown to be significa nt in this t-test. The details of the demographic information of the respondents are displayed in Appendix J. Table 21: T-test for Equality of M eans among Respondents who filled out the Questionnaire fully versus those who did not Demographic Item tdfSig. (2-tailed) Gender -1.4741490.143 Education 1.0251480.307 Years in Position 0.3821450.703 Years at Employer 1.2331450.220 Age -1.5391330.126 Position 0.4581320.647 # of Employees 1.2171470.225 Org. Industry 0.8481490.398 System Scope 1.9721440.051 System Vendor 0.9571480.340 Days of Training -0.7981360.426 Days before Usage -0.0981290.922 The Q-sort responses indicated on each que stionnaire were ente red into PQMethod, a statistical program specifically tailored fo r use with Q-methodology. This software is often used with Q-methodology studies. In the analysis of the data, the intercorrelations of the Q-sorts were calculated, then factor analyzed. A principal components factor analysis was first conducted to view the eigenvalues and percentage of variance explained by each factor, shown in Table 22 below.
100 Table 22: Principal Components Factor Analysis Factor # Eigenvalues Percent age Cumulative Percentage 1 30.22 23.61 23.61 2 7.91 6.18 29.78 3 7.39 5.77 35.56 4 6.63 5.18 40.74 5 6.41 5.01 45.75 6 5.35 4.18 49.93 7 5.15 4.02 53.95 8 5.00 3.90 57.86 For the purposes of this research, eight f actors were selected, explaining a cumulative 58% of the variance. Eight factors were sel ected for two reasons: 1) there was a slightly larger gap between the eigenvalues of the ei ghth and ninth factors th an there was between the other factors; and 2) eight is a sufficient number of groups to analyze, since the purpose of this research is to identify the ma in groups that form from the data analysis, not to explain every group/fact or that exists. The eight f actors identified were then rotated using a Varimax rota tion, commonly used with Q-me thodology studies to identify the factors that maximize the amount of variance.
101 Research Question 4a: User Groups Table 23 below shows the factors that were identified using the PQMethod software with Varimax rotation. For the analys is of the factors, the concourse statements that were most representative of the user gr oupsÂ’ ES experience were identified (-3 is the most representative of their experiences, +3 is the least representative of their experiences). The highlighted factors in Table 23 are the top third of concourse statements that respondents indicated were representative of their ES experience.
102 Table 23: Factors of User Groups (Normalized Factor Scores and Statement Rankings) Concourse Statement REAS-Uncertainty-0.91300.2915-0.79271.007-0.59290.63131.1160.2121 REAS-Lack Input-1.5834-0.5425-0.8028-0.5427-0.10230.25161.392-0.4826 REAS-Lose Control-0.9129-1.0130-0.29221.283-0.76301.0951.315-0.9631 REAS-Self Efficacy0.8080.2816-0.3823-0.0620-0.2726-1.7837-0.53280.2520 REAS-Changed Job-1.7536-1.0332-0.4524-0.2925-2.2437-1.5634-1.1932-1.4234 REAS-Workload-1.9837-0.0120-1.5636-0.9530-2.0736-1.1030-1.5034-0.7728 REAS-Technical Problems-0.85280.2914-0.9630-1.4433-0.9732-1.7536-1.9037-1.3133 REAS-Environment-0.29240.7911-0.8129-1.2832-0.0522-1.41311.353-0.8830 REAS-Lack of Fit-1.01310.809-1.3034-1.9636-0.9231-0.7729-1.4133-1.1132 REAS-Communication-0.5526-0.0623-0.2320-1.63350.28150.19200.50120.2919 REAS-Training-0.53251.305-0.7026-0.84290.7580.22191.344-0.3225 REAS-Complexity-0.14231.136-1.4435-0.9631-1.0733-0.6227-1.7636-0.8529 BEH-Challenged-0.0922-0.80270.1217-0.20230.2118-0.0124-0.22230.3916 BEH-Dont Follow Processes0.6113-1.05340.61130.37160.9360.679-0.26240.4015 BEH-Shadow System0.0720-1.83360.61140.85111.1040.64120.00170.697 BEH-Old System0.05210.33130.6712-0.23240.50121.016-0.34250.6010 BEH-Avoid0.65110.8480.9580.40150.67100.23170.49130.4812 BEH-Inappropriately1.007-0.54260.37150.25170.09201.692-0.16210.659 BEH-Hack1.194-1.04331.1560.87100.25171.234-0.05200.3217 BEH-Refusal1.791-1.02310.78111.2440.36140.67100.06160.934 BEH-complain0.30161.383-0.48250.4714-0.32270.1023-0.40260.4114 BEH-Defensive0.7590.02170.35160.17180.27160.6790.06160.726 BEH-Demotivated0.20180.7911-0.0619-0.1421-0.13240.64120.73110.1322 BEH-Less productivity0.48150.00190.8790.72120.03210.2915-0.05200.845 BEH-Impatient1.105-0.0322-0.2521-0.56280.3613-0.5826-0.1922-0.5827 BEH-Quit0.1619-0.49241.5331.5111.712-0.1225-0.05200.0424 BEH-Dont want to learn system0.55141.3441.1650.9480.60110.15211.0971.093 BEH-Turnover Intention0.29170.57121.3340.8991.015-0.72280.36140.0424 BEH-Procrastinated0.7210-0.81280.05180.09190.6990.11220.8790.2918 MGMT-Communication-1.13320.01181.057-0.36262.001-1.6435-1.72351.202 MGMT-Feedback-1.6335-0.0221-1.14331.175-0.39280.827-0.64290.678 MGMT-Provide Support0.64120.877-1.06311.3920.7970.2219-1.05310.4113 MGMT-Training1.303-1.8537-2.0337-0.17221.283-1.44321.058-2.1937 MGMT-Incentives1.742-0.82290.82101.006-2.0335-1.48332.1312.851 MGMT-Clear Plan-0.8127-1.8336-1.0632-2.08370.14191.692-0.64300.5211 MGMT-Expertise-1.25331.6021.692-1.5734-0.27250.30140.7410-1.5035 MGMT-Customizations1.0262.1311.6910.6713-1.84341.473-0.5227-2.0536 12345678
103 In Table 23 above, the responses greatly varied depending on the group. In group 1, resistant behaviors were not among the top third of concourses selected. This group identified various reasons for resistance and management strategies to minimize user resistance, but did not exhibit resistant behaviors. Group 2 exhibited the most resistant behaviors. Six of the seve n behaviors highlighted are ac tive behaviors, with only one behavior that is passive (pro crastination). From management Â’s perspective, this is the group that is most resistant. In order to minimize the resistance, the top three management strategies identified by this gr oup are training, incentives and a clear plan. For group 3, only the overt, passive behavi or of complaining was identified. To minimize the complaining, management can pr ovide better feedb ack, support, training, and a clear plan. Group 4 exhi bited only the covert, passive behavior of impatience as part of the top behaviors identified. This group identifies that better communication, a clearer plan, and management expertise woul d have been the most useful management strategies. Group 5, like group 3, only has complaining as the resistant behavior exhibited. However, there we re different reasons for us er resistance identified among these two groups as well as different ma nagement strategies. Group 6 identified impatience, turnover intention, and actual turnov er (quitting) as the most representative behaviors. This group identified manageme nt communication, training, and incentives as the most important management strategies th at should have been implemented better. Group 7 had complaining and using the old syst em as the most representative behaviors, and identified five management strategies This group had system complexity and technical problems as the top reasons for user resistance. Group 8 identified impatience
104 as the top resistant behavior, similar to group 4, but had different reasons for user resistance and management strategies. To further understand each of these groups, the qualitative portion of the questionnaire was analyzed to determine if these groups made sense based on the supporting qualitative data. The qualitative portion is comprised of three questions asking the reasons for why the respondent choose the most and least representative statements. Each of the eight groups had res pondent statements that supported the results of the quantitative analysis shown in Table 23 above. Appendix I pr ovides examples of qualitative quotes from each of the eight groups. An additional analysis was conducted to examine if any of the demographic variables might have a sta tistically significant effect on the user group. Thus, an ANOVA was conducted to determine if there are demographics th at have significant effects between factor groupings. As s hown in Table 24 below, no statistically significant effects were f ound at alpha=0.05. Although demographic information could be provided for each group, this analysis indicated that none of the groups have statistically significant differences from the demographics of the overall questionnaire. Therefore, no groups were found to have any demographics different than those found in Appendix J, which provides tables on all the demographic data.
105 Table 24: Analysis of Vari ance based on Factor Grouping Sum of Squares df F Sig. Between Groups 3.007 1.898 0.083 Within Groups 15.3668 Gender Total 18.3675 Between Groups 11.417 1.699 0.124 Within Groups 65.2668 Education Level Total 76.6775 Between Groups 123.697 1.329 0.251 Within Groups 877.2666 Years in Current Position Total 1000.9573 Between Groups 183.107 0.567 0.780 Within Groups 3046.6666 Years in Organization Total 3229.7673 Between Groups 9.657 1.502 0.183 Within Groups 56.9362 Age Total 66.5969 Between Groups 15.417 1.015 0.430 Within Groups 132.2461 Position Total 147.6568 Between Groups 13.127 1.361 0.236 Within Groups 93.6268 System Vendor Total 106.7475 Between Groups 5644.627 1.203 0.315 Within Groups 41568.3262 Days of Training Total 47212.9469 Between Groups 101865.137 1.360 0.239 Within Groups 641950.1160 Days Between Training and Using Live System Total 743815.2467 Research Question 4b: Resisting Groups Another step was performed on each of thes e eight groups in order to understand the resistant behaviors of each group. All of the resistant behaviors were categorized by the Overt-Covert-Active-Passive 2x2 matrix de veloped by (Bovey and Hede 2001). Based on the ranking of all the re sistant behaviors for each gr oup, the relative difference was calculated for each cell of the 2x2 matrix. This was calculated for each group by adding
106 1 to the Z-score for each of th e four resistant behaviors (so that there would not be any negative values). The results of the rela tive resistant behaviors among the groups are shown in Figure 7 below. Clearly, group 2 show ed the most resistant behaviors, and in particular, overt-active behaviors. This is followed by group 7, which had a high degree of overt-active, covert-active, and overt-passi ve behaviors. Figure 7: Resistant Beha viors by Group Number 0 0.5 1 1.5 2 2.5 3 3.5 4 4.5 12345678 Group #Level of Resistant Behavior Covert-Passive Overt-Passive Covert-Active Overt-Active Research Question 4c: Management Strategies In order to determine what management strate gies are identified by users that will be most effective in minimizing the level of resi stance, Z-scores were calculated. As shown in Table 25 below, a clear concise plan is the most desirable management strategy for users. The second most desired strategy is for the managers to have more expertise in the
107 system and in rolling out the system. The third most desired strategy is better top-down communication. Table 25: Rank Ordering of Management Strategies Z-Score Concourse Statement -0.805 MGMT-Clear Plan -0.680 MGMT-Expertise -0.227 MGMT-Communication -0.086 MGMT-Feedback -0.039 MGMT-Training 0.453 MGMT-Customizations 0.531 MGMT-Provide Support Management Strategies 0.797 MGMT-Incentives In addition to the management strategies as shown in Table 25 above, there are several reasons for user resistance that emer ged as the most important reasons. As shown in Table 26 below, the additional workload was the most significant reason for user resistance, followed by a lack of fit, technical problems, and changed jobs. In regards to resistant behaviors, challenging the management plan was the most representative of ES usersÂ’ experiences, followed by impatience, complaints, and then trying to use the old system.
108 Table 26: Rank Ordering of Reasons for User Resistance and Resistant Behaviors Z-Score Concourse Statement -1.836 REAS-Workload -1.516 REAS-Lack of Fit -1.500 REAS-Technical Problems -1.406 REAS-Changed Job -1.141 REAS-Complexity -0.703 REAS-Environment -0.633 REAS-Lack Input -0.508 REAS-Communication -0.195 REAS-Training -0.078 REAS-Uncertainty -0.039 REAS-Self Efficacy Reasons for User Resistance 0.023 REAS-Lose Control -0.203 BEH-Challenged -0.047 BEH-Impatient 0.039 BEH-Complain 0.344 BEH-Old System 0.352 BEH-Defensive 0.359 BEH-Procrastinated 0.391 BEH-Unmotivated 0.578 BEH-Inappropriately 0.594 BEH-DonÂ’tFollow Processes 0.641 BEH-Less productivity 0.758 BEH-Shadow System 0.758 BEH-Avoid 0.828 BEH-Hack 0.859 BEH-Turnover Intention 0.992 BEH-DonÂ’twant to learn system 1.023 BEH-Quit Resistant Behaviors 1.344 BEH-Refusal
109 CHAPTER V. CONCLUSIONS AND FUTURE DIRECTION The implementation of an ES in organizations has forced many employees to adopt a system that changes their job duties a nd reward structure. These mandatory, roletransforming systems have faced considerable resistance, though often it is covert. There are obviously contributing factors that aff ect employee responses to a system, as mentioned earlier in this paper, such as l ack of top management support and project team competence (Akkermans and Van Helden 2002). However, even when the appropriate planning, analysis, and design have been pe rformed, there are still many times that implementations have failed or faced unwarrant ed difficulties because of user resistance. The following sections discuss the results from both studies, the cont ributions of the two studies, the limitations of this dissertation, and future directions for this research. Discussion of Study 1 Underlying reasons for user resistance, resistant behaviors, and management strategies to minimize user resistance were found and described in Study 1. Furthermore, the constructs were classified into categor ies. These findings s howed several unique aspects of an ES change that do not exist in organizational cha nge not facilitated by technology. For example, the ES-enabled ch ange added the complexity of technical problems and employees needing to learn a co mplicated system. Furthermore, a portion of the employees face a lack of computer self-efficacy and additional skills are required
110 for jobs, such as needing to know how to perform queries. Also, in regards to management strategies, employees expect mana gement to have expertise in the system, and to perform system customizations. Th ese expectations would not exist in an organizational change not facilitated by technology. Markus (2004) revolves discusses organiza tional change facilitated by technology and differentiates technology enab led change versus redesigni ng organizational structures without technology. In differen tiating these two type s of change, Markus (2004) suggests that there are different target outcomes, solutions, role of managers, and key success factors. Using the term Â“technochangeÂ” to describe the use of technology to drive organizational change, Markus (2004) descri bes Â“technochangeÂ” as different from most IT projects, since many IT projects merely ad just work processes minimally, rather than driving organizational change. Furthermor e, Markus (2004) suggests that misuse, nonuse, and failure risks are very high with t echnology-enabled organizational change, yet IT project management approaches do not focus on these issues. Comparing the Results to Other Studies As mentioned in the liter ature review, there was no publication found that conducts a research study to understand the underlying reasons for why users resist. However, there are several publications th at have discussed potential reasons for resistance based on literature reviews. In the following paragraphs, first of all the findings of reasons for user resistance ar e compared to two publications. Next, the findings of management strategies to mini mize user resistance are compared to one
111 publication. Finally, the results are compared to user acceptan ce literature. As seen in the comparisons, this dissertation has identified several constructs not identified in the publications, and modifies some constructs th at were discussed in these publications. The first comparison is with Hirschheim and Newman (1988), which focuses on ten reasons for user resistance that are ba sed on a literature review. Table 27 below identifies the constructs s uggested from the Hirschheim and Newman (1988) literature review and the definition of each construct. These are compared with the reasons for user resistance found in the results of this dissertation. Table 27: Comparison with Hirschhe im and Newman (1988) Â– Reasons Hirschheim and Newman (1988) Construct Hirschheim and Newman (1988) Definition This Dissertation Innate Conservatism Â“reluctance to change the status quoÂ” (p. 399) Job/Job Skills Change Lack of Felt Need Â“individualsÂ…have not been convinced of the merits of the changeÂ” (p. 399) Communication Lack of Involvement in the Change Â“Individuals [feel] that they have been excluded from the decision-making process associated with the changeÂ” (p. 399) Input Redistribution of Resources Â“disruption of the status quo [including] departmental budgets, equipment, staff, and territoryÂ… status, salary, roles, etc.Â” (p. 399) Control/Power Organizational Invalidity Â“mismatch between spec ific features of system design and characteristics of the existing organizationÂ” (p. 400) Lack of fit & Facilitating Environment Poor Technical Quality Â“systems which areÂ…Â‘unfriendlyÂ’, unreliable, lack functionality and slowÂ” (p. 400). System Complexity, Technical Problems Uncertainty Â“see change as a threat and possess a fearÂ” (p. 399) Uncertainty Poor Training Â“users are no t properly trained to use the systemÂ” (p. 400) Training
112 As seen in the table, there are eight cons tructs that are similar to the constructs found in this dissertation, which are discusse d below. There are also two constructs found by this dissertation not found in Hirschheim and Newman (1988) and two constructs discussed in Hirschheim and Newman (1988) not found in the dissertation, which also are discussed below. The first construct identified, Â“Innate Conser vatismÂ”, is a different construct from Â“Job/Job Skills ChangeÂ”, but these reasons have similar roots. The job/job skills change may result in user resistance because of the innate conservatism, but since the job/job skills change is what initiates the resistance, it is the underl ying reason for user resistance that management can control. Â“Lack of Felt NeedÂ” and Â“Com municationÂ” also are differe nt constructs, but have similar roots. Poor communication may lead to a lack of felt need since users do not understand the benefits of the system or why it is being implemented. Thus, poor communication is likely the underlying reason for the lack of felt need experienced by the users. The constructs Â“Lack of Involvement in the ChangeÂ” and Â“InputÂ” are similar. However, Hirschheim and Newman (1988) is more focused on involvement on the system decision and participation in developm ent in the system. This dissertation found that the users were not very interested in th e initial system decision or participation in
113 development of the ES; rather, they were interested in management seeking their thoughts and opinions for the implementation. There is also similarity between Â“Redistribution of ResourcesÂ” and Â“Control/PowerÂ”. However, Hirschheim a nd Newman (1988) is focused more on power issues whereas this dissertation includes both loss of power and loss of recognition as an expert. Thus, the findings of this disserta tion lead to a slightly broader construct. For Â“Organizational InvalidityÂ” and Â“Lack of Fit & Facilitating EnvironmentÂ”, Hirschheim and Newman (1988) identified one broad construct. However, this dissertation identified two diffe rent constructs which both are part of what is described by Â“Organizational InvalidityÂ”. The difference betw een the two constructs identified in this dissertation is that Â“Lack of FitÂ” is the process problems th at occurs when new processes are implemented, but Â“Facilitating Environmen tÂ” is the affect of the organizational culture and the ability of an orga nization to infuse a technology. In regards to Â“Poor Tech nical QualityÂ” and Â“Syste m Complexity & Technical ProblemsÂ”, the construct discussed by Hirs chheim and Newman (1988) encompasses the two constructs of system complexity and tech nical problems that ar e described in this dissertation. These are separated in this dissertation since it was found that system complexity often exists with an ES even if technical problems do not exist. For other types of systems, there may be technical problems even though the system is not
114 complex. Therefore, system complexity a nd technical problems should be addressed as two separate reasons for user resistance. The last two constructs, Â“Uncertainty Â” and Â“Poor TrainingÂ” are identical constructs to the Â“Uncertain tyÂ” and Â“TrainingÂ” found in this dissertation. For many systems, these two reasons for user re sistance are common and management should address these issues. There were also two reasons identified by Hirschheim and Newman (1988) not found in this dissertation. One reason not found in this dissertation is Â“Lack of Management SupportÂ”, which Hirschheim a nd Newman (1988) defines as failure of management Â“to support and encourage the changeÂ” (p. 400). Th is dissertation found Â“Providing Help/SupportÂ” as a management stra tegy useful in minimizing user resistance that arises from various reasons for user resi stance, rather than identifying it as a reason for user resistance. The second reason iden tified by Hirschheim and Newman (1988) not found in this dissertation is Â“P ersonal Characteristics of th e DesignerÂ”, which is defined as Â“difficulties that many system developers have in interacting with usersÂ” (p. 400). This was not found in the interviews becau se every organization implemented an ES software package rather than designed their own system. Finally, there are two re asons found in this dissertation not identified by Hirschheim and Newman (1988). These two reasons for user resistance are Â“SelfEfficacyÂ” and Â“WorkloadÂ”. These are important reasons for user resistance, and very
115 applicable to an ES implementation, but were not identified in the description of Hirschheim and Newman (1988). One other study that the dissertation resu lts are compared to is Markus (1983), which discusses how there are some user attributes, technical attributes, and power/sociotechnical issues that affect the leve l of resistance. In pa rticular, this article focuses on power issues, discussing how user resistance remains until the users feel compensated for the lost power. In Table 28 below, the results of this dissertation are compared and contrasted with the reasons fo r user resistance identified by Markus (1983). As seen in the table, there are some similar issues between this dissertation and Markus (1983), but there are also reasons fo r user resistance found in this dissertation that were not identified in Markus (1983).
116 Table 28: Comparison with Markus (1983) Â– Reasons Issues identified by Markus (1983) Issues identified in dissertation Discussion User Attributes Â– cognitive style, personality traits, human nature Individual Issues Â– Uncertainty, Input, Control/Power, SelfEfficacy The dissertation results identify specific constructs that can be used to measure individual issues rather than identifying general categories of reasons Technical Attributes Â– Lack of userfriendliness, poor human factors, inadequate technical design or implementation System Issues Â– Technical Problems, Complexity Although these cover similar areas, the dissertation identifies specific constructs that can be measured Power/Sociotechnical Issues Â– Interaction of the system and the context Organizational Issue Â– Facilitating Environment The power that Markus (1983) discusses is included in the Individual Issues for the dissertation, because of its dependency on the individualÂ’s desire for power/control. The sociotechnical issue is similar to the facilitating environment identified in the dissertation Process Issues Â– Job/Job Skills Change, Workload, Lack of Fit; Organizational Issues Â– Communication, Training Markus (1983) does not address the process issues that are inherent to an ES implementation or the organizational issues of communication and training In regards to management strategies, Kotter and Schlesinger (1979) discusses management strategies to deal with resistance, and also prov ides examples of situations where combinations of management strategies would be used. Table 29 below compares the suggested management strategies of Kotter and Schlesinger (1979) with the management strategies identified in this dissertation. As seen in the table below, Kotter and Schlesinger (1979) addresse s two management strategies that were not found in this
117 dissertation (Manipulation/C ooptation and Explicit/Implic it Coercion). Although there are situations in which these strategies may be useful in minimizing resistance, Kotter and Schlesinger (1979) also warns that these strategies can lead to future problems if employees feel that they are manipulated and that the strategies are risky, since employees may be angry at change initiators. Due to the long-term results of such strategies, these strategies are not used of ten. For example, Hunton and Beeler (1997) notes that coerced participation may be ine ffective in gaining th e positive involvement, responsibility, intention to use, and ownershi p that ultimately affects system success. From a userÂ’s perspective, thes e two strategies are not desired, and thus were not found in the interviews that were c onducted with users in Study 1.
118 Table 29: Comparison with Kotter and Schl esinger (1979) Â– Management Strategies Issues identified by Kotter and Schlesinger (1979) Issues identified in dissertation Discussion Education and Communication Top-down Communication Kotter and Schlesinger (1979)Â’s explanation of Education and Communication focused on informing employees about the change, which is similar to Communication in this dissertation Participation and Involvement Listen to Feedback These are very similar issues, which is basically involving employees in the change Facilitation and Support Training; Provide Help/Support These are similar. Kotter and Schlesinger (1979) includes emotional support when referring to Â“supportÂ”, which is included in Â“Provide Help/SupportÂ” in the Dissertation Negotiation and Agreement Incentives Kotter and Schlesinger (1979) focuses more on working with unions and thus providing incentives an d negotiating with the union in order to support the change Manipulation and Cooptation Not in Dissertation Kotter a nd Schlesinger (1979) suggests including a leader, su ch as a union leader, in a desirable role in the change in order to gain support from other employees Explicit and Implicit Coercion Not in Dissertation Kotter a nd Schlesinger (1979) suggests implicitly or explicitly threatening employees with a potential loss of job or lack of promotion Not in Kotter and Schlesinger (1979) Clear Consistent Plan Although Kotter and Schlesinger (1979) refers to educating users of the plan when referring to Â“Education and CommunicationÂ”, it does not mention a clear, consistent plan in order to minimize resistance Not in Kotter and Schlesinger (1979) Management Expertise Kotter and Schlesinger (1979) does not suggest increasing th e understanding of managers in regards to the processes and/or system Not in Kotter and Schlesinger (1979) System Customizations Since Kotter and Schlesinger (1979) is just referring to organizational change, system issues are not addressed.
119 The results also are compared to a user acceptance study. Perhaps the most comprehensive user acceptance study is Ve nkatesh et al. (2003) which includes 32 potential independent variable s based on eight different models. These variables are examined and synthesized to develop the Unified Theory of Acceptance and Use of Technology. Part of the study examines the effect of the 32 independent variables on intention to use, examined in a mandatory a doption setting. The resu lts are displayed in a table that displays the signif icance of these independent vari ables on intention, tested in three different time periods (Venkatesh et al. 2003, p. 441). In Table 30 below, the independent variables that were found to be significant in at least two of the three time periods are shown and compared to the reasons for user resistance found in this dissertation.
120 Table 30: Comparison with Venkatesh et al. (2003) Significant Independent Variables that Predict Intention Comparable Reason for User Resistance Found in Dissertation Comparison of the results Attitude Toward Using Technology None Although attitude is likely to affect resistant behaviors, only the root causes of user resistance were sought out in this dissertation Subjective Norm Facilitating Environment Although these are different constructs, there are external forces that affect the attitudes and behaviors of users Perceived Usefulness Communication Through the communication of the benefits of the ES, users form their opinion on its perceived usefulness Perceived Ease of Use Communication, Technical Problems, Complexity The communication to the users as well as the technical problems or complexity of the system likely affect the userÂ’s perceived ease of use Extrinsic Motivation None Extrinsic motivation was not identified as a root cause of user behaviors in both the interviews or questionnaires Intrinsic Motivation None Intrinsic motivation was not identified as a root cause of user behaviors in both the interviews or questionnaires Job-Fit Lack of Fit, Job/Job Skills Change These are similar constructs. When the system does not fit th e job, or users need to develop new skills or perform new tasks, there is likely to be negative behaviors Social Factors Facilitating Environment These are similar constructs as they both revolve around the environment of the user Relative Advantage Workload, Job/Job Skills Change Perceptions of relative advantage can stem from the changes in workload, job tasks, or job skills Image None This construct was not found in any of the interviews Outcome Expectations Control/Power Outcome expect ations is a broad category that includes the gain /loss of control or power Self-Efficacy Self-Efficacy Same constructs Anxiety Uncertainty Similar constructs since uncertainty is a cause of anxiety
121 As shown in Table 30 above, there are some si milarities to what user acceptance research has proposed as the predictors of intention and what this di ssertation research has found to affect user resistance. Despite some si milarities, Table 30 also provides comments on the differences between the constructs. Fu rthermore, Input and Training were found to be reasons for user resistance, but are not re lated to any of the constructs identified in Venkatesh et al. (2003). Previ ously in this dissertation it wa s stated that the opposite of user resistance is not user acceptance, since users can resist while seemingly accepting or using the system. However, there are some si milarities between the driving forces of user acceptance and user resistance, as shown in Table 30. The user acceptance research stream may benefit by considering some of the reasons for user resistance as antecedents to a userÂ’s intention to use a system. Managing the Reasons for User Resistance The first reason for user resistance describe d in the results sec tion is Uncertainty. Users often are unclear of the fu ture and view the system as a potential threat to their job and/or work life. Management can addre ss this issue through top-down communication and clear, consistent plans. Through conve ying important details and clearly addressing issues such as why the system is being impl emented and the extent of the project, users will better understand what is required and the changes that will occur. A clear vision may entail promoting the system as able to provide seamless integration among the multiple departments and numerous employees connected through the ES. In an ES implementation, a Â“sponsorÂ” can also be usef ul in convincing those involved how the benefits of the ES outweigh the costs. Through credibility and trust, it is likely that this
122 leader can create strong alliances throughout the organization. Mana gement should also clearly demonstrate their commitment and support for the ES implementation since a long-term commitment keeps employees from bei ng distracted from the project. Ross et al. (2000) notes that managers demonstrated commitment to the project by assigning their best people full time to the project, clearly developing a business case for system use that has clear objectives, demand status reports based on well-estab lished objectives, communicate goals and scope of the project cl early, and establish a nd articulate a longterm vision. Newman and Sabherwal (1996) focuses on commitment to a project and found that psychological and project determin ants were the most influential in an employeeÂ’s commitment to a project. If empl oyees perceive chronic problems to exist without a solution, commitment will dimini sh. Newman and Sabherwal (1996, p. 27) provides a list of managerial determinants of commitment. The second reason for user resistance is a lack of Input, as th ere are a number of times a userÂ’s opinions are not considered or sought out by management. User involvement has been studied in a number of research publications. For example, Ives and Olson (1984) found that ES implementations are more likely to succeed when user involvement is high. This is different from us er participation; Bark i and Hartwick (1989) distinguishes between user participation and user invol vement, stating that user participation is Â“a set of behaviors or ac tivities performed by us ers in the system development processÂ” while user involvem ent is Â“a subjective psychological state reflecting the importance and personal releva nce of a system to the user (Barki and Hartwick 1989, p. 53). In regards to ES res earch, one study stated th at ES research has
123 not studied user involvement and satisfacti on in depth (Esteves and Pastor 2001). In order to better seek the input of users, co mmunication channels must be available to receive communication from users. Salopek (2001) suggests that management needs to involve users from the beginning as well as re define leadership role s and negotiate with users. The facilitation of these management strategies can be improved through tactics such as opening the communication lines be tween management and users (De Jager 1994). The third reason for user resistance is Control/Power, since some users end up losing control or recognition as an expert. Thus, often times, there is a leveled playing field because someone who is newly hired may have as much expertise as someone employed for many years. Green, Collin s and Hevner (Under Review) found that perceived level of control affect s the level of user satisfaction. This reason for resistance is difficult for management to mitigate, as bringing in a new system often requires the loss of expertise of the old system. Howeve r, through listening to feedback from the Â“expertÂ” users and conveying the necessity of the new system, th e usersÂ’ level of resistance may be reduced. The fourth reason for user resistance lis ted is Self-Efficacy. A lack of selfefficacy may exist because of a lack of confidence in the skillset needed for the new system, such as a lack of computer skills/abilities. Computer self-efficacy has been studied in various publications and has been defined as Â“an indi vidualÂ’s judgment of efficacy across multiple computer application domainsÂ” (Marakas, Yi and Johnson 1998,
124 p. 129). Marakas et al. (1998) also points out that there is a difference between taskspecific and general computer self-efficacy. Ev en for users with general computer selfefficacy, they may lack self-efficacy in regard s to the ES because of task-specific selfefficacy. One study that examined comput er self-efficacy found that it affects the perceived ease of use towards new systems (Agarwal, Sambamurthy and Stair 2000). In fact, Kotter et al. (1979) states that one reason for resistance is that users feel their skilllevel is inadequate. One manage ment strategy to deal with low self-efficacy is to provide training to increase the skills and confidence of the users. Also, providing user support mechanisms can be effective (Bendoly 2000). The training and the user support mechanisms can be complementary. It is an understandable huma n nature that people resist situations if they feel unskilled or that their abilities are lacking. Thus, a lack of training or lack of support may manife st itself through user resistance. A fifth reason for user resistance is the technical problems with the system, such as bugs in the system and features that do not work right. This can be minimized through increased management expertise, such as bringing in consultants and experienced decision-makers who develop an appropriate tim eline that allows for testing the system. Furthermore, through effectively providing he lp and support, technical problems can be dealt with promptly, which should mitig ate the level of user resistance. A sixth reason for user resistance is the complexity of the system, such as the difficulty to access data or a poorly designed user interface that is not intuitive. Initially, the analysis of various ESs and selecti ng a less complex ES would be useful in
125 minimizing future problems. However, once a system is selected, useful training should be able to minimize the impact of the comp lexity on user resistance. Furthermore, through communication channels that receive fe edback from users, appropriate system customizations can be made to minimize the complexity. A seventh reason for user resistance is th e Facilitating Environment, such as an organization that has a bureaucracy whic h is not conducive to change. Large organizations usually are not able to change their envir onments quickly. However, through training management to gain expertis e in the system and organizational change, the impact of user resistance may be minimi zed. In addition, customizing the system to better fit the organiza tion may be useful. An eighth reason for user resistance is Comm unication to users, such as a lack of communication or not conveying to users the be nefits of the system and the Â“whysÂ” of the change. One way to address this issue is through frequent and repetitive communication to users regardi ng the vision, the plan, and pote ntial outcomes of the ES. Planning is a very important part of the vision as it can weave the implementersÂ’ and organizationÂ’s culture together. A communication plan can also be used to facilitate the vision and goals. Oliver a nd Romm (2002) discusses the vi sion of integration that is presented to employees as a reason for ES adoption. This vision may encompass conceptions of teamwork and synergy, and suggest that the ES may bring about harmony for the organization (Oliver and Romm 2002). The decision makers should share the vision and goals, and clearly articulate the means to achieve the goals. The plan should
126 emphasize the benefits to the individuals w ho are to follow the vision and achieve the goals and be flexible enough as to encompass all necessary tasks and permit delays. The communication of the benefits may also lead to users supporting the system to a greater degree. Baronas et al. (1988) writes that Â“More important than the actual changes implementers might make are their skills at communicating them to users, and linking them into usersÂ’ experiencesÂ” (p. 121). Comm unicating with all invol ved parties, setting suitable expectations, and frequent progre ss report meetings can be useful in communication. A ninth reason for user resistance is Trai ning. Training is problematic when users perceive training to be a waste of time, th at trainers are incompetent, the timing of training is inappropriate, or a lack of training. A case study found that although users were briefly trained in using the new system all employees did not feel comfortable, which led to the fear of being laid off, decreased morale, as well as decreased job satisfaction (Mainiero and DeMi chiell 1986). Umble et al. ( 2002) argues that a failure to train users to take advantage of the system Â’s features guarantees that implementation problems will arise. Bingi et al. (1999) also identifies the importance of training, and states that although adequately training employe es for ES use is a major challenge, it is necessary as employees need to know how to do their job and how the data they enter affects the rest of the organization. Appropr iate training is an important management strategy to mitigate these issues.
127 A tenth reason for user resistance is J ob/Job Skills Change, since users often undergo revised job descriptions or must perform different job tasks or develop new skills and new ways of thinking for the job. Kotter et al. (1979) notes that performing new behaviors, working with different pe ople, or assuming different roles makes employees uneasy, contributes to low toleran ce for change. However, it has been found that when users have realistic expectations, ES implementations are far more likely to succeed (Ives and Olson 1984). For example, Ginzberg (1981, p. 475) found that the Â“degree of realism of usersÂ’ pre-implementation expectations was positively correlated with a range of project success measures, both attitudinal and behavioral.Â” Another management strategy to minimize the degree of changed jobs is to customize the system. However, often times this will not be done b ecause part of the reason for the ES is to change inefficient processes. For example, Ross and Vitale (2000) discusses how a CEO talked about during the firmÂ’s first imp lementation, customization requests were considered, but how the steer ing committee rejected customizations during the second implementation. Although system customiza tions often are not performed because of cost, performing the customizations mitigates this reason for user resistance. Additionally, a company may consider a strate gy to provide incentives to users so that they feel compensated for the change they en counter as they adjust to new job tasks and skills. An eleventh reason for user resistance is Workload, as users often need to exert additional effort to perform the same task or ne ed to take work home in order to complete it on time. Employers could address this issue by setting forth appropriate incentives that
128 compensate users for the extra work. Some organizations, whether or not they are implementing an ES, provide bonuses based on performance and/or workload. This could also be used for users that learn a nd adjust to a new system. Furthermore, management may be able to minimize some of the resistance through effective training. In the short-term, training requires additional efforts; however, in the long-term, trained employees should be more productive and ab le to accomplish tasks in less time. The final reason for user resistance is Lack of Fit, due to the problematic changes to processes and new processes not working as planned. Leifer (1988) describes how the technology needs to fit an organization, that a technology may fit some organizations and not others, and that many organizations must change their organizati onal structure to fit the technology. Although this is difficult, it may be necessary in order to remain competitive or to implement strategic change Through customizations, the system can better fit the organizational structure. Furthermore, training managers and the implementation team to be more knowledgeab le in understanding both the processes and the system leads to a better fit between th e system and new orga nizational processes. Discussion of Study 2 Study 2 examined the types of users, focusing on the characteristics of users, the types of resisting users, and the desired manage ment strategies identi fied by these groups. There were eight groups that were examin ed, two of which had a greater degree of resistant behaviors. Due to th e lack of other studies examin ing groups of users, there are not other studies to which results can be compar ed. There clearly are groups of users that
129 emerge from the analysis, which is consistent with both the quotes identified in Study 1 and previous studies that suggest types of us ers exist. Because of the lack of other studies in this area, this study sets the groundwork showing that user groups exist and describes the user groups. In comparing the pilot data and the full data collection, both sets of data collection suggests that there are a wide va riety of user groups. Due to the limited data collected in the pilot study, an in-depth comparison betw een the two data collections has not been performed. However, for the pilot study, five of the groups exhibited a small degree of resistant behaviors, one group exhibited a medium degree of resistant behaviors, and two groups exhibited a large degree of resistant behaviors. Fo r the full data study, two groups exhibited a large degree of resistant behavior s while the other six groups exhibited small or small/medium levels of resistant behavior s. Despite the different users that were examined in the two data collections, user gr oups with resistant beha viors were identified in both data collections. A pract ical implication is that resistan t user groups are likely to exist and management should s eek to understand these groups. From a managerÂ’s perspective, knowing that various groups exist in an ES implementation can lead to strategies that better meet the needs of the various groups. For example, each of the eight groups identifi ed in the results had a different set of reasons for user resistance. To some groups, a lack of input was im portant while to other groups, the uncertainty was important. Ther efore, depending on the employees, some may want to be on a planning committee while others do not; others need some computer
130 training classes while others do not; and some want to have more top-down communication while others do not mind having only minimal communication. In regards to the overall re sults for the respondents, the management strategies that were shown to be the most desirable to users are a clear plan, management expertise, and top-down communication. Although each group has different preferences, these three were shown to be the most important ove rall to users. Mana gers should also be aware of the reasons for resistance that were most often present during implementations. The top five reasons, in order of representativeness to ES implementations, are additional workload, lack of fit, technical problems, changed job, and system complexity. If possible, managers should try to minimize th e potential problems that arise from these areas. For example, the problem of lack of fit could be minimized through spending more time to find the best system suitable to the organization a nd have organizational change management in place to alter any necessary processes prior to the system implementation. There are many other sugge stions provided in the Â“Managing the Reasons for User ResistanceÂ” section of this chapter. Despite the collection of various demogr aphic data, one surprising finding was that there were not any responde nt demographics identified th at differentiated the groups of users. Although this could be due to in sufficient statistical pow er, the implication is that both resisters and non-resisters exis t from all demographic backgrounds. For example, age, gender, education level, and ye ars with employer do not affect the level of resistance or the user group which best fits an employee.
131 Contributions There are several contributions of this di ssertation. First, th e ES implementation is examined from a user resistance perspective; as user resistance is a reason why a technology is not adopted, this research modifies the curren t understanding of the user acceptance literature. As the second chapter points out, there are many studies that have examined user acceptance, with user resist ance sometimes considered the opposite of user acceptance. This study argues that us er resistance is not the opposite of user acceptance and differentiates the two concepts, since user resistance can still occur, even when acceptance appears to have occurred. Base d on the user resistance findings of this study, researchers and practitioners can ha ve a better understandi ng of the difference between user acceptance and user resistance. A second contribution is a bett er understanding of why users resist an ES. In spite of the recent increases in the number of ES publications, there is not a compelling explanation in describing the phenomenon of us er resistance and its underlying causes. This study conceptualizes user resistance, providing a framework that includes an explanation as to why it occurs during ES implementations. A third contribution is providing an unde rstanding of how user resistance manifests itself through behavi ors. Although some studies have suggested ways users may resist a system, this study looks specifically at ES impl ementations and the types of
132 behaviors that are exhibited by users. Through a qualitative analysis, specific resistant behaviors are identified, descri bed, and set into a framework. A fourth contribution is the identification and analysis of management strategies to minimize user resistance. A framework was developed for these management strategies and their effects on user resistan ce are described. As identified previously, there are many critical success factors and management strategies that have been identified that may or may not work depending on the contextual factors. This study has suggested specific management strategies useful in minimizing the level of user resistance in ES implementations. A fifth contribution is the unde rstanding of types of ES users. There has not been any research found that has been conducted in this area. Yet, an understanding of types of users, and in particular, resistant groups, is key to understanding how to mitigate user resistance. Users ranked the reasons for user resistance that were most representative of their ES experience. This research both explores the area of resistant groups and the characteristics of these groups. Furthermor e, it sets the groundwork upon which future theories can be built. A sixth contribution is an understanding of the management strategies most desired by users and perceived to be the most important in minimizing the level of user resistance. Based on the eight general strategies identifi ed in Study 1, users provided feedback regarding the management strategi es desired during the implementation. The
133 assumption is made that the most desired management strategies are also the most useful in minimizing the level of user resistance. Limitations Although the researcher strived to minimi ze potential limitations of the research, there are several limitations. In regards to the qualitative interviews in study 1, generalizability is an inherent limitation. Ther e is no assurance that the individuals that were interviewed are representative of the population. To minimize the impact of this limitation, interviewees were sought out from multiple organizations and in multiple positions within those organizations. In regards to the interpretation of the in terview transcripts, independent coders were used for reliability purposes. Howeve r, only the researcher and two independent coders analyzed the transcripts in depth. Sin ce both coders were trai ned by the researcher and used the coding scheme developed by the researcher, there could be bias in the coding. To minimize this limitation, multiple quo tes from the interviewees covering each construct were selected and shown to othe r researchers who checked the statements. Another potential limitation for both studies is based on the bias of interviewees and questionnaire respondents, which were reflecting on their own ES experiences. One aspect of this bias results from some respondents responding to the questionnaire regarding an experience they had two years previously. Even though respondents may be trying to provide accurate information, they may have a skewed view concerning what
134 actually happened. Furthermore, as Lapoi nte and Rivard (2005) found, resistance may change over time and thus it is possible that some respondents reflected on resistance at an early point while others reflected on resistance at a later point. A nother bias is social desirability, which may have occurred in th e interviews and may have affected the responses of some of the interviewees. For example, intervie wees may not have discussed their own resistance to the system in order to present a certain image about themselves. This impact of this limitati on was minimized through the use of interviewing multiple people within the same organizati on as well as distributing questionnaires to multiple users within the same organization. In regards to the Q-methodology, respondent s were asked to rank the concourse statements by placing them into a Q-sort, wh ich has a fixed distri bution. Although there are advantages to this form of respons e which are discussed in the Q-methodology description, the limitations ar e that respondents may feel th at the concourse statements should be distributed in a di fferent way. For example, so me respondents may feel that several concourse statements ar e highly representative of th eir ES experience while the rest are not representative of their experi ence and have a hard time figuring out how to arrange the various concourse st atements into the fixed dist ribution. This limitation was minimized through having questionnaire responde nts fill out several qualitative questions describing why they had chosen the statements at either end of th e fixed distribution. In regards to the genera lizability of Study 2, a conve nience sample was used. Packets of questionnaires and business reply envelopes were distri buted to members of
135 various user groups. The user s which received the questionnaire from these user group members were not randomly selected among th e general population; rather, they were people known to the user group members who may ha ve filled it out as a favor to the user group member. The result is that certain gr oups may have been underrepresented, such as small businesses that do not have a user group member. Thus, although the respondents represented many di fferent positions within many different organizations, it may not be representative of the overall popu lation. To minimize the impact of this limitation, user groups from various ES vendors were selected in order to represent a wide variety of businesses that implement an ES. Future Research This dissertation provides a foundati on upon which future research on user resistance can be built. One future directi on for this line of res earch is developing a model of user resistance based on the key drivers for user resi stance. This line of future user resistance research would also examin e and identify which reasons are the most important in the determination of behavior s. Although there were a number of reasons and behaviors discussed in this dissertation, it is likely that there are certain reasons that are the key drivers. This line of research would encompass more empirical research. Another direction for user resistance resear ch is understanding the lifecycle of an implementation and how user resistance cha nges throughout the lifecycle. For example, both Markus and Tanis (2000) and Markus et al. (2003) discuss the phases of an implementation. Markus et al. (2003) expands on Markus and Tanis (2000), but both of
136 these discuss problems and successes in the phases of ES implementations. ES success does not just occur from a one-time im plementation, but rather through on-going improvements (Kraemmergaard and Rose 2002). Understanding the lif ecycle of an ES would be useful in developing effective manage ment strategies and ultimately affect the level of user resistance. An equally important future direction is a psychological understanding of the usersÂ’ perspectives. For example, Eagly and Chaiken (1995) discuss Attitude Strength, Attitude Structure, and Resistance to Cha nge. For a user, there may be negative perceptions towards the ES and the change; how ever, the attitude st rength and structure has not been examined. It is possible that if an attitude is not strong enough, even though users may have negative perceptions, resistan t behaviors will not ex ist. On the other hand, users with negative perceptions and a st rong attitude may exhibit a greater degree of resistance.
137 REFERENCES Agarwal, Ritu, V. Sambamurthy and Ral ph M. Stair (2000). "Research Report: The Evolving Relationship between General a nd Specific Computer Self-Efficacy An Empirical Assessment." Information Systems Research 11(4): 418-430. Ainsworth, William (1977). "The Primacy of the User." Infosystems : 46-48. Ajzen, Icek and Martin Fishbein (1980). Understanding Attitudes and Predicting Social Behavior Englewood Cliffs, NJ, Prentice-Hall. Akkermans, H. and K. Van Helden (2002) "Vicious and Virtuous Cycles in ERP Implementation: A Case Study of Inte rrelations between Critical Success Factors." European Journal of Information Systems 11: 35-46. Aladwani, Adel M. (2001). "Change Mana gement Strategies for Successful ERP Implementation." Business Process Management Journal 7(3): 266-275. Alvarez, Rosio and Jacqueline Urla (2002). "Tell me a Good Stor y: Using Narrative Analysis to Examine Information Requirements Interviews during an ERP Implementation." Database for Advances in Information Systems 33(1): 38-52. Anderson, Daun (2003). The Application of Q Met hodology to Understanding the Success of Female Executives in the Biotechnology Industry in Massachusetts: A Contextual Perspective International Leadership Studies Conference. Bagchi, Shantanu, Shivraj Kanungo and Subha sish Dasgupta (2003). "Modeling Use of Enterprise Resource Planning Systems: A Path Analytic Study." European Journal of Information Systems 12(2): 142-158. Barker, Traci and Mark N. Frolick (2003). "E RP Implementation Failure: A Case Study." Information Systems Management 20(4): 43-49. Barki, Henri and Jon Hartwick (1989). "Ret hinking the Concept of User Involvement." MIS Quarterly 12(1): 53-63.
138 Baronas, Ann-Marie K. and Meryl Reis L ouis (1988). "Restoring a Sense of Control during Implementation: How User Involve ment Leads to System Acceptance." MIS Quarterly 12(1): 111-124. Benbasat, Izak, David K. Goldstein and Me lissa Mead (1987). "The Case Research Strategy in Studies of Information Systems." MIS Quarterly 11(3): 369-386. Benbasat, Izak, David K. Goldstein and Me lissa Mead (2002). The Case Research Strategy in Studies of Information Systems. Qualitative Research in Information Systems M. D. Myers and D. Avison. London, Sage Publications: 79-99. Bendoly, Elliot (2000). "A Framework for Investments in Suppor t Building Activities (SBAs): Support Mechanisms and Strategic Scenarios." Decision Support Systems 27: 343-354. Bhattacherjee, Anol (2000). "Beginni ng SAP R/3 Implementation at Geneva Pharmaceuticals." Communications of the Associa tion for Information Systems 4(2): 1-39. Bingi, Prasad, Maneesh K. Sharma and Ja yanth K. Godla (1999) "Critical Issues Affecting an ERP Implementation." Information Systems Management 16(3): 714. Bjorn-Andersen, Niels, Ken Ea son and Daniel Robey (1986). Managing Computer Impact: An International Study of Management and Organizations Norwood, NJ, Ablex Publishing. Blake and Margrethe H. Olson (1984). "Use r Involvement and MIS Success: A Review of Research." Management Science 30(5): 586-603. Bonoma, Thomas V. (1985). "Case Research in Marketing: Opportuni ties, Problems, and a Process." Journal of Marketing Research 22: 199-208. Bovey, Wayne H. and Andrew Hede (2001). "R esistance to Organizational Change: The Role of Defense Mechanisms." Journal of Managerial Psychology 16(7): 534548.
139 Bovey, Wayne H. and Andy Hede (2001). "Res istance to Organizational Change: The Role of Cognitive and Affective Processes." Leadership & Organization Development Journal 22(8): 372-382. Brancheau, James C., Brian D. Janz and Ja mes C. Wetherbe (1996). "Key Issues in Information Systems Management: 1994-95 SIM Delphi Results." MIS Quarterly 20(2): 225-242. Brouwer, Marten (1992-1993). "Validity: Q vs. R." Operant Subjectivity 16(1/2): 1-17. Brown, Carol V. and Sharon L. Magill (1998) "Reconceptualizing the Context-Design Issue for the Information Systems Function." Organization Science 9(2): 176-194. Brown, Maureen (2004). Illu minating Patterns of Percep tion: An Overview of Q Methodology, Carnegie Mellon University: 1-21. Brown, S.A., A.P. Massey, M.M. Montoya-Weiss and J.R. Burkman (2002). "Do I Really Have To? User Acceptance of Mandated Technology." European Journal of Information Systems 11: 283-295. Brown, Steven R. (1971). "The Forced -Free Distinction in Q Technique." Journal of Educational Measurement 8(4): 283-287. Brown, Steven R. (1980). Political Subjectivity Applic ations of Q Methodology in Political Science Westford, MA, Yale University Press. Brown, Steven R. (1986). Q Technique a nd Method: Principles and Procedures. New Tools for Social Scientists W. D. Berry and M. S. Lewis-Beck. Beverly Hills, CA, Sage: 57-76. Brown, Steven R. (1993). "A Primer on Q Methodology." Operant Subjectivity 16(3/4): 91-138. Brown, Steven R. (1996). "Q Method ology and Qualitative Research." Qualitative Health Research 6(4): 561-567. Brown, Steven R. (Under Review). "Q Technique and Questionnaires." 1-9.
140 Callahan, Terry (2002). Credit Research Foundation Examines ERP Technology. 2004. Chalmers, Raymond E. (1999). "Small Ma nufacturers Seek Best ERP Fit." Manufacturing Engineering 123(4): 42-46. Chen, Hung-Chen and Arbee L.P. Chen (2005). "A Music Recommendation System Based on Music and User Grouping." Journal of Intelligen t Information Systems 24(2/3): 113-132. Cissna, Tami (1998). "ERP Software Implem entation brings Pains with its Gains." Electric Light & Power 76: 43. Conner, Daryl R. (1993). Managing at the Speed of Change: How Resilient Managers Succeed and Prosper Where Others Fail Toronto, Random House of Canada. Cooke, Dudley P. and William J. Peterson ( 1998). SAP Implementation: Strategies and Results, The Conference Board: 1-22. Cooper, Randolph B. (2000). "Information Technology Development Creativity: A Case Study of Attempted Radical Change." MIS Quarterly 24(2): 245-276. Cule, Paul E. and Daniel Robey (2004). "A Dual-Motor, Constructiv e Process Model of Organizational Transition." Organization Studies 25(2): 229-260. Davidson, Julia O'Connell (1994). The Sources a nd Limits of Resistance in a Privatized Utility. Resistance and Power in Organizations J. M. Jermier, D. Knights and W. R. Nord. London, Routledge: 69-101. Davis, Fred D. (1989). "Per ceived Usefulness, Perceived Ease of Use, and User Acceptance of Information Technology." MIS Quarterly 13(3): 319-340. Dawson, Patrick and Ian McLaughlin (1986). "Computer Technology and the Redefinition of Supervision: A Study of the Effects of Computerization on Railway Freight Supervisors." Journal of Management Studies 23(1): 116-132. De Brabander, B. and G. Thiers (1984). "S uccessful Information System Development in Relation to Situational Factors which Affect Effective Communication between MIS-Users and EDP-Specialists." Management Science 30(2): 137-155.
141 De Jager, P. (1994). "Communi cating in Times of Change." Journal of System Management 45(6). Dennis, Alan R. and Bryan A. Reinicke ( 2004). "Beta versus VHS and the Acceptance of Electronic Brainstorming Technology." MIS Quarterly 28(1): 1-20. Dennis, Karen E. (1988). Q-Methodology: Ne w Perspectives on Estimating Reliability and Validity. Measurement of Nursing Outcomes O. L. Strickland and C. F. Waltz. New York, Springer. 2: 409-419. Dong, Linying (2001). "Modeling Top Manageme nt Influence on ES Implementation." Business Process Management Journal 7(3): 243-250. Doppler, Klaus (2004). Managing Change Su ccessfully: Core Questions, Issues, and Strategies. Dynamics of Organizational Change and Learning J. J. Boonstra. West Sussex, England, John Wiley & Sons: 115-132. Dos Santos, B.L. (1989). "Information Syst ems: Similarities and Difference across Organizations." OMEGA 17(1): 9-20. Dos Santos, Brian L. and Stephen R. Hawk (1988). "Differences in Analyst's Attitudes towards Information Systems Development: Evidence and Implications." Information & Management 14: 31-41. Dryden, Patrick (July 27, 1998). ER P Failures Exact High Price. Computerworld Accessed: November 16, 2005. Dunphy, Dexter and Doug Stace (1993). "The Strategic Management of Corporate Change." Human Relations 46(8): 905-920. Eagly, Alice H. and Shelly Chaiken (1995). A ttitude Strength, Attitude Structure, and Resistance to Change. Attitude Strength: Antecedents and Consequences R. E. Petty and J. A. Krosnick. Mahwah, NJ Lawrence Erlbaum Associates: 413-432. Edosomwan, Johnson A. (1996). Organizational Transformation and Process Reengineering Delray Beach, Florid a, St. Lucie Press.
142 Eisenhardt, Kathleen M. ( 1989). "Building Theories from Case Study Research." Academy of Management Review 14(4): 532-550. Esteves, Jose and Joan Pastor (2001). "Ent erprise Resource Planning Systems Research: An Annotated Bibliography." Communications of the Asso ciation for Information Systems 7(8): 1-51. Falbe, Cecilia M. and Gary Yukl (1992). "Consequences for Managers of Using Single Influence Tactics and Combinations of Tactics." Academy of Management Journal 35(3): 638-652. Ferman, John E. (1999). "Strategie s for Successful ERP Connections." Manufacturing Engineering 123(4): 48-60. Fiorelli, Joseph S. and Howard Margolis (1993). "Managing a nd Understanding Large Systems Change: Guidelines for Executives and Change Agents." Organization Development Journal 11(3): 1-13. Floyd, Christiane (1993). "STEPS--A Methodical Approach to PD." Communications of the ACM 36(4): 83. Folger, Robert and Daniel P. Skarlicki ( 1999). "Unfairness and Resistance to Change: Hardship as Mistreatment." Journal of Organizational Change Management 12(1): 35-50. Ford, Jeffrey D. and Laurie W. Ford ( 1994). "Logics of Identity, Contradiction, and Attraction in Change." Academy of Management Review 19(4): 756-785. Ford, Jeffrey D., Laurie W. Ford and Randa ll T. McNamara (2002). "Resistance and the Background Conversations of Change." Journal of Organizational Change Management 15(2): 105-121. Franz, Charles R. and Daniel Robey (1984) "An Investigation of User-led System Design: Rational and Political Perspectives." Communications of the ACM 27(12): 1202-1209. Franz, Charles R., Daniel Robey and Robert R. Koeblitz (1986). "User Response to an Online Information System: A Field Experiment." MIS Quarterly 10(1): 29-42.
143 Gattiker, Thomas F. and Dale L. Goodhue (2005). "What Happens After ERP Implementation: Understanding the Impact of Interdependence and Differentiation on Plant-Level Outcomes." MIS Quarterly 29(3): 559-585. Ginzberg, Michael J. (1981). "Early Di agnosis of MIS Implementation Failure: Promising Results and Unanswered Questions." Management Science 27(4): 459478. Glaser, Barney G. and Anselm L. Strauss (1967). The Discovery of Grounded Theory: Strategies for Qualitative Research Chicago, Aldine. Goltz, Sonia M. and Amy Hietapelto ( 2002). "Using the Operant and Strategic Contingencies Models of Power to Understand Resistance to Change." Journal of Organizational Behavior Management 22(3): 3-22. Gosain, Sanjay (2004). "Enterprise Informa tion Systems as Objects and Carriers of Institutional Forces: The New Iron Cage?" Journal of the Association for Information Systems 5(4): 151-182. Gottschalk, Petter (2001). "Key Issues in IS Management in Norway: An Empirical Study Based on Q Methodology." Information Resources Management Journal 14(2): 37-45. Gravenhorst, Kilian Bennebroek and Roeland in 't Veld (2004). Power and Collaboration: Methodologies for Working Together in Change. Dynamics of Organizational Change and Learning J. J. Boonstra. West Sussex, England, John Wiley & Sons: 317-341. Green, Gina C., Rosann Webb Collins and Alan R. Hevner (Under Review). "Perceived Control and the Diffusion of So ftware Process Innovations." submitted to Journal of Management Information Systems. Greiner, Larry E. (1972). "Evolution and Revolution as Organizations Grow." Harvard Business Review 50(4): 37-46. Gupta, Atul (2000). "Enterpr ise Resource Planning: The Emerging Organizational Value Systems." Industrial Management & Data Systems 100(3): 114-118.
144 Hammer, Michael (1990). "Reengineering Work: Don't Automate, Obliterate." Harvard Business Review 68(4): 104-112. Hammer, Michael and James Champy (1993). Reengineering the Corporation: A Manifesto for Business Revolution New York, HarperCollins. Hartwick, Jon and Henri Barki (1994). "Explai ning the Role of User Participation in Information System Use." Management Science 40(4): 440-465. Hersey, Paul and Kenneth H. Blanchard (1988). Management of Organizational Behavior: Utilizing Human Resources. Englewood Cliffs, NJ, Prentice Hall. Hill, Kimberly (March 26, 2003). "System Designs That Start at the End (User)." CRM Daily Accessed: October 13, 2004. Hines, Matt (2002). PeopleSo ft 8.8 to be More Usable, Easier to Integrate. 2004. Hirschheim, R. and M. Newman (1988). "Information Systems and User Resistance: Theory and Practice." The Computer Journal 31(5): 398-408. Hitt, Lorin M., D.J. Wu and Xiaoge Zhou (2002). "Investment in Enterprise Resource Planning: Business Impact a nd Productivity Measures." Journal of Management Information Systems 19(1): 71-98. Hu, Paul Jen-Hwa, Theodore H.K. Clar k and Will W. Ma (2003). "Examining Technology Acceptance by School T eachers: A Longitudinal Study." Information & Management 41: 227-241. Hunton, James E. and Jesse D. Beeler (1997). "Effects of User Participation in Systems Development: A Longitudinal Field Experiment." MIS Quarterly 21(4): 359-388. Hussain, Donna and K.M. Hussain (1984). Information Res ource Management. Homewood, IL, Richard D. Irwin. Ives, Blake and Margrethe H. Olson (1984) "User Involvement and MIS Success: A Review of Research." Management Science 30(5): 586-603.
145 Jasperson, Jon (Sean), Traci A. Carte a nd Carol S. Saunders (2002). "Power and Information Technology Research: A Metatriangulation Review." MIS Quarterly 26(4): 397-459. Jermias, Johnny (2001). "Cognitive Dissonance a nd Resistance to Change: The Influence of Commitment Confirmation and Fee dback on Judgment Usefulness of Accounting Systems." Accounting, Organizations, and Society 26: 141-160. Jiang, James J., Waleed A. Muhanna and Gary Klein (2000). "User Resistance and Strategies for Promoting Acceptance across System Types." Information & Management 37: 25-36. Joshi, Kailash and Thomas W. Lauer ( 1998). "Impact of Information Technology on Users' Work Environment: A Case of Computer Aided Desi gn (CAD) System Implementation." Information & Management 34(6): 349-360. Jurison, Jaak (2000). "Perceived Value a nd Technology Adoption across Four End User Groups." Journal of End User Computing 12(4): 21-28. Kaplan, Bonnie and Dennis Duchon (1988). "Combining Qualitative and Quantitative Methods in Information Syst ems Research: A Case Study." MIS Quarterly 12(4): 571-586. Keen, Peter G.W. (1981). "Information Systems and Organizational Change." Communications of the ACM 24(1): 24-33. Kegan, Robert and Lisa Laskow Lahey (2001) "The Real Reason People Won't Change." Harvard Business Review : 85-92. Kendall, Julie E. and Kenneth E. Kendall ( 1993). "Metaphors and Methodologies: Living Beyond the Systems Machine." MIS Quarterly 17(2): 149-171. Kendall, Kenneth E. (1997). "The Signifi cance of Information Systems Research on Emerging Technologies: Seven Informa tion Technologies that Promise to Improve Managerial Effectiveness." Decision Sciences 28(4): 775-792. Kirk, Jerome and Marc L. Miller (1986). Reliability and Validity in Qualitative Research. Beverly Hills, CA, Sage.
146 Klatzky, S.R. (1970). "Automation, Size, and the Locus of Decision Making: The Cascade Effect." The Journal of Business 43(2): 141-151. Klein, Heinz K. and Michael D. Myers (1999) "A Set of Principles for Conducting an Evaluating Interpretive Field St udies in Information Systems." MIS Quarterly 23(1): 67-94. Kling, Rob (1980). "Social Analyses of Compu ting: Theoretical Perspectives in Recent Empirical Research." Computing Surveys 12(1): 61-110. Kling, Rob and Suzanne Iacono (1984). "T he Control of Information Systems Developments after Implementation." Communications of the ACM 27(12): 12181226. Knights, David and Theo Vurdubakis (1994). Fo ucault, Power, Resistance, and All That. Resistance and Power in Organizations. J. M. Jermier, D. Knights and W. R. Nord. London, Routledge: 167-198. Koh, Christine, Soon Ang and Detmar W. Straub (2004). "IT Outsourcing Success: A Psychological Contract Perspective." Information Systems Research 15(4): 356373. Kotter, John P. and Leonard A. Schlesinge r (1979). "Choosing Strategies for Change." Harvard Business Review: 106-114. Kraemmergaard, Pernille and Jeremy Rose (2002). "Managerial Competences for ERP Journeys." Information System Frontiers 4(2): 199-211. Krasner, Herb (2000). "Ensuring E-Busine ss Success by Learning from ERP Failures." IT Professional 2(1): 22-27. Landis, J. Richard and Gary G. Koch (1977). "The Measurement of Observer Agreement for Categorical Data." Biometrics 33: 159-174. LaNuez, Danny and John M. Jermier (1994). Sabotage by Managers and Technocrats: Neglected Patterns of Resistance at Work. Resistance and Power in Organizations J. M. Jermier, D. Knights and W. R. Nord. London, Routledge: 219-251.
147 Lapointe, Liette and Suzanne Rivard (2005). "A Multileve l Model of Resistance to Information Technology Implementation." MIS Quarterly 29(3): 461-491. Larsen, Kai R.T. (2003). "A Taxonomy of Ant ecedents of Information Systems Success: Variable Analysis Studies." Journal of Management Information Systems 20(2): 169-246. Lawler, Edward E. (2000). Rewarding Excellence. San Francisco, Jossey-Basse Inc. Legare, Thomas L. (1995). "Minimizing Re sistance to Technological Change: A Power and Politics Approach." Information Systems Management 12: 59-61. Leifer, Richard (1988). "Mat ching Computer-Based Information Systems with Organizational Structures." MIS Quarterly 12(1): 63-73. Lewis, William, Ritu Agarwal and V. Sa mbamurthy (2003). "Sources of Influence on Beliefs about Information Technology Us e: An Empirical Study of Knowledge Workers." MIS Quarterly 27(4): 657-678. Mainiero, Lisa A. and Robert L. DeMich iell (1986). "Minimizi ng Employee Resistance to Technological Change." Personnel : 32-37. Marakas, George M., Mun Y. Yi and Rich ard D. Johnson (1998). "The Multilevel and Multifaceted Character of Computer Self -Efficacy: Toward Clarification of the Construct and an Integrative Framework for Research." Information Systems Research 9(2): 126-163. Marakas, GM and S Hornik (1996). "Pa ssive Resistance Misuse: Overt Support and Covert Recalcitrance in IS Implementation." European Journal of Information Systems 5: 208-219. Markus, M. Lynne (2004). "Technochange Management: Using IT to Drive Organizational Change." Journal of Information Technology 19: 3-19. Markus, M. Lynne, Sheryl Axline, David Pe trie and Cornelis Tanis (2003). Learning from Adopters' Experiences with ER P: Problems Encountered and Success Achieved. Second-Wave Enterprise Resource Planning Systems. G. Shanks, P. B. Seddon and L. P. Willcocks. Cambridge, Cambridge University Press.
148 Markus, M. Lynne and Daniel Robe y (1988). "Information Technology and Organizational Change: Causal Stru cture in Theory and Research." Management Science 34(5): 583-598. Markus, M. Lynne and Cornelis Tanis (2000). The Enterprise System Experience From Adoption to Success. Framing the Domains of IT Management R. W. Zmud. Cincinnati, Ohio, Pinnaflex E ducational Resources: 173-208. Markus, M.L. (1983). "Power, Politics, and MIS Implementation." Communications of the ACM 26(6): 430-444. Martinsons, Maris G. and Patrick K.C. C hong (1999). "The Influence of Human Factors and Specialist Involvement on Information Systems Success." Human Relations 52(1): 123-152. Mathieson, K. (1991). "Predicting User Intentions: Comparing the Technology Acceptance Model with the Theory of Planned Behavior." Information Systems Research 2(3): 173-191. Maurer, Rick (2002). Plan for the Human Part of ERP. Workforce Online McKeown, Bruce and Dan Thomas (1988). Q Methodology Newbury Park, CA, Sage. Miles, Matthew B. and A. Michael Huberman (1994). Qualitative Data Analysis: An Expanded Sourcebook Thousand Oaks, CA, Sage Publications. Mullins, Laurie J. (1999). Management and Organisational Behaviour. London, Financial Times Professional Limited. Myers, Michael D. and David Avison (2002). An Introduction to Qualitative Research in Information Systems. Qualitative Research in Information Systems M. D. Myers and D. Avison. London, Sage Publications: 3-12. Nah, Fiona Fui-Hoon and Janet Lee-Shang La u (2001). "Critical Factors for Successful Implementation of Enterprise Systems." Business Process Management Journal 7(3): 285-296.
149 Newman, Michael and Rajiv Sabherwal ( 1996). "Determinants of Commitment to Information Systems Development: A Longitudinal Investigation." MIS Quarterly 20(1): 23-54. Nunnally, Jum C. (1978). Psychometric Theory New York, McGraw-Hill Book Company. O'Leary, Daniel Edmund (2000). Enterprise Resource Planning Systems: Systems, Life Cycle, Electronic Commerce, and Risk Cambridge, UK, Cambridge University Press. Oliver, Dave and Celia Ro mm (2002). "Justifying Enterp rise Resource Planning Adoption." Journal of Information Technology 17: 199-213. Orlikowski, Wanda J. and Stephen R. Barley (2001). "Technology and Institutions: What Can Research on Information Technology and Research on Organizations Learn from Each Other?" MIS Quarterly 25(2): 145-165. Pardo del Val, Manuela and Clara Martinez Fuentes (2003). "Resistance to Change: A Literature Review and Empirical Study." Management Decision 41(2): 148-155. Piderit, Sandy Kristin (2000) "Rethinking Resistance and Recognizing Ambivalence: A Multidimensional View of Attitudes Toward an Organizational Change." Academy of Management Review 25(4): 783-794. Pimchangthong, Daranee, Michel Plaisent and Prosper Bernard (2003). "Key Issues in Information Systems Management: A Comparative Study of Academics and Practitioners in Thailand." Journal of Global Information Technology Management 6(4): 27-44. Prasad, Pushkala (1993). "Symbolic Proce sses in the Implementation of Technological Change: A Symbolic Interactionist Study of Work Computerization." Academy of Management Journal 36(6): 1400-1429. Ranchhod, Ashok and Fan Zhou (2001). "Comparing Respondents of E-mail and Mail Surveys: Understanding the Implications of Technology." Marketing Intelligence & Planning 19(4): 254-262.
150 Rao, Siriginidi Subba (2000). "Enterpris e Resource Planning: Business Needs and Technologies." Industrial Management & Data Systems 100(2): 81-88. Robey, Daniel (1979). "User Attitudes and Management Information System Use." Academy of Management Journal 22(3): 527-538. Robey, Daniel (1981). "Computer Informati on Systems and Organization Structure." Communications of the ACM 24(10): 679-687. Robey, Daniel and Marie-Claude Boudreau (1999). "Accounting for the Contradictory Organizational Consequences of Informa tion Technology: Theoretical Directions and Methodological Implications." Information Systems Research 10(2): 167-185. Robey, Daniel, Jeanne W. Ross and Mari e-Claude Boudreau (2002). "Learning to Implement Enterprise Systems: An E xploratory Study of the Dialectics of Change." Journal of Management Information Systems 19(1): 17-46. Rogers, Everett M. (1976). "New Product Adoption and Diffusion." Journal of Consumer Research 2(4): 290-301. Ross, Jeanne W. and Michael R. Vitale ( 2000). "The ERP Revolution: Surviving Vs. Thriving." Information System Frontiers 2(2): 233-241. Salopek, Jennifer J. (2001) "Give Change a Chance." APICS The Performance Advantage 11(7). Schein, Edgar H. (1992). The Role of the CE O in the Management of Change: The Case of Information Technology. Transforming Organizations. T. A. Kochan and M. Useem. New York, NY, Oxford University Press: 80-95. Shang, Shari S. C. and Tim C. C. Su (2004). Managing User Resistan ce in Enterprise Systems Implementation. Americas Conference on Information Systems, New York. Sheth, Jagdish N. (1981). Psychology of I nnovation Resistance: The Less Developed Concept (LDC) in Diffusion Research. Research in Marketing. J. N. Sheth. Greenwich, Connecticut, Jai Press. 4: 273-282.
151 Shields, Donald C. and John F. Cragan (1981). A Communicati on Based Political Campaign: A Theoretical and Methodological Perspective. Applied Communication Research: A Dramatistic Approach. D. C. Shields and J. F. Cragan. Prospect Heights, IL, Waveland Press: 177-196. Sia, Siew Kien, May Tang, Christina Soh and Wai Fong Boh (2002). "Enterprise Resource Planning (ERP) Systems as a Technology of Power: Empowerment or Panoptic Control?" The DATA BASE for Advances in Information Systems 33(1): 23-37. Silver, Mark S., M. Lynne Markus and Cynt hia Mathis Beath (1995). "The Information Technology Interaction Model: A Founda tion for the MBA Core Course." MIS Quarterly 19(3): 361-390. Soh, Christina, Siew Kien Sia, Wai Fong Boh and May Tang (2003). "Misalignments in ERP Implementation: A Dialectic Perspective." International Journal of HumanComputer Interaction 16(1): 81-100. Somers, Toni M., Klara Nelson and Jahangi t Karimi (2003). "Confirmatory Factor Analysis of the End-User Computing Satis faction Instrument: Replication within an ERP Domain." Decision Sciences 34(3): 595-621. Steelman, Toddi A. and Lynn A. Maguire (1999). "Understanding Participant Perspectives: Q-Methodology in National Forest Management." Journal of Policy Analysis and Management 18(3): 361-388. Stein, Tom (1999). "Making ERP Add Up." InformationWeek : 59-68. Stephenson, W. (1935). "Technique of Factor Analysis." Nature 136: 297. Stephenson, William (1953). The Study of Behavior: Q-T echnique and Its Methodology. Chicago, IL, University of Chicago Press. Stratman, Jeff K. and Aleda V. Roth ( 2002). "Enterprise Resource Planning (ERP) Competence Constructs: Two-Stage Multi-Item Scale Development." Decision Sciences 33(4): 601-628. Strauss, Anselm L. (1990). Basics of Qualitative Research Newbury Park, CA, Sage Publications.
152 Surmacz, Jon (2002). "Mix-and-Match ERP." CIO http://www2.cio.com/metrics/2002/m etric381.html(Printed 10/22/04). Taylor, S. and P.A. Todd (1995). "Understand ing Information Technology Usage: A Test of Competing Models." Information Systems Research 6: 144-176. Thomas, Dominic M. and Richard T. Wats on (2002). "Q-Sorting and MIS Research: A Primer." Communications of the Associat ion for Information Systems 8: 141-156. Todd, Peter and Izak Benbasat (1999). "Evaluat ing the Impact of DSS, Cognitive Effort, and Incentives on St rategy Selection." Information Systems Research 10(4): 356374. Tractinsky, Noam and Sirkka L. Jarve npaa (1995). "Information Systems Design Decisions in a Global vers us Domestic Context." MIS Quarterly 19(4): 507-534. Trader-Leigh, Karyn E. (2002). "Case St udy: Identifying Resistance in Managing Change." Journal of Organizational Change Management 15(2): 138-155. Turley, Richard T. and James M. Bieman (1995). "Competencies of Exceptional and Nonexceptional Software Engineers." Journal of Systems Software 28: 19-38. Umble, Elisabeth J. and M. Michael Umble (2002). "Avoiding ERP Implementation Failure." Industrial Management 44(1): 25-33. Valenta, Annette L. and Ulrike Wi gger (1997). "Q-Methodol ogy: Definition and Application in Health Care Informatics." Journal of the American Medical Informatics Association 4(6): 501-510. Van de Ven, Andrew H. and Marshall Sco tt Poole (1995). "Explaining Development and Change in Organizations." Academy of Management Review 20(3): 510-540. Venkatesh, Viswanath and Fred D. Davis ( 2000). "A Theoretical Extension of the Technology Acceptance Model: Four Longitudinal Field Studies." Management Science 46(2): 186-204.
153 Venkatesh, Viswanath, Michael G. Morris, Go rdon B. Davis and Fred D. Davis (2003). "User Acceptance of Information Technology: Toward a Unified View." MIS Quarterly 27(3): 425-478. Wah, Louisa (2000). "Give ERP a Chance." Management Review 89(3): 20-24. Walsham, Geoff (2002). Interpretive Case Studi es in IS Research: Nature and Method. Qualitative Research in Information Systems. M. D. Myers and D. Avison. London, Sage Publications: 101-113. White, Kathy Brittain (1984). "MIS Project Te ams: An Investigation of Cognitive Style Implications." MIS Quarterly 8(2): 95-101. Willcocks, Leslie P. and Richard Sykes (2000). "The Role of the CIO and IT Function in ERP." Communications of the ACM 43(4): 32-38. Wingreen, Stephen C., J. Ellis Blanton, Sa ndra K. Newton and Madeline Domino (2005). Assessing the IT Training and Development Climate: An Application of the QMethodology SIGMIS-CPR, Atlanta, GA. Wu, Ing-Long (2003). "Understanding Senior Ma nagement's Behavior in Promoting the Strategic Role of IT in Process Reengi neering: Use of the Theory of Reasoned Action." Information & Management 41: 1-11. Yin, Robert K. (2003). Case Study Research: Design and Methods Newbury Park, CA, Sage. Yoon, Youngohc, Tor Guimaraes and Quinton O'Neal (1995). "Exploring the Factors Associated with Expert Systems." MIS Quarterly : 83-106. Zhang, Xiangmin and Hui Han (2005). "An Em pirical Testing of User Stereotypes of Information Retrieval Systems." Information Processing and Management 41: 651-664. Zuboff, Shoshana (1988). In the Age of the Smart Machine. New York, NY, Basic Books.
154 APPENDIX A: SAMPLE QUOTES FOR REASONS FOR RESISTANCE Names and other identifying information have been changed from these quotes
155 Table A-1: Sample quotes for reasons for resistance Construct Sample Quotes U17People did perceive from my e xperience it as a threat to their job. Oh, theyÂ’re not going to need me quite as much or Oh, IÂ’m not going to be able to use this system and theyÂ’re going to fire me if I canÂ’t use it kind of thing U18it was just the c oncept of the unknown. How is this really going to work, how is this really going to function, is really going to do what theyÂ’re telling us itÂ’s going to do. U19Some new features in the sy stem were unclear on how theyÂ’re going to workÂ… I think a lot of that fear and concern had to do with that they werenÂ’t sure how they were going to get their jobs done Uncertainty U12I think people were afraid of how it was going to change their jobs because they were very scared F1(3)Most people really hated it and didnÂ’t understand why they werenÂ’t asked about it or anythi ng. It was just Â“hereÂ’s the new system-enjoy.Â”Â… The general idea is that the people who actually do the work have had the least amount of input on how the system turns out. U3You may not want my input, but itÂ’s important to meÂ… you donÂ’t ask my opinion then fine I donÂ’t want to be involved with it, but July 1st is approaching and your job is changing and you have to. Those were the folks that you had to smooth over. Input U14We tried very hard to tell people. People said your system itÂ’s not my system, I didnÂ’t buy it, I didnÂ’t implement it anymore than I could, I did the vanilla like everybody told us to do. U3Using the state system, having it for 20 years they were experts in their field all of the sudden, you have leveled the playing field. And the new person coming off the street knows just as much about the system as you do, so you are no longer an expert. F1(6)Some people liked the old system a lot better, who were the Â“expertsÂ” Â– they would resist mo re, because with the changes, everybody starts at square one Â– you donÂ’t have that advantage or comfort zoneÂ… For example, custom er service people wanted to be the Â“go-toÂ” people. Control/Power U12in their bossÂ’s eyes they were the experts, they knew they could hand them anything and their bo sses just thought they were wonderful. Now theyÂ’re faced with a system they donÂ’t know that well, they donÂ’t want to look inco mpetent and sometimes you have to look incompetent until you learn a system. And people donÂ’t want to go through that, they donÂ’t want to disappoint their supervisors, or look incompetent in front of their supervisors.
156 U8thereÂ’s a large population at the university th at theyÂ’re not computer savvy. Whatever computer us e they have is here at work. They do very little at homeÂ… IÂ’m going to retire in a couple of years, I donÂ’t need to learn this. I just don Â’t like computers is what one lady told us, she wasnÂ’t going to use the system. She was going to have someone else in her department learn it, computers scared her. U1I'm not the account person so it probably was more difficult for me. Say, a young person coming in at this point whoÂ’s account savvy wouldn't have a problem. U6I feel very uncomfortable for th e fact that I feel like I cannot balance my accounts like I used to balance them before. Self-Efficacy F1(2)because they were afraid of entering the wrong code, and so they didnÂ’t want to take part in any of the user acceptance testing. U2they werenÂ’t able to acce ss their budgets for six months. U7There were lots of glitches at the beginning. Very often we found it just wasnÂ’t working. It just wasn Â’t doing what it was supposed to be doing. U14I can tell you what happened and I canÂ’t tell you why it happened and I canÂ’t necessarily fix it. Which is the biggest frustration that we have. We s ee its wrong, the system let you do it wrong, but now it wonÂ’t let you fix it. You know, so itÂ’s very frustrating. People are frustrated with it. U22if you get to a certain point, yo u canÂ’t print it, but then if you do one of two things and then you go to print it and wonÂ’ t print and itÂ’s been a nightmare. I hate it. I absolutely hate it. Technical Problems U11-The system had a lot of bugs in the beginning and it had a lot of bugs at the training, it didnÂ’t help us sell this thing, even at the training it would crash, so. U2The system is so complicatedÂ… It doesnÂ’t make sense for most of us. Let me know when you find so mebody who can read one of their reports and access it. The hardest part is to access. U9In terms of how we derive the information, how we get the numbers that we need, itÂ’s much more complexÂ… ItÂ’s much more difficult, much more frustrating a nd I have many more people driving me crazy with questions U2I have a secretary in the na val ROTC program, who is going to use the system for the second time in two years, because the system is so complicatedÂ… people in the trench es can write a better interface and they know what people want to see and how it reads cleanly. Complexity U20some people were excited by it because it was new, but by the same token there were some people w ho were afraid of it because itÂ’s new, because itÂ’s definitely a more complex system that we had previously.
157 U3You want to talk about change I think itÂ’s just a paradigm shift, especially in this environment the university is slow to change. The bureaucracy just creates that. U8for some reason itÂ’s a struggle to make that change here and I think it goes back to there was no one ever in place empowered to do it before, so any kind of change to try and give that person that power is met with resistance. And out in the private world, there is no, IÂ’m not going to do it that way, itÂ’s you will do it this way or you will work some place else. And thatÂ’s not the culture here at the university. Facilitating Environment U12thereÂ’s a lot of, what I would ju st call, self service attitudes, Â… ItÂ’s not in my job description, IÂ’m not doing it. And the system introduces and lot of crossover wher e you need to be kind of able to do more, if youÂ’re going to be kind of like me, doing many things, youÂ’ve got to be able to interface wi th a lot of different departments and lot of different skills. U15I donÂ’t think it was communicate d, maybe at a very high level it was communicated clearly, but not down to the trenches. F1(4)I know at our organization with the system, they didnÂ’t communicate. There really wasnÂ’t much communication as far as the goals or benefits or anything like th at. Now, in hindsight, you can see that Â… there are benefits afte r the fact, but that was not communicated, so there was pain wh en there was no system at all Â– there was no discussion of Â“We kno w it sucks now, but itÂ’s going to be great in 6 monthsÂ” Â– thereÂ’s not even that type of communication. U6No, I didnÂ’t have no knowledge whatsoever of the system. A lot of information went on with e-ma il. Communication with e-mail, but as far as the system itself, I le arned about it when I attended the training. Communication U3communication is very bad here at the university and it gets filtered down person by person. U10It seems like they did not o ffer enough training after the system was put in. U2And those [training] classes dr ive me nuts. Because they work with the lowest common denominator the slowest person in the class drives the class. U7I witnessed some people getti ng just exasperated because the people who were training them were not that knowledgeable in the subject matter and uh, you know itÂ’s ha rd to say whose fault that isÂ… IÂ’ve got to tell you some of those tr ainings were terrible and when you walked out of there you didnÂ’t know much more than when you walked in. Training U9there has not been enough trai ning, there continues to be not enough training, I mean the traini ng just gets less and less.
158 U2-Â… from the high muckety-mucks, that's wonderful. But from the little people's level, thatÂ’s a fucking pain in the ass. Because instead of just processing the piece of paper we now become purchasing agents, payroll clerks, and HR reps. It's a nice concept Â… itÂ’s great if you're at the top. U1some people have been here 20 or 30 years Â– itÂ’s just real hard for them to change. It was hard to just start using a computer. I mean we used to use reams and reams of pape r with stuff on it. You get a hard copy Â– somebody else did the programming for you now you do it yourselfÂ… Everybody here had to unlearn and look at it from a different perspective and that's not always easily. U19now the skill sets required on the part of organizationÂ’s staff have changed. You know, itÂ’s less about going to the file draw and rifling through or pulling out re ports and building some sort of spreadsheet. Now itÂ’s a query, but y ou need to understand what tables the information resides on, what the field names might be. Job/Job Skills Change U12a lot of the issue now you real ly do have to know a little bit of accounting to be able to operate e fficiently in the system and people donÂ’t know that and accounting debits and credits are a mystery to most people. U9Some of those people are really, really struggling. So itÂ’s definitely made our jobs more time consuming. More frustrating alsoÂ… IÂ’ve had to stay late plen ty and do things at homeÂ… what I used to be able to do in a short amount of time takes much, much longerÂ… you still have the same amount of work to do, itÂ’s just taking more time to do it. U22It takes much; much longer to do the same the thing, to get a requisition in here takes about 90 stepsÂ… The previous system was very easyÂ… It was very, very differe nt, but once I got into it, it was very easy to move around in. U8I have a 40-hour a week job for the department that pays me and now you want me to do this system work as well. Workload U6-[[So it sounds like everything take s longer.]] Of course, definitely, definitely.
159 U8In the past the invoices came to the individual departments. We checked them, made sure they were correct; if they were wrong we got the vendor to send us a new invoice, whatever. Well during the solution design labs the central unit said no all invoices were notifying the vendors that all invo ices were coming to accounts payable. You will not see them any more, if you put in the correct amount on the purchase order there will be no problems. We said you donÂ’t realize what we do with the in voice. Nope. As Eric whoÂ’s in charge of accounts payable said, you Â’re all resisting change. It took them one week of receiving all th e invoices to be overwhelmed. As Eric later said, I just didnÂ’t real ize how much you worked with the vendors on getting the correct invo ices or getting discounts, you know, and that was that whole ment ality carried across all of the different modules. The people in the trenches out in the departments were saying, well these are the thi ngs you need and instead trying to listen to them to meet them half way, it was nope and because of that there was that really resistant when the system got turned on. U14For a university this large to have only three or four people doing purchasing is ridiculous. Bu t thatÂ’s because the money is basically in the administrative units and they get the responsibility of handling the details. The systemÂ’s not made for that. ThatÂ’s one of the reasons we were resistant. U7Another difference Â… was the departments and colleges would not receive the invoices directly, th at accounts payable would receive the invoices. And people were not receptive to that either because how does accounts payable know that we received everything. You know, that really should be someth ing that should stay with the departments and the colleges. Lack of Fit U3We had accounts payable that was back-logged, they couldnÂ’t pay invoices, we were spending $100s of $1,000s in late feesÂ… Because they didnÂ’t change anything. They didnÂ’t know how to pay the invoices. ThereÂ’s supposed to be a three-way match.
160 APPENDIX B: SAMPLE QUOTES FOR RESISTANT BEHAVIORS Names and other identifying information have been changed from these quotes
161 Table B-1: Sample Quotes for Resistant Behaviors Behavior Category Sample Quotes U2[The behaviors I have seen are] Quitting or not using it. I have three secretaries who wonÂ’t use it Â– flat out will not use the system. And theyÂ’ve gone to the training. But they will not use itÂ… I fired one person in 38 years. I find it a lot easier to make their lives miserable and get them to quit. U4There were some that very aggressively challenged us and actually had an effect of the de sign itselfÂ… A couple of things that we wanted to do in the billing area were challenged from a couple of colleges around the campus and we backed off and said okay, thatÂ’s not going to serve you well and we came up with an alternative plan. F1(6)They think it is their right to do whatever they want and they donÂ’t have to participate in any of our systemsÂ… So they resist by just not doing it the way we want them to. U2there are people who have quit rather than learn new systems Â– theyÂ’ve retired. Overt-Active U22Well people were getting very, very frustrated, Very frustrated, in fact, I mean, Andrea just got so frustrated and they were just so overwhelmed over th ere so she just found a job in another department and she doesnÂ’t use the system at all and I have some friends who have left areas where they were and we donÂ’t use it. U11They would like to complain that they couldnÂ’t do it. They wouldnÂ’t go to the training, but they would also complain. U8it was a whole two or three months of I donÂ’t like this, IÂ’m not going to use it U7People were very frustrated. It affected morale. People were saying that they who chose to initiate this system into the university were not those working with it. U15-I had many people call me. IÂ’m not sure why they called me, but they said IÂ’m going to quit, because I canÂ’t handle the system. Okay and what are you going to do, why would you want to do that? U7-here are 100 angry people walk ing in [to the orientation session], they donÂ’t know what to expect, theyÂ’re all defensive Overt-Passive U9thereÂ’s a lot of people wa nting to make job changes. Covert-Active U3And everybody was saying, I want to keep my shadow system, because I know what's in this, I can report off this. It's double work, and we were encourag ed to get rid of the shadow system, we need to quit this double data entry.
162 U4I guess one way we discovered who they were they were coming to the cashier office the old way; even bring the codes from the old system. U20I think people could do what they needed to do in the system, but were somewhat afraid, but they avoided what they needed to do in the system. U22Most people avoid the sy stem; most people do avoid the system. U8even though you know its wrong, instead of figuring out what was happening with the wrong information, they would just go in and change it enough to ma ke the transaction go through and therefore they considered th eir work done, but then we had bad data out there and no oneÂ’s gone back to correct that bad data U20They werenÂ’t as productive as they needed to be from the university standpoint because they were hesitant and unsure about themselves in using the system to its fullest capabilityÂ… part of the issue does fall back on us to pr ovide training to the best extent we can on some things that we havenÂ’t done yet like queries in the system and people understand ing the tables in the data warehouse. U21Impatient, especially in th e training. Okay, just show me; just get it over with, why doe s it take so longÂ… you see people become impatient and make little jokes about the system is not really fast and you know all th e time spent, so impatience probably with most of us. F1(3)It was Â… waiting until the very last moment to go to training. They had to extend the window for training since nobody signed up for training until the last two weeks. So they had to redo their whole schedule and make more people available to do the training, so it ki nd of passive resistance. U18-[[In trainingÂ…]] I could hear typing when there was nothing to be typing, so I know they were answering their e-mails or whatever they were doing. Covert-Passive U11they donÂ’t want to be in the training. TheyÂ’re not as receptive and the information takes a lot longer to get in there and itÂ’s a lot harderÂ… if I donÂ’t sign up for it itÂ’ll go away, if I donÂ’t learn this IÂ’ll be able to keep my old way of doing it. It was just like a refusal to admit weÂ’re moving on. I would see that. People would wait till the last minute. Ag ain, just trying to refute the whole thing.
163 APPENDIX C: SAMPLE QUOTES FOR MANAGEMENT STRATEGIES Names and other identifying information have been changed from these quotes
164 Table C-1: Sample Quotes for Manageme nt Strategies to Minimize Resistance Management Strategy Sample Quotes U4we tried first to convince them that the change was one mandatory, two needed and three beneficial to them. U8I think the key things to have the implementation go better would have been communication and involvement of the larger organizational community U13there was communication going out all the time and I think they went to various management meetings saying this is were we are and this is when itÂ’s coming U17from the change management perspective we were trying to communicate the benefits and the whys, the compliancesÂ… [it was] possibly over communicated with to the point where people may have deleted the e-mail without reading it. It was more about the changes, more about the news, the benefits were in it though, IÂ’m sure. U18I think communication was one of the aspects that they used. I know leading up to the go live, th ere were constant e-mails and information going out on our organizationÂ’s home page to organizational newslettersÂ… In terms of helping to at least let the people know that this was coming. Top-down communication [Positive comments] U14Yea, they used to have a lo t of meetings and theyÂ’d come and tell us what their long term goals were. U18we had literally people that I received calls from after the system went live, a week or tw o weeks after the system went live that were still trying to log into the old sy stems to do their requisitions and, you know, I was like, you know, have you been on an island or in a cave or were you on vacation because the old requisition system is gone. ThereÂ’s a new day a coming. I have to do today for my department. So, sorry, youÂ’re out of luck. There we re literally people that just paid no attentionÂ… thereÂ’s still people that just literally chose either not to listen or just paid no attention to it because they didnÂ’t think it applied to them. Top-down communication [Negative comments] U3communication is very bad here Â… they did try to improve it they created web sites, they create d lists of questions and answers. U10I know that the management team listens to what the people have to say and their complain ts and they try to address Listen to Feedback [Positive comments] U3they kind of came in and met with us as a group, getting our concerns, what are you concerned a bout, what are you afraid of, what do you want to see happen, what donÂ’t want to see happen.
165 U11Using the feedback instruments, using the formal communication, using what we experien ced in the training class. A lot of empathy. I can tell you there was a lot of that because we could see people struggle with this thing. We werenÂ’t meaning to cause people stress or, on the contrary we wanted to help them through this thing. Empathy. A lot. U3they did a survey of computer knowledge. I took the survey myself and I think they pretty much put it out to the whole community, and if you said you were going to be a user of this, then you had to take this survey. Â“Do you know how to turn on a computer?Â” I mean it was absurd, a nd I just laughed at that. Do I know how to turn it on? And when I got involved as a trainer that was one of things that came up Â– was this really the talent we have here at the university that you have to as k that question and they said, Â“Unfortunately, yes.Â” You have pe ople in the past that have not gotten onto the computer, but theyÂ’re going to need to now with this system. So, yes, there were just ce rtain people that just did not have the capacity or ability to use the computer, and to get into such a complex system as this was overwhelming for them. F1(6)And so complaints flow uphill to a point, and then they stop there, and then they donÂ’t go any higher, because these people in between canÂ’t make decisions anyw ay, so theyÂ’re not going to make bossÂ’ day bad by complaining about it, because they donÂ’t have to deal with it and they donÂ’t want to make their life hard by making VPÂ’s life hard, so it just stops part-way up the tree. Listen to Feedback [Negative comments] U3[[Did they distribute a questi onnaire to solicit opinions?]] They did do a little bit of that, but proba bly too late, youÂ’d already closed people up. U7if [the shadow system that si mplifies the creation of reports] had not been implemented, I think things would have been worse. I really do. U21they had certain hours set up, they had specific questions, you can go in, they had computers set up so you could actually show facilitators what your problem was Provide Help/Support [Positive comments] U18I went from just a packed open lab to now IÂ’m running it every other month and itÂ’s probably five or six people at a time. So I think that really helped. U6You were just left on your own. You could go to this one or that one, but you were practically on your own. Provide Help/Support [Negative comments] U8[with the new system] thereÂ’s 12 ways to get the same information. Out of 12 reports, thereÂ’ s one that really has everything that you know. We havenÂ’t instruct ed the community on how to go to that one report, so the community ge ts frustrated because they tried report number one through five and it just didnÂ’t give them the information, so they just forget it.
166 U7ThatÂ’s why I went back to the tr ainers. They have to be confident; they have to be told they are knowledgeable of the material. They really have to go above and beyond. They have to take the work home with them to learn it; they have to spend weekends learning. You know, they really have to be dedi cated to it and I really donÂ’t think that was the case. Aside, and of c ourse, case in point is then when I had James it was obvious that he work ed on this 24/7 or close to it, you know and thatÂ’s what so impressive about him and you could just tell. Because you know how hard he worked to learn it. U10as soon as the trainers we re in there and found out what the differences were and they immediat ely got that information out to people and incorporated it into their training. Training [Positive comments] U3So, going in there and knowing already who is going to be my problem child Â– trying to greet them as they come in and encourage them to sit in the front of the cla ss, so that youÂ’re more closely to them, that you can just take one step back and look at their screen and make sure that theyÂ’re on trackÂ… On e of the things that I did when I first started day one when I was te aching this class, I had a little Power Point thing and I showed a bunch of runners to say that this is the race to learn the sy stem and here are you guys back here in this little cluster, right b ack here at the end and guess what IÂ’m just a couple of steps ahead of you. Then you got these other folks up here, youÂ’ve got the vendorsÂ’ experts a nd then youÂ’ve got some of our experts and then you got us, but we only started this only two months before you started to walk through th e door, so donÂ’t expect too much from me. So at least tried to lowe r their expectations quite a bit. Training [Negative comments] U3itÂ’s day two, you need to review what we went over day one and what are we going to do on day two and so he was throwing out candy and he did a little quiz and so I t hought that was great and I came up with questions on day two, so I donÂ’ t feel like sitting up here and doing a boring recap of yesterday, so how about can you tell me, da, da, da Â…. So these people, I can do that, and I threw out a piece of candyÂ… they were fighting to get th e question then. Yeah, that was a good icebreaker. Yeah people like candy. [--The following is the perspective of U2 regarding the candy--] U2ThatÂ’s the kind of stuff that just drove me nuts with those things. And when you have to sit there for three days, folks guess what? My favorite one was when we answered the questions right they threw candy at us. That was our prize fo r getting the right answer Â– they threw candy at us theyÂ’re lu cky I didnÂ’t throw anything else backÂ…ItÂ’s an insult to the menta lity. Hey Suzanne, in your system training, did they throw candy at you ? [Suzanne: Â“Yeah.Â”] See that was their rah-rah thing. [Suzanne: Â“And youÂ’re like Â‘Dude, get that candy out of my face.Â’Â”]
167 U20we probably started the traini ng a little bit late, trainingÂ’s an interesting issue because we trained people before hand and that might minimized you know user, that might fostered user acceptance a little bit better, but the issu e was if you train them too far in advance and they donÂ’t start the proj ect until here then they kind of forget what theyÂ’ve done. F1(2)they had to raffle off mini coopers[car] and so everyone who submitted their timesheets correctly three times in a row were automatically submitted to this raffle for a car. U20I donÂ’t know for a fact, but I thi nk that some of the colleges and departments did provide some financial incen tives to people and I think they sent them to training. U19we tried to take a look at the volume of transactions that they were processing and then tried to gi ve them some financial reward and some recognition for the new skill sets that they had developed and things of that nature. But, it was something that we did with our own resources within our own college that wasnÂ’t done necessarily in other colleges Incentives [Positive comments] U3[As a trainer] Incentives? Yea h, It kind of worked out. Yeah, I think we got $300 a class. So for every class I taught I got a $300 bonus. And I sat there and I looked at the hours I spent and I said I think I made $10 an hour over a period of time. Well, it was a lot of work and I gotta say I was pleasan tly surprised at the university communityÂ… If youÂ’re not a self-m otivated person, youÂ’re just not going to do it. ThereÂ’s just no pay for performance, good job, bad job or whatever you were getting your 2% increase. This year was the first time there was a pay for performance. U21I donÂ’t believe there were any incentives. Incentives [Negative comments] U2[Were there any incentives put in place for you?] Not a thing. HereÂ’s the work Â– do it. U17[the V.P., said] weÂ’re doing this period, get on board, regardless of consequences, re gardless of, weÂ’re doing it periodÂ… Figure it out. Clear Consistent Plan [Positive comments] U11-[[Management consistency]] I think the goals remained pretty much the same. They would maybe sh ift a little bit and maybe delay, have to push a date somewhat, but pretty much remained. Clear Consistent Plan [Negative comments] U8-there was no planning ahead as to what our strategies were going to be. So one moment itÂ’s this, then depending on some meeting they attended, something they read, all of a sudden our direction went this way. So it wasnÂ’t and I certainly think that you can make changes along the way, but youÂ’re talking going from, youÂ’re heading down path A and all of sudden they want you to jump to path Z.
168 U13They were going to implement one version 7.4 I think, I forget now. And then six months into th at they changed and decided to implement 7.8, which was available when they decided to go to 7.4, so we lost a whole bunch of stu ff and a lot of testing time. U5about 9-10 months before the go live date, they had determined that this version of the system th at we were doing that theyÂ’d come out with a new version that was an online web based and at the last minute the university kind of made a decision to go with the web based one, as to the other one which was not really web based and that kind of threw things, not quite out of whack, but you know I think it got quite a few people, you know, not really upset, but concerned that we were planning on this and now al l of a sudden they said we were doing this and now weÂ’re going to do thisÂ… it kind of threw people for a loop that we were going to make this quantum leap you know 9 months ahead of time to go from so mething that theyÂ’d worked about a year on. Management Expertise [Positive comments] [although no comments were identified that directly pertained to the expertise of management, it was demonstrated through the positive comments regarding how management was implementing the system well.] U14the higher level was saying to the masses, kick them in the buttock if they donÂ’t give you what you want, make them create this system the way you want and they were telling us, you canÂ’t change anything, you have to sell it and use it the way it is. How do you reconcile that. From my perspe ctive our biggest enemies, OUR BIGGEST ENEMIES, are the VPs. They have never logged into this system, they have no idea what it means to use this system and they donÂ’t want to know. And they also, again, my opinion, only hear what they want to hear because they tell the lower level echelons this is what youÂ’re going to tell me and that Â’s what youÂ’re going to get told. U19-the vision and view from executive management at that 60,000 foot level is very different from what it is at the grassroots ground level. DevilÂ’s in the details. And th at couldnÂ’t be more true with these software implementations. And, there may have been a little lack of understanding on the part of execu tive management on exactly how many details need to be in place for this thing to work smoothly and maybe a little bit of lack of recogni tion on their part in terms of the talent Management Expertise [Negative comments] U5I think at certain levels the goals were very articulate. At the very high level. And when you get dow n to the unit level maybe there wasnÂ’t that real understanding Â… Th eyÂ’re just concerned about howÂ’s it gonna affect the work that I have to do. And so, you know, these high flutingÂ’ goals, theyÂ’re really good for the right people, but for other people theyÂ’re not. System U8We did not take and change our processes to fit what was now
169 Customizations [Positive comments] going to be in the new system. We took and made the system change to fit the processesÂ… ThereÂ’s quite a few customizations that were done. U14Basically we were told beca use of difficulties with the Banner Oasis project that they did not want to modify this system, that we were to try to make changes in th e university, to change our business practices to work with the system U8[If there were more customizations for the individual departments rather than just the central unit, ] I think Â… there would have been more of a buy in to the system. System Customizations [Negative comments] U20Our system vendor comes up with upgrades all the time. So we made the decision to implement the system vanilla, which means that the system worked a certain way an d we really had to adjust our business process to agree to the way the system worked and you know people were used to doing things the way they wanted to do them
170 APPENDIX D: CODING SCHEME
171 R easons for User Resistance IND: Individual Issue (for users) UN: Uncertainty [User is unclear of the future] (Unknown future, potential threat, lack of clarity) LI: Lack of Input [Users opinions are not considered] (The thoughts and opinions of users were not sought ou LC: Loss of Control/Power [User loses control or loss of recognition as the expert] (leveled playing field, not t expert anymore) SE: Self-Efficacy [perceived lack of capability] (lack of confidence, lack of computer skills/abilities) SYS: System Issue TP: Technical Problems [Problems with the system] (Bugs in system, features that dont work right) CO: Complexity [System is complicated to use] (Difficult to access, Poor user interface that lacks logic or is n intuitive) ORG: Organizational Issue LE: Lack of Facilitating Environment [Organizational cu lture is not conducive to the change] (bureaucracy th a is slow to change) PC: Poor Communication [Communication to users is problematic] (lack of communication, users not hearing benefits of system, users not understanding why) PT: Poor Training [Training does not meet organizational needs] (Lack of training, training seems to be a was t of time, incompetent trainers timing of training, sufficiency of training) PRO: Process Issue CJ: Changed Job/Job Skills [Users job or job skill requirements changes] (Revised job description, different j o tasks, new skills, new way of thinking) AW: Additional Workload [User is required to put fort h additional effort] (extra work, more work to get same extra time) LA: Lack of Fit [Process problem between the system and organizational structure] (problematic changes to processes, new processes not working as planned) R esistant Behaviors OA: Overt-Active [clearly expressed behavior that takes action] (Refusal to use system, challenge system/pla n at system, dont follow process, quit job/job change) CA: Covert-Active [minimally expressed behavior that takes action] (Use shadow system, try to use old syste m avoid system use, enter in info inappropriately) OP: Overt-Passive [clearly expressed behavior that is inert] (Complaints, lower morale, defensive, turnover intention) CP: Covert-Passive [minimally expressed behavior that is inert] (Not motivated, less productive, impatient, no p aying attention, procrastinate, dont want to learn) M ana g ement Strate g ies to Minimize Resistance --Include + or when coding ECO: Effective Communication TD: Top-down communication [Top management/implementation team communicating to users] (communicat the types of changes to occur, the benefits of the system, the goals and vision, the whys, managers shari n information with subordinates) LF: Listen to Feedback [Management listening and res ponding to the input of users] (distribute/collect questionnaires, address complaints) EES: Effective Education/Support PH: Provide Help/Support [Management offering assistance to users] (availability of consul t ants or helpline, providing a support system to interface with the system) UT: Useful Training [Train the users at an appropriate time in a way that is suitable for their needs] (Trainers w knowledge/communication skills, address the needs of trainees, appropriate time frame) AI: Appropriate Incentives [Suitable motivators to users to learn and use the system] (incentives to take trainin g and to do extra work) EDP: Effective Direction/Planning CC: Clear Consistent Plan [Straightf orward consistent strategies] (Clear direction, consistent management strategies, following through with plans || opposite: confusion, failure to carry out plans) ME: Management Expertise [Management understanding of processes and system] (Decision makers understa n system and processes, Decision Makers understand the details) SC: System Customizations [Customize the system to the processes in place] (tailor system to fit user needs)
172 APPENDIX E: INTERVIEW SCRIPT
173 Interview Script The interview script is the following questions although follow up questions will also be included based on these questions when appropriate. Background of interviewee Please describe your involvement in th e Enterprise System implementation, the amount of time you were involved in the project, and the name/type of system Change What degree of change has the Enterprise system had on your job? To what extent were employees of your organization affected by changing jobs and responsibilities because of the system implementation? For you, what were the advantages and disadvantages of the project? What did you gain and lose becaus e of the system implementation? Resistance Describe the type or types of resistan ce that occurred during the implementation. Why do you think this resistance occurred? Do you think anything could have been done differently to reduce the level of resistance? How does the phase of implementation aff ect the level or ty pe of resistance? Describe the conflict between management and users (what type of conflict, how was it resolved, etc.) What types of things, if any, attracte d users to embrace the system and change? What was the nature of goal-setting? For example, did management set all goals near the beginning, or were some goals se t, then change Â– was one option more conducive to resistance? Management Strategies What strategies did management take in dealing with resistance? To what degree is the vision and plans of management clear to you? How consistent are the vision and plan s of management in your organization? Did management appear to be committed to seeing this system implemented and used? How was the training? Extra To what degree was there training in using the system and what are the strengths/weaknesses of the training?
174 APPENDIX F: Q-METHODOLOGY QUESTIONNAIRE
175 STUDY INFORMATION SHEET The following information is being presented to help you decide wh ether or not you want to be a part of a minimal risk research study. Please read carefully. If you do not understand anything, ask the pers on in charge of the study. Title of Study: Rethinking User Acceptance: An Examination of User Resistance in Mandatory Adoption of Enterprise Systems Principal Investigator: Timothy Klaus The purpose of this research study is to be tter understand user resistance in the implementation of an Enterprise (ERP) System. Your participation will include completion of this questionnai re and will take approximately 15 minutes. You will not receiv e benefits from participating in this research and there are no known risks involve d. Your privacy and research records will be kept confidential to the extent of the law. Authorized research pe rsonnel, employees of the Department of Health and Human Servi ces and the USF Institutional Review Board, its staff, and others acting on behalf of USF, may inspect the records from this research project. The results of this study may be publis hed. However, the data obtained from you will be combined with data from other pe ople in the publication. The published results will not include your name or any other info rmation that would personally identify you. Your decision to participate in this study is completely voluntary. You are free to participate in this study or to withdraw at any time. If you have any questions after completing this study or would like to revi ew the results of the study upon completion, please contact: Tim Klaus Â– (813)974-6751 or firstname.lastname@example.org If you have questions regard ing your rights as a person who is taking part in a research study, you may contact a member of the Division of Research Compliance of the University of South Florida at 813-974-5638. You are guaranteed total anonymity All information you provide will be used exclusively within th e bounds of this study and nothing will be used to identify you. None of the information you provide will be shared with your employer, or any other person or entity. Participation in this study is voluntary, and will not adversely affect your job.
176 Thank you for participating in th is study! This is a study on employees interacting with an Enterprise System (i.e., Peoplesoft-Oracle, SAP, Baan, and a number of other large-scale systems). You should complete this questionnaire only if you have been employed while one of these systems was put in place and you used the system for your job. If you can not participate in the study, please ta ke a minute to complete this page of the questionnaire. Doing so will help validate the quality of the sample by providing some quick demographic information. General Information What is your gender? Male Female What is your highest level of education? High School AssociateÂ’s Degree BachelorÂ’s Degree MasterÂ’s Degree Doctoral Degree How many years have you been in your current position? ___________ How many years have you been in your organization? ___________ What is your age? Under 25 26-35 36-45 46-55 Above 55 What best describes your position? Clerical/Data Entry Support Staff IT Staff Supervisor Mid-level manager Top management How many employees are in your organization? Under 50 50 to 100 101 to 500 501 to 1,000 1001 to 5,000 Over 5,000 What is the industry of your organization? Government Financial Services Utilities Manufacturing Insurance Healthcare Retail Education High-tech Other ___________________ The remaining questions on this questionna ire are regarding your experiences with the implementation of an En terprise System/ERP System. What is the scope of the organizationÂ’s system? One location Regional National Global What is the vendor of the system? SAP Peoplesoft/Oracle Baan Computer Associates Siebel J.D. Edwards DonÂ’t Know Other _________________________ How many days were you in training to learn the system? ___________ How many days were there between when you finished training and when you starting using the live system? _____ What modules of the system have you used ( all that apply)? Purchasing Production Finance Customer Management Maintenance Human Resource Inventory Shipping/Distribution Receiving B2B Commerce Billing Other
177 For the next two pages, you are asked to sort the sets of statements. Sort the items "from the outside in". Start with step 1 by selecting two statements that are the Â“Most RepresentativeÂ” of your experience during the system implementation. Next fill in the boxes for step 2 by entering the Â“Least RepresentativeÂ” statements. Continue by filling in the four boxes for both steps 3 and 4. Finally, fill in the five boxes for both steps 5 and 6. Please pay attention to make sure that you enter an item only once. Rate which statements are representative of your experience during the system implementation Step 1 Most Representative (2 items) Step 3 Representative (4 items) Step 5 Somewhat Representative (5 items) Step 6 Minimally Representative (5 items) Step 4 Slightly Representative (4 items) Step 2 Least Representative (2 items) # Statements # Statements 1 I was not comfortable with the level of certainty regarding how the system would affect my future 16 My organizationÂ’s internal environment is not conducive to changes brought about by the system 2 I did not have sufficient input into how the system implementation would occur 17 There was poor or problematic communication to me during the system implementation process 3 I lost control/recognition of my expertise 18 Training was poor 4 I refuse to use the system 19 I complain to others about the system 5 The system required capability/skills that I lacked 20 I am defensive because of the system 6 I try to hack at the system 21 I am demotivated by the system implementation 7 The use of the system required that my job or required job skills changed 22 I decrease my level of productivity in protest because of the system 8 I donÂ’t follow the system processes I was told to follow 23 I am impatient during the system training 9 I intentionally perform my job in a different way than IÂ’m supposed to in protest 24 I quit my job or changed to a different position at my job because of the system 10 I try to do my job the old way 25 I do not want to learn the system 11 I avoid using the new system whenever I can 26 I intend to quit my job, but never took action on it 12 I inappropriately enter information into the system 27 The system seemed complicated to use 13 I had to put forth additional effort because of the system 28 I procrastinate when I can 14 I experienced technical problems with the system 15 I challenge the system implementation plan 29 There were problems with the new processes that were put in place because of the system Please double-check to make sure that the items you entered have only been entered once and that all boxes are filled. For the following two questions, pl ease answer each question with a minimum of two sentences: Why did you choose the two Â“Most RepresentativeÂ” st atements? [space provided to answer question in actual questionnaire] Why did you choose the two Â“Least RepresentativeÂ” statements? [space provided to answer question in actual questionnaire]
178 Rate the following statements regarding how much you would have desired the following management strategies in the system implementation Step 1 Most Desirable (1 item) Step 3 Desirable (2 items) Step 4 Slightly desirable (2 items) Step 2 Least desirable (1 item) # Statements Examples 1 Top management/ implementation team communicates to users Communicating the types of changes to occur, the benefits of the system, the goals and vi sion, the Â“whysÂ”, managers sharing information with subordinates 2 Management listens and responds to the input of users Distribute/collect questionnaires, address complaints 3 Management offers assistance to users Availability of consultants or helpline, providing a support system to interface with the system 4 Users are trained in a way that is suitable for their needs Trainers with knowledge/comm unication skills, address the needs of trainees, appropriate time frame 5 Suitable motivators are offered to users to learn and use the system Incentives to take traini ng and to do extra work 6 There is a clear and consistent implementation plan Clear direction, consistent management strategies, following through with plans 7 Management understands the work processes and the system Decision makers understand system and processes, Decision Makers understand the details 8 The system is customized to the processes in place Tailor the system to fit th e usersÂ’ preferences/needs Why did you choose the Â“Most DesirableÂ” and Â“Least DesirableÂ” statements? ___________________________________________________________________________________________ ___________________________________________________________________________________________ ___________________________________________________________________________________________ ___________________________________________________________________________________________ Thank you! Your participation is greatly appreciated. Please take a moment to make sure you have answered all questions. Do you wish to receive a copy of the results of this study? If so, please provide your email address below or send an email to email@example.com to request a copy: Email: ___________________________________________________ Please return the completed form to: Tim Klaus University of South Florida 4202 E. Fowler Ave., CIS 1040 Tampa, FL 33620-7800 (813) 974-6751 firstname.lastname@example.org 2005 Tim Klaus
179 APPENDIX G: SAMPLE OF CONS ISTENT/INCONSISTENT CODING Names and other identifying information have been changed from these quotes
180 Table G-1: Examples of Consiste nt Coding Â– Reasons for Resistance Quote # Quote Reason for Resistance 1 G6-those staff working in these sections are showing some resistance because they have fear they might loose their job Uncertainty 2 G1-instead of spending a lot of time in terms of entries and all that, things will be captured, probably more time will be spent in terms of analysis, review things a nd sort of better improve things Job/Job Skills Change 3 G2maybe they are not so comfortable, the uncertainty of whatÂ’s lying ahead. Uncertainty 4 N3instead of having to go to two pages to enter a purchase order, in the new system, theyÂ’d have to go to like four different sc reens to capture all the information they had to capture. So in that respect itÂ’s taking longer to do data entry. Workload 5 N2-a lot of the reports that weÂ’re spitting out like journal entries that we pr epare, thereÂ’s was data that was not on them or the formatting was rather awkward and then it was very messy. Technical Problems 6 N2-[the training] was really kind of just a waste of time and it was well after we had already started closing our first month anyways. Training 7 N3a lot of key users pr obably felt like their input was not solicited. Input
181 Table G-2: Example of Inconsiste nt Coding Â– Reasons for Resistance Quote # Quote Coder1 Coder2 1 G6-the main fear for them is that there will be reduction in manpower. Uncertainty Control/Power 2 G2-I think itÂ’s because youÂ’re threatening their comfort zone Â… will I be able to do my job in the future or is it going to be very complicated Uncertainty Self-Efficacy 3 N3-Some of their jobs actually became more complex Job/Job Skills Change System Complexity 4 N3People are uncomfortable when their jobs change and have to learn new tools, etc. Job/Job Skills Change Self-Efficacy 5 N3I think there might have been an impression that we were going to gain there, but it actually became less efficient. Workload Technical Problems 6 N2-teams discussed, well, do we really need it and how are they going to get it for us and in a lot of cases it turns into a customization. ThereÂ’s custom reports that they have to build, which we actually just this week finally got the custom report we requested a year ago or over a year ago to work. So, it took awhile Facilitating Environment Management Strategy Â– System Customizations 7 N2-[there were issues such as] response time, lag time, really slow, very very slow, Technical Problems Workload 8 N4That was the biggest change when this happened. I know how to do this in with another vendorÂ’s software, but now I donÂ’t know how to do it with our new vendorÂ’s software. Self-Efficacy Control/Power
182 Table G-3: Example of Consiste nt Coding Â–Resistance Behavior Quote # Quote Resistance Behavior 1 G6-They take a month to do the process when this information can be provided within two hours. Covert-Passive 2 G2-[What types of resistance was there?] Mostly complaints. Overt-Passive 3 N3-[There were] lots of tickets late in the system when it wasnÂ’t really an e rror with the system, it was just not following the new process Overt-Active 4 N3-I would say it was almost a level below middle management where it was the worker bees complaining how things didnÂ’t work. Overt-Passive 5 N1What type of behaviors? In some cases they would revert to their old way of doing things Covert-Active 6 People went back to the ol d school and still tried to do things the way they did before Covert-Active Example of Inconsistent C oding Â– Resistance Behavior None found Â– Coders co nsistently coded ever y resistant behavior.
183 Table G-4: Example of C onsistent Coding Â– Manageme nt Strategy to Minimize Resistance Quote # Quote Management Strategy 1 G7-The only thing that I remember we have done is just customize some of the reports System Customization 2 G1-[have the vision and plans been pretty consistent over time or they have changed during the re-engineering of the processes?] This time it has been consistent, the way IÂ’ve seen it, it has been consistent, yes. Clear Consistent Plan 3 G6-everybody in the company knows what we are doing and why we are doing this. Top-Down Communication 4 N3We had to customize qu ite a bit, yes, for it to do everything that we needed it to do. System Customization 5 N3we had quite a few super users out in the field into specific locations and to assist the users when they had issues in the system and in effect they helped them resolve it. Provide Help/Support 6 N3There should have been more incentive to take the training. Incentives 7 G3-[so you will essentially be the expert of your area and train everyone who is going to be using that module?] True. True. Management Expertise
184 Table G-5: Example of In consistent Coding Â– Manage ment Strategy to Minimize Resistance Quote # Quote Coder1 Coder2 1 once weÂ’ve completed the phase where weÂ’ve designed the system and weÂ’ve got an environment where we can actually be testing, te st all the processes and so forth. At that time, you know, weÂ’ll be contacting every training positions for all the end user community. Training Top-Down Communication 2 I think 18 to 24 months would have been a much more realistic time frame. It would not have been necessarily cost more it just would have spread out more so people had more time to review and provide feedback to the system Listen to Feedback Training 3 N3-[Was the training optional then?] It was and it probably should have been required. Training Reason for resistance Â– training 4 N1At one point, ri ght before we went live, we had a lock down of customizations Â… when you have to make your case [for customizations] in front of the CIO, corporate control youÂ’d better have a pretty strong case. System Customizations Listen to Feedback 5 N2-[was it the customizations that they put in that made it hard?] Yes, the rules and the customizations and the way they wanted things built. System Customizations Reason for resistance Â– Lack of Fit
185 APPENDIX H: CONCOURSE STATEMENTS
186 Table H-1: Concourse Statements for Reasons for User Resistance Reason for User Resistance Concourse Statement Uncertainty I was not comfortable with the level of certainty regarding how the system would affect my future Lack of Input I did not have sufficient input into how the system implementation would occur Loss of Control/Power I lost cont rol/recognition of my expertise Self-Efficacy The system required capability/skills that I lacked Technical Problems I experienced t echnical problems with the system Complexity The system seemed complicated to use Lack of Facilitating Environment My organizationÂ’s intern al environment is not conducive to changes brought about by the system Poor Communication There was poor or problematic communication to me regarding the system implementation Poor Training Training was poor Changed Job/Job Skills The use of the sy stem required that my job or required job skills changed Additional Workload I had to put forth additional effo rt because of the system Lack of Fit There were problems w ith the new processes that were put in place because of the system
187 Table H-2: Concourse Statemen ts for Resistance Behaviors Behavior Type Concourse Statement Refusal to use system I refuse to use the system Challenge system/plan I challenge the system implementation plan Hack at system I try to hack at the system DonÂ’t follow process I donÂ’t follow the sy stem processes I was told to follow Quit job/job change I quit my job or ch anged to a different position at my job because of the system Use shadow system I intentionally perf orm my job in a different way than IÂ’m supposed to in protest Try to use old system I try to do my job the old way Avoid system use I avoid using the new system whenever I can Enter in info inappropriately I inappropri ately enter information into the system Complaints I complain to others about the system Defensive I am defensive because of the system Turnover Intention I intend to quit my job, but never took action on it Not Motivated I am demotivated by the system implementation Less Productive I decrease my level of productivity in prot est because of the system Impatient I am impatient during the system training Procrastinate I procrastinate when I can DonÂ’t want to learn I do not want to learn the system Table H-3: Concourse Statements for Management Strategies Management Strategy Concourse Statement Top-down communication Top managemen t/ implementation team communicates to users Listen to Feedback Management listens and responds to the input of users Provide Help/Support Management offers assistance to users Useful Training Users are trained in a way that is suitable for their needs Appropriate Incentives Suitable motivators are offered to users to learn and use the system Clear Consistent Plan There is a clear and consistent implementation plan Management Expertise Management understands the work processes and the system System Customizations The system is customized to the processes in place
188 APPENDIX I: QUOTES FROM QU ESTIONNAIRE FOR EACH GROUP
189 Table I-1: Quotes from Questionnaire Respondents Group Questionnaire# Quote 36 My knowledge base was taken away, system that was in place we had for yearsÂ…To implement a new system, the decisions makers need to understand what the system does and what we need the system to do. 61 I tried to be optimistic about the use of the system. I thought it would have been better for the use of the company in the long run...MGM T must not only listen to input but must respond and act. 71 Must have user input to be successful. 150 We are afraid of all changesÂ…The way of working is difficult to change...the usage is easier if you have a clear implementation plan. 161 The implementation was rushed and not effectively communicated within the or ganization, when questions were asked; they were not addressed. 1 177 Without a clear and strong implementation plant the project will fail. I think peopl e are either motivated or not. Incentives are short term fix for people. 109 One reason for resistance is the lack of good training. 133 I had to change the way in which I was organized in order to attend to the important issues Most of the people at the company have been around and fear changes. 133 Because we implemented first and looked at the processes later this complicated thi ngs and were caused by not knowing things. 140 There is a strong resistance to change for fear of learning new and more efficient methods. 2 140 Good planning marks the Institu tion's future on educational and managerial levels, and it improves service. Success depends on mentality changes from the top. 70 We still have problems that have not been solved, that the new system is not designed forÂ…The advantages were apparent and instead of rejec ting the change we tried to cooperate with the transition. 101 It is a hard system if you are not trained and additional training was needed...Everyone needed more training...Not all systems used in the plant would communicate with the new system. 3 116 I only received 1.5 hours of traini ng ... for a total of 6 days in class.
190 119 It seemed complicated at first, more effort was expended at figuring our the new processes. 152 The system was new and unknown, not sure what the future would hold. 171 No clear directions on proce sses and technical process...I believe management should liste n to the users, since they are using the system everyday. I don't believe in motivators to help the user learn the system. 182 Need to know how to do job; not why. 87 I had not needed financials in the old system I had to put forth effort in this one...Planni ng is critical to the success. 93 During the implementation pro cess, I witnessed confusion & chaos...My excitement of learning this new process diminished when I saw poor management. 110 A clear project management pl an helps to keep things moving. 138 We had to change policies and processes which had been in use for many years. 139 We have had technical problems due to poor infrastructure...A consistent plan everyone involved knows is important so we all go towards the same goal. 155 there were issues with the new system had they communicated properly to the implementation team, they could have been resolved. 4 165 It required extra work to define requirements, learn processes, & report issues... system basics required to process changes...implementati on plan being understood is critical to success & acceptan ce by users. should not have to offer incentives if system improves things. 42 System is complicated and requires a good deal work to accomplish the same purpose. 42 I never intended to quit due to the system I was somewhat perturbed...If management understood the system better there would be less redunda ncy and the system would work better. 47 I think there should have been some incentive to acquire the knowledge, management fails to comprehend the amount of time we had to spend away from our jobs. 68 Customizing the system to fit the users is of course the most desirable. 5 82 The implementation was difficu lt but I never thought of changing itÂ…There needs to be some type of motivation [for] using the system.
191 115 The system changed the way our university did business. It took 4 times the steps to do the same processes. 147 It was difficult to change our working habits and did not know if it would affect my job...I would have liked my boss to understand the system and avoid misunderstandings. 120 [I] had to learn additional skills ... there were problems which I did not understand. 143 We have had to adapt because the system has not been tailor made, we have had t echnical problems when running some of the processes. 143 We did not have communicati on at all levels of the organizationÂ… I believe there was a plan. 173 My position changed from having very little to do with the computer with having more. 6 173 There was only incentives for those directly involved with the implementation not for everyone else. Had to change our whole system to match the program. 62 [A] clear implementation plan needs to meet expectations. 65 I was unfamiliar with the syst em. On initial use of the system it did not functionÂ…[I placed] technical support calls to fix the problem. 65 Tailoring the system is necessary to complete the job. Motivators are just fluff, you are either motivated or you are not. 7 81 The system was intense and overwhelming at the start. 49 The new process is more cumbersome than the old and requires learning more processes for the new system...I don't feel Mgmt. understood the work process. 55 The new system is complicated, and is stressful additional training is needed. 84 The system was new to me and had lots of technical problems. [I] had to put forth more effort in learning this system...a system that is not catered to your company's needs ... is useless. 122 We had a lot of technical problems system Â…We had a trainer but her knowledge of what we did and why was not good. We needed to know where our information was going and how to extract it. 154 Our skill set changed and there was no recognition of past experience. 8 180 I had to play around with it to find how to do things, and read a 500 word page manual for each function...Motivators do not work with a poor system.
192 APPENDIX J: DEMOGRAPHICS OF QUESTIONNAIRE RESPONDENTS
193 Table J-1: Gender Gender Frequency Percent Valid Percent Cumulative Percent Male 58 38.2 38.4 38.4 Female 93 61.2 61.6 100.0 Valid Total 151 99.3 100.0 Missing System 1 .7 Total 152 100.0 Table J-2: Education Education Frequency Percent Valid Percent Cumulative Percent High School 27 17.8 18.0 18.0 Associate's Degree 17 11.2 11.3 29.3 Bachelor's Degree 66 43.4 44.0 73.3 Master's Degree 36 23.7 24.0 97.3 Doctoral Degree 4 2.6 2.7 100.0 Valid Total 150 98.7 100.0 Missing System 2 1.3 Total 152 100.0 Table J-3: Age Age Frequency Percent Valid Percent Cumulative Percent Under 25 1 .7 .7 .7 26-35 21 13.8 15.6 16.3 36-45 39 25.7 28.9 45.2 46-55 50 32.9 37.0 82.2 Above 55 24 15.8 17.8 100.0 Valid Total 135 88.8 100.0 Missing System 17 11.2 Total 152 100.0
194 Table J-4: Position Position Frequency Percent Valid Percent Cumulative Percent Clerical/Data Entry 8 5.3 6.0 6.0 Support Staff 43 28.3 32.1 38.1 IT Staff 11 7.2 8.2 46.3 Supervisor 13 8.6 9.7 56.0 Mid-level Manager 55 36.2 41.0 97.0 Top Management 4 2.6 3.0 100.0 Valid Total 134 88.2 100.0 Missing System 18 11.8 Total 152 100.0 Table J-5: Employees in Organization Employees in Organization Frequency Percent Valid Percent Cumulative Percent Under 50 19 12.5 12.8 12.8 50 to 100 4 2.6 2.7 15.4 101-500 32 21.1 21.5 36.9 501-1000 19 12.5 12.8 49.7 1001-5000 49 32.2 32.9 82.6 Over 5000 26 17.1 17.4 100.0 Valid Total 149 98.0 100.0 Missing System 3 2.0 Total 152 100.0
195 Table J-6: Industry of Employer Organization's Industry Frequency Percent Valid Percent Cumulative Percent Government 1 .7 .7 .7 Manufacturing 22 14.5 14.6 15.2 Healthcare 2 1.3 1.3 16.6 Retail 2 1.3 1.3 17.9 Education 105 69.1 69.5 87.4 High-tech 10 6.6 6.6 94.0 Other 9 5.9 6.0 100.0 Valid Total 151 99.3 100.0 Missing System 1 .7 Total 152 100.0 Table J-7: Scope of OrganizationÂ’s System Scope of System Frequency Percent Valid Percent Cumulative Percent One location 31 20.4 21.2 21.2 Regional 46 30.3 31.5 52.7 National 33 21.7 22.6 75.3 Global 36 23.7 24.7 100.0 Valid Total 146 96.1 100.0 Missing System 6 3.9 Total 152 100.0
196 Table J-8: System Vendor System Vendor Frequency Percent Valid Percent Cumulative Percent SAP 4 2.6 2.7 2.7 Peoplesoft/Oracle 131 86.2 87.3 90.0 J.D. Edwards 9 5.9 6.0 96.0 Don't Know 2 1.3 1.3 97.3 Other 4 2.6 2.7 100.0 Valid Total 150 98.7 100.0 Missing System 2 1.3 Total 152 100.0 Table J-9: Statistics for Numeric Demographics Statistics Years in Current Position Years in the Organization Days in Training Days between finishing training and using live system Valid 147 147 138 131 N Missing 5 5 14 21 Mean 5.64 10.58 14.667 55.10 Median 5.00 8.50 5.000 7.00 Std. Deviation 4.761 7.699 26.5729 132.038 Minimum 0 1 .0 0 Maximum 30 34 180.0 730
197 Table J-10: ES Modules Used by Respondents FrequencyPercentValid Percent Did not use module 8455.356.0 Used module 6643.444.0 Total 15098.7100.0 Missing 21.3 Purchasing Total 152100.0 Did not use module 11374.375.3 Used module 3724.324.7 Total 15098.7100.0 Missing 21.3 Production Total 152100.0 Did not use module 8555.957.0 Used module 6442.143.0 Total 14998.0100.0 Missing 32.0 Finance Total 152100.0 Did not use module 10468.469.3 Used module 4630.330.7 Total 15098.7100.0 Missing 21.3 Receiving Total 152100.0 Did not use module 11374.375.8 Used module 3623.724.2 Total 14998.0100.0 Missing 32.0 Customer Management Total 152100.0 Did not use module 12280.381.3 Used module 2818.418.7 Total 15098.7100.0 Missing 21.3 Billing Total 152100.0 Did not use module 13588.890.0 Used module 159.910.0 Total 15098.7100.0 Missing 21.3 Maintenance Total 152100.0 Did not use module 9965.166.9 Used module 4932.233.1 Total 14897.4100.0 Missing 42.6 Human Resource Total 152100.0 Did not use module 11676.377.9 Used module 3321.722.1 Total 14998.0100.0 Missing 32.0 Inventory Total 152100.0
198 Did not use module 14696.198.0 Used module 32.02.0 Total 14998.0100.0 Missing 32.0 B2B Commerce Total 152100.0 Did not use module 12682.984.0 Used module 2415.816.0 Total 15098.7100.0 Missing 21.3 Shipping/ Distribution Total 152100.0 Did not use module 8656.657.7 Used module 6341.442.3 Total 14998.0100.0 Missing 32.0 Other Total 152100.0
199 ABOUT THE AUTHOR Timothy Paul Klaus is currently an Assist ant Professor of Management Information Systems at Texas A&M University Â– Corpus Ch risti. He received a Master of Business Administration degree and Master of Science Degree in Computer Science from Illinois State University. While completing his sec ond MasterÂ’s Degree, he conducted seminars for businesses on Enterpreneurship, Business Plan Development, and Business Growth Strategies and has been a c onsultant to several organiza tions. His current research interests are in User Resistance, IT Personnel, and Global IT Systems. After living the first 18 years of his life in Japan, he moved to the U.S. to start his college career. His undergraduate de grees are in International Business and Organizational Leadership with a minor in Japanese Studi es. He has enjoyed traveling and meeting people from various cultures. After getting married, his wife often travels with him and has seen much more of the world than she ever thoughtÂ…