78168532

RESEARCH ARTICLE
INFORMATION TECHNOLOGY IMPLEMENTERS RESPONSES 1 TO USER RESISTANCE: NATURE AND EFFECTS

Suzanne Rivard
HEC Montral, 3000 Chemin-de-la-Cte-Ste-Catherine, Montral H3T 2V7 CANADA {suzanne.rivard@hec.ca}
Liette Lapointe
Desautels Faculty of Management, McGill University, 1001 Sherbrooke Street West, Montral H3A 1G5 CANADA {liette.lapointe@mcgill.ca}
User resistance has long been acknowledged as a critical issue during information technology implementation. Resistance can be functional when it signals the existence of problems with the IT or with its effects; it will be dysfunctional when it leads to organizational disruption. Notwithstanding the nature of resistance, the implementersbusiness managers, functional managers, or IT professionalshave to address it. Although the literature recognizes the importance of user resistance, it has paid little attention to implementers responsesand their effectwhen resistance occurs. Our study focuses on this phenomenon, and addresses two questions: What are implementers responses to user resistance? What are the effects of these responses on user resistance? To answer these questions, we conducted a case survey, which combines the richness of case studies with the benefits of analyzing large quantities of data. Our case database includes 89 cases with a total of 137 episodes of resistance. In response to our first research question, we propose a taxonomy that includes four categories of implementers responses to user resistance: inaction, acknowledgment, rectification, and dissuasion. To answer our second question, we adopted a set-theoretic analysis approach, which we enriched with content analysis of the cases. Based on these analyses, we offer a theoretical explanation of how implementers responses may affect the antecedents that earlier research found to be associated with user resistance behaviors. Keywords: User resistance, information technology implementation, implementers response, theory building, case survey, set-theoretic analysis
Introduction1
Since the start of the industrial revolution, as processes have evolved from manual labor to operator-dependent machines,
1
Ola Henfridsson was the accepting senior editor for this paper. Suzanne Pawlowski served as the associate editor. The contribution of the authors was equal. The appendices for this paper are located in the Online Supplements section of the MIS Quarterlys website (http://www.misq.org).
semi-automated machines, and robots, the introduction of technologies has often been met with worker resistance. In the 19th century, when industrialization brought about fundamental changes such as the use of machinery and collective work hours, factory owners faced much resistance, manifested as criticism, vandalism, and strikes (Littler 1982). Since the second half of the 20th century, information technology implementersbusiness or functional managers and IT professionalhave often faced user resistance to IT innovations, including electronic data-processing equipment (Mann and Williams 1960), management information systems
MIS Quarterly Vol. 36 No. 3 pp. 897-920/September 2012
897
Rivard & Lapointe/IT Implementers Responses to User Resistance
(Dickson et al. 1974), decision support systems (Alter 1980), clinical information systems (Poon et al. 2004), and enterprise systems (Kitto and Higgins 2010). In this study, we espouse the view that user resistance to IT implementation is neither good nor bad (Hirschheim and Newman 1988; Marakas and Hornik 1996; Markus 1983). At times, it is a means for users to convey the existence of problems with the IT or with its effects; in such instances, resistance is functional. At other times, however, resistance can be destructive, because it generates conflict and ill-will (Markus 1983, p. 433). Furthermore, when resistance prevents the adoption of an IT that could benefit the organization, it is dysfunctional. Notwithstanding whether resistance is functional or dysfunctional, implementers must address it. If resistance is functional and conveys a message about problems with the IT or with its effects, implementers bear the responsibility for taking this message into account and responding in a way that will keep the organization out of jeopardy. If resistance is dysfunctional, implementers are responsible for taking appropriate action to ensure that it does not result in major organizational disruptions. Although models have been proposed to explain why and how resistance to IT implementation develops, there is a paucity of studies pertaining to implementers responsesand their effectwhen resistance occurs. The sparse literature on implementers actions with regard to user resistance focuses on actions to prevent it. Only a handful of authors have referred, albeit tangentially, to implementers responses to user resistance behaviors (e.g., Ferneley and Sobreperez 2006; Marakas and Hornik 1996; Markus 1983; Martinko et al. 1996). Given the significance (for both research and practice) of user resistance to IT implementation as well as the scarcity of research on the nature and effects of implementers responses to user resistance, this study is intended to advance knowledge of the issue. More specifically, it focuses on implementers responses once user resistance has occurred. We first adopt a descriptive stance to address the question: What are implementers responses to user resistance to IT? We then adopt an explanatory stance to address the question: What are the effects of these responses on the level of user resistance behaviors? Our study answers these questions by first proposing a taxonomy of implementers responses to user resistance. Second, it proposes an explanation of the effects of implementers responses on user resistance behaviors, one that captures the influence of these responses on the intensity of resistance. We adopted a case survey methodology, which combines the richness of case studies with the benefits of
analyzing large quantities of data. Our case database includes 89 IT implementation cases comprising a total of 137 episodes of resistance. We coded each episode using both an initial coding scheme and open codes. During this process, we conducted a two-round Delphi study to refine our level of resistance behavior measure. The data was analyzed in three steps. First, from the implementers responses identified in the coding phase, we generated a taxonomy. Second, we conducted content and set-theoretic analyses of our data (Ragin 2008) to identify patterns of influence of implementers responses on user resistance. Third, enfolding extant literature to refine our understanding of the patterns of influence, we developed a theoretical explanation. Our study makes a number of contributions. The first contribution is empirical, since our taxonomy advances current knowledge on how implementers actually react to user resistance. The second contribution is theoretical, in that by developing an explanation of the effects of implementers responses to user resistance behaviors, our study expands on prior IT resistance research. Indeed, through a fine-grained analysis of the data, we develop a theoretical explanation of how implementers responses may affect the antecedents that earlier research found to be associated with user resistance behaviors. The third contribution is methodological: by adopting a case survey method and a set-theoretic approach to analyze the case data, both infrequently used by IS researchers, the study extends the array of methodological tools used in IS research. Our research also offers contributions to practice. It first reveals that inaction, which is the most common response of implementers, is systematically associated with an increase in resistance. Therefore, our study highlights the importance of implementers actively responding to user resistance. Second, by shedding light on the effects of implementers responses to user resistance, it can help them select the responses that are most likely to be effective when addressing user resistance.
Conceptual Background
User Resistance to IT Implementation
Numerous studies refer to user resistance as an important issue whenever it occurs during an IT implementationfrom initiation of a project to post-installation of a system. Although most studies simply acknowledge the significance of this issue, some devote more attention to it, focusing on its antecedents. Among these, eight models propose an explanation as to why and how user resistance develops (Ferneley and Sobreperez 2006; Hirschheim and Newman 1988; Joshi 1991; Kim and Kankanhalli 2009; Lapointe and Rivard 2005;
898
MIS Quarterly Vol. 36 No. 3/September 2012
Marakas and Hornik 1996; Markus 1983; Martinko et al. 1996). We synthesize this literature in light of the five basic elements of resistance proposed by Lapointe and Rivard (2005): manifestations of resistance, subjects of resistance, object of resistance, perceived threats, and initial conditions. Manifestations of resistance represent the core element in resistance to IT, which is generally defined as a set of behaviors enacted by users to manifest some discontent with the implementation of a new IT. The literature suggests a broad range of manifestations, from apathy (Keen 1981; Lee and Clark 1997) to sabotage (Day 2000; Moreno 1999) and destructive behaviour (Ferneley and Sobreperez 2006), and including denial (Kim and Kankanhalli 2009; Munkvold 1999), persistence of former behavior (Dos Santos and Sussman 2000; Kim and Kankanhalli 2009), and the formation of coalitions (Joshi 1991; Lapointe and Rivard 2005). Some manifestations are mild or weak forms of resistance, others are strong but not destructive behaviors, while still others seek to create disruptions and may even be destructive. The subject of resistance refers to the actor or actors exhibiting resistance behaviors. In some instances, the subject is an individual (Joshi 1991; Marakas and Hornik 1996; Martinko et al. 1996) but it may also be a group (Lapointe and Rivard 2006; Markus 1983) or an organization (Ang and Pavri 1994). The object of resistance is the target of the resistance behaviors. In some cases, it is the system itself and its features (Gunawardane 1985; Wagner and Newell 2007). In other cases, it is associated with the significance that the system has to the user, such as a loss of power (Markus 1983) or a loss of status (Lapointe and Rivard 2005). Finally, the implementers themselves may become an object of resistance when a situation has been politicized to the point where the system becomes a pawn in a power struggle between the users and the implementers (Lapointe and Rivard 2006). Perceived threats correspond to the negative assessments that users make of the IT implementation. For instance, Joshi (1991) builds on equity theory and proposes a model of resistance wherein individuals assess an IT implementation in terms of equityor the lack thereof. The model posits that resistance stems from negative user assessments of the fairness of the exchange between their inputs and the outcomes of their interaction with an IT. Focusing on covert resistance, Marakas and Hornik (1996) propose that passive resistance misuse is a response to threats that individuals associate with the introduction of a new IT in their world. They suggest that individuals who are unwilling or unable to adjust may enact overt cooperation and acceptance of the proposed system combined with covert resistance and likely sabotage of the
implementation effort (p. 208). Kim and Kankanhalli (2009) propose a model in which the perceived value of a system has a negative effect on user resistance. Central to the model are switching costs, defined as the disutility associated with switching to a new system, which negatively impact perceived value, thereby indirectly increasing user resistance. Initial conditions refer to characteristics of the environment that interact with the object of resistance and influence the assessment that users make of the situation. For instance, Martinko et al. (1996) propose an attributional explanation of individual resistance to IT. Their model suggests that a new IT interacts with environmental influences and users prior success or failure at using IT to evoke causal attributions. These attributions influence expectations regarding outcomes in terms of future performance. Expectations will then drive individuals reactions to the IT. In a similar vein, Hirschheim and Newman (1988) emphasize the social and political nature of IT implementation, and state that resistance has multiple causes, such as innate conservatism, lack of involvement, organizational invalidity, and personal characteristics of the designers or users, that interact in a tangle of different threads (p. 400). These five elements are not isolated and some researchers propose explanations of the dynamics of resistance to IT. At the individual level, Martinko et al. (1996) posit that individuals make causal attributions for the success or failure of a new IS based on internal and external influences. These attributions lead to outcome and efficacy expectancies; negative expectancies will lead to user resistance which, in turn, will influence future attributions. Focusing on group resistance, Lapointe and Rivard (2005) posit that resistance behaviors follow perceived threats resulting from the interaction between an object and initial conditions. During an IT implementation, some triggers can modify or activate initial conditions; a modification of the object of resistance may ensue. Based on the interaction between this new object and the new set of initial conditions, different resistance behaviors may follow. Ferneley and Sobreperez (2006) suggest that resistance is often manifested in user workarounds. They propose a dynamic model with four antecedent conditions: enforced proceduralization, organizational and personnel issues, discipline, and non-engagement with the system. Any given condition may lead to resistance, which in turn may result in different kinds of workarounds.
IT Implementers Responses
While these elements are clearly important to understanding the nature and sources of resistance, it has been suggested that
899
key interventions made by implementers may influence how resistance evolves (Lapointe and Rivard 2005). Here, we adopt Gottschalks (1999) definition of IT implementers as those responsible for the introduction of the technology to prospective users (p. 208), to which we add: as well as those responsible for the successful use of the system implemented. Indeed, IT implementers, whether business and functional managers or IT professionals, have responsibilities for system implementation, including, for example, responsibility for implementing on time and within budget, but also for implementing a system that will deliver the intended benefits (Gottschalk 1999). A few researchers have studied the link between user resistance to IT implementation and implementers actions (Jiang et al. 2000; Lederer and Nath 1991; Teng et al. 1996). Their focus, however, was on actions taken to prevent resistance, such as analyzing contextual factors, creating and communicating a vision, determining the optimal pace of change, and providing training and emotional support. Our literature review did not yield any study focusing specifically on implementers responses once resistance has occurred, nor on the effects of these responses on user resistance. Some studies, however, allude to the idea that implementers responses play a role in the dynamics of resistance. For instance, identifying the causes of resistance and determining what can be done about them is considered one of the first responses that implementers should have to resistance once it has occurred (Markus 1983). Modifying the system, even if it means that implementers may have to lose the battle and sacrifice a pet system project in order to win the war (Markus 1983, p. 400), is deemed an appropriate response when the actual object of resistance is the system (Martinko et al. 1996), when it is burdensome (Ferneley and Sobreperez 2006), or when some of its features are defective (Lapointe and Rivard 2006). Other effective responses are to look beyond modifications to the system for solutions (Lapointe and Rivard 2006; Lauer and Rajagopalan 2002) and aim responses at the users themselves or at their work environment (Joshi 1991). This includes modifying users perceptions of the IT (Martinko et al. 1996) or of their own and others inputs/outcomes related to the IT (Joshi 1991).
was deemed appropriate. To meet our objectives, we needed a relatively large sample of rich cases. Operationally, conducting such a large number of original case studies is very resource-intensive and difficult to achieve. Consequently, and given the availability of high-quality cases on IT implementation, we adopted a case survey strategy (Larsson 1993). Case surveys integrate qualitative studies, transforming qualitative data into (semi-) quantitative data, using a coding scheme and expert judgments by multiple coders (Newig and Fritsch 2009, p. 4).
Case Selection
A case survey requires the systematic selection, coding, and analysis of written case studies (Larsson 1993). To ensure quality data, we selected cases from scholarly sources (Yin and Heald 1975), including refereed journals, books, proceedings, research reports, working papers, and doctoral theses. The initial sample included all of the cases that we could find in these sources with mentions of resistance to IT implementation; over 250 cases were identified. The selection process was refined along specific inclusion and exclusion criteria (Lucas 1974; Yin and Heald 1975) to ensure the quality of the material selected and allow for an in-depth analysis of each case and comparisons across cases. The inclusion criteria were (1) that the case reported an instance of IT implementation, (2) that it reported evidence of user resistance, and (3) that the narrative provided a rich description of the events. A case was excluded when (1) we could not identify episodes of resistance complete with the initial resistance behaviors, the subject of resistance, the object of resistance, the implementers identity, the implementers response, and ex post resistance behaviors, or (2) the case was not set in an organizational context. We selected 89 cases published between 1974 and 2010 (sources listed in Appendix A). The majority (51) were from articles published in scholarly journals, mostly in information systems, but also in other domains including management, medicine, and operations management. Of the remaining 38 cases, 3 were from IS conference proceedings, 24 were drawn from 7 unpublished Ph.D. theses, 10 from a published Ph.D. thesis, and 1 from a scholarly book. The cases are varied in terms of sectors (e.g., finance, health, education, and manufacturing) and types of users (e.g., clerks, nurses, officers, physicians, managers, financial analysts, and researchers). Two types of implementers were identified in the cases: business or functional managers, such as senior and middle managers, and IT professionals, including IT personnel and IT consultants.
Research Method
As this studys objectives are first to identify how implementers respond to user resistance behaviors and then to investigate the effects of these responses on user resistance, an in-depth perspective such as that offered by case studies
900
In this study, the unit of analysis is the episode of resistance, which starts when an initial resistance behavior is identified and ends when an ex post resistance behavior is observed following the implementers response. Overall, our case database of 89 cases includes a total of 137 episodes, as our sample includes longitudinal cases in which multiple episodes of resistance were identified and coded. In these cases, the implementers responses were enacted in a sequential manner, and were coded as such. In a very small number of episodes (3 out of 137), the case data revealed that there were two concomitant responses; in these instances, we strived to identify the most salient one.
the score applied to a given resistance behavior, we conducted a two-round Delphi study. Prior to this, we reviewed the IS literature and produced a list of 27 different resistance behaviors, which we enriched with additional behaviors identified in the cases. We distributed the list to 25 IS faculty members and Ph.D. candidates. The respondents, several of whom had prior experience as IT professionals, were asked to indicate the degree of severity of each behavior on a six-point scale. A broad consensus was reached for most of the items; however, responses to five items varied widely. The wording of these items was refined or the items split; they were then submitted for a second round in which a general consensus was reached. The final list of 46 resistance behaviors and their levels of severity is presented in Appendix B.
Coding
We used an initial coding scheme that included the following categories derived from the literature: level of initial resistance behavior, subject of resistance, object of resistance, implementers identity, implementers response, and level of ex post resistance behavior. The coding process was iterative, using both the initial coding scheme and open codes. Each of the two authors independently coded each episode of resistance with the list of codes and completed a coding report that included bibliographical information, the code value attributed to each identified component, and excerpts from the case to justify the chosen codes. In this first step, all coding disagreements were reconciled by adopting a consensus approach. Resolving discrepancies in this way is said to be a superior way to correct coding mistakes (Larsson 1993, p. 1521). We remained open to new codes and categories when appropriate (Miles and Huberman 1994). Once again, we relied on inter-coder agreement and consensus to validate the new codes. Four codes associated with the implementers response category emerged: inaction, acknowledgment, rectification, and dissuasion. Two new categories were created: the object of rectification, with three codes (system, system significance, and users), and the degree of congruence between the object of resistance and the object of rectification, with two codes (congruent and noncongruent). Together we reviewed the final coding of each episode and discussed any discrepancies until we had reached a consensus; this helped eliminate individual disparities and solve problems (Bullock and Tubbs 1987). The final coding reports were recorded in a repository database that was used in the data analysis. One category, the level of resistance, received special treatment due to its ordinal nature. To ensure the validity of
Data Analysis
Creating a Taxonomy of Implementers Responses Using attributes that allowed us to differentiate one category from another (Shaw et al. 2005), we classified each of the implementers responses into one of four categories: INACTION, ACKNOWLEDGMENT, RECTIFICATION, and DISSUASION. We further refined these categories into subcategories to better specify the nature of the responses, using simple and parsimonious rules to assign instances to each subcategory and ensure that the categories were mutually exclusive, were easily understood, and appeared natural (Gregor 2006, p. 619). Determining Patterns Using Set-Theoretic Analysis Given the nature of the data and the theory-building objective of our study, we adopted set-theoretic analysis as our analytic approach, as it allows the identification of commonalities in the relationships between causal conditions and an outcome across cases (Ragin 2008). Here, the term causal condition refers to an aspect of a case that is relevant in some way to the researchers account or explanation of some outcome (Ragin 2008, p. 18). Set-theoretic analysis focuses on set relations rather than on correlations. It is a means for reducing the complexity of data sets by using Boolean algebra (Cronqvist 2007). The approach is applicable to small-N situations (from as few as 2 cases up to 15 cases) as well as intermediate-N (up to 100 cases) and large-N research designs (Ragin 2008). Set-theoretic analysis is based on the assumption of causal complexity between conditions and an outcome (Ragin 2008). Under this assumption, most often it is a combination of conditions, rather than a unique condition, that generates the
901
outcome. Causal complexity comprises the notion of equifinality, under which different paths can lead to the same outcome. In addition, the causal effect of a condition is not uniform, as a given condition may sometimes act in favor of an outcome and at other times act against it, depending on the other conditions with which it is combined. Finally, causality is not assumed to be symmetrical: the presence and the absence of the outcome, respectively, may require different explanations (Berg-Schlosser et al. 2009, p. 9). Although the assumption of causal complexity posits that most often it is a combination of conditions, rather than a single condition, that causes a given outcome, the use of settheoretic analysis to evaluate monocausal arguments is deemed appropriate. Indeed, set-theoretic analysis has been said to be relevant to any social scientific argument or statement that is formulated in terms of sets (Ragin 2006) when the aim is to identify the common causal conditions whether constituted by a single causal factor or combinations of causal factors across cases (Greckhamer et al. 2008, p. 697). Because in our data set a given episode of resistance corresponds to a single response, we first evaluated implementers responses as monocausal arguments. Set-theoretic analysis is concerned with whether a given condition, or a combination of conditions, is necessary and/or sufficient for an outcome to occur. A condition is necessary when it is always present when the outcome occurs, and it is sufficient when the outcome always occurs when the condition is present (Rihoux and Ragin 2009). Two strategies can be used to search for commonalities across cases (Ragin 2008). One is to examine cases that share a given outcome and attempt to identify their shared causal conditions. The other is to examine cases that share a causal condition, or a combination of conditions, and assess whether they display the same outcome. Given that the objective of our analysis was to determine the effect (outcome) of each category of implementers responses to user resistance (causal condition), we adopted the second strategy. Set-theoretic analysis is said to offer not only a method for analyzing causal relationships, but also a language for expressing such relationships (Fiss 2007, p. 1190), thereby providing new insights and allowing theory development (Berg-Schlosser et al. 2009). In theory development, a settheoretic approach involves an iterative analysis, wherein the researcher conducts an elaborate dialogue of ideas and evidence that leads to a progressive refinement of understanding of the relevant cases and to a more nuanced elaboration of the relevant causal conditions (Ragin 1999, p. 1232). During this dialogue, theoretical knowledge (knowledge of extant literature) and knowledge of the empirical field embedded in
the cases help the researcher interpret the analysis results, identify candidate causal conditions, and generate theoretical arguments. We adopted such an iterative approach. We first analyzed the data to determine whether, as a monocausal argument, a response category was sufficient to produce a given outcome (i.e., an increase or decrease in resistance). When a response category was not a sufficient condition, we drew on qualitative content analysis of the cases to identify additional candidate conditions that, combined with the implementers response, would form a configuration of conditions sufficient for a given outcome. Through careful coding and examination of the data, content analysis allows inferences to be made from data to their context in order to provide new knowledge and insights (Hsieh and Shannon 2005). This helped us further capitalize on the richness of the case survey material (Reuver et al. 2009). Here again, we relied on consensus to resolve any inter-coder discrepancies. Based on the results of this content analysis, we conducted a new round of settheoretic analysis, which was again followed by qualitative content analysis. The latter involved reexamining the narrative of each episode of resistance and the patterns of influence that emerged in the set-theoretic analysis to validate and enrich our observations. Finally, we enfolded extant literature. Our theoretical explanation emerged as a result of this iterative process.
Results
A Taxonomy of Implementers Responses
In response to our first research question, which sought to identify implementers responses to user resistance, we propose a taxonomy that comprises four categories: INACTION, ACKNOWLEDGMENT, RECTIFICATION, and DISSUASION (see Table 1). INACTION indicates a lack of action. Fully 38 percent (52) of the implementers responses fell into this category. A close examination of the instances of INACTION led us to derive three subcategories. The first, UNAWARENESS, refers to situations in which the implementers were not conscious that the users were exhibiting resistance. In these episodes, not only was there no explicit mention of the actual responses of the implementers, there were clear indications that the project continued as before, including doing nothing (Gunawardane 1985), continuing the implementation (Joshi 2005), and fumbling through the implementation (Hennen 2002).
902
Table 1. A Taxonomy of Implementers Responses

Response Category N (%) Response Subcategory
UNAWARENESS INACTION
N (%) 18 (13.1%) 27 (19.7%) 7 (5.1%) Doing nothing
Examples
52 (38.0%)
DELIBERATE IGNORANCE IMPOTENCE
Not caring, deciding to wait and see, adopting a laissez-faire approach Feeling unable to do anything Discussing issues; administering a questionnaire; organizing round tables, a task force, or focus group Redesigning the system (when the object of resistance is system features), training (when the object of resistance is system features and users skills are deficient), changing the work schedule or making concessions (when the object of resistance is system significance) Fixing the system (when the object of resistance is system significance), providing explanations as to how to use the system or adding personnel (when the object of resistance is system features and when the system is flawed) Forcing use of the system using coercive power or threatening users Reprimanding users or mandating use Reassurance, top management support, explanations, benefit rationalization
ACKNOWLEDGMENT
5 (3.6%)
CONGRUENT
38 (27.7%)
RECTIFICATION
49 (35.8%)
NONCONGRUENT
11 (8.0%)
COERCION DISSUASION
6 (4.4%) 8 (5.8%) 17 (12.4%)
31 (22.6%)
AUTHORITATIVE PERSUASION SUPPORTIVE PERSUASION
The second subcategory, DELIBERATE IGNORANCE, corresponds to situations where the implementers knowingly chose to disregard resistance. For some, managing resistance was not their responsibility, and others simply refused to take any corrective action. This category includes responsesor a lack thereofsuch as ignoring complaints (Van Akkeren and Rowlands 2007), not caring (Silva and Backhouse 2003), deciding to wait and see (Roy 1999), and adopting a laissezfaire approach (Par 1995). The third subcategory, IMPOTENCE, refers to situations where the implementers did not seem to have the competencies or resources to adequately respond to user resistance. In some episodes, the implementers seemed to feel powerless; in others, they appeared concerned by the prospect of investing additional resources: feeling impotent to rectify the situation (Rowe 1985), feeling confused (Hennen 2002), and feeling that there is too much water under the bridge to go back (Wagner and Newell 2007, p. 515).
ACKNOWLEDGMENT refers to responses that are limited solely to the recognition of user resistance. The implementers reacted with this response in five episodes (3.6 percent), including discussing issues (Wagner and Newell 2007), administering a questionnaire (Gunawardane 1985), creating a task force (Markus 1983), or forming focus groups (Bondarouk 2004). Although this response sometimes reflected a genuine effort to develop a better understanding of the issues underlying user resistance, it remained rather superficial. RECTIFICATION designates corrective responses intended to correct the system, the environment, or the users themselves. The implementers enacted this type of response in 49 episodes (35.8 percent). Examples of RECTIFICATION included redesigning the system (Wagner and Newell 2007), developing a new system (Alter 1980), fixing inadequacies (Aarts et al. 2004), providing explanations as to how to use the system (Bondarouk 2004), training (Boudreau 2000), turning off the
903
old system (Newman and Robey 1992), changing the work schedule (Miller 1983), and adding personnel (Lapointe 1999). A close examination of the data suggested the creation of two subcategories within RECTIFICATION, depending on the degree of congruence between the object of resistance and the object of the rectification. Three objects of rectification were identified: system, system significance, and users. When the object of resistance was some system feature and this feature was rectified (e.g., when an unfriendly user interface was improved), RECTIFICATION was deemed CONGRUENT. Similarly, when the object of resistance was system significance and the implementers responses were targeted at significance (e.g., improving work conditions when users felt that their work environment had deteriorated), RECTIFICATION was considered CONGRUENT. In some situations, resistance resulted from users feeling that they lacked the skills necessary for efficient use; the implementers responded with additional training. In such situations, RECTIFICATION was also deemed CONGRUENT. All other instances of RECTIFICATION were deemed NONCONGRUENT. DISSUASION refers to efforts meant to divert users from resisting. Fully 31 responses (22.6 percent) fell into this category. We further refined DISSUASION into COERCION and PERSUASION to reflect the valence of the dissuasive response. The coercive responses were aimed at forcing users to stop resisting and included a threateither implicit or explicitto the users if they continued resisting. This included forcing use of the system (Miller 1983), using coercive power (Allen et al. 2000), and threatening users (Lapointe 1999). Persuasive responses were aimed at convincing users that the new system was appropriate. In some episodes, PERSUASION was SUPPORTIVE and included reassurance (Joshi 2005), explanations (Hennen 2002), and benefit rationalization (Par 1995). In other episodes, PERSUASION was AUTHORITATIVE and involved reprimanding users (Alter 1980) or mandating use (Carasik and Grantham 1988). Although authority was exercised in these responses, no threats were involved.
Descriptive Statistics Table 2 presents the joint frequency distribution of the 137 episodes of resistance according to each of the four categories and the associated change in the level of resistance. As shown in the table, none of the episodes that involved INACTION as an implementers response was followed by a decrease in the level of resistance. Indeed, fully 90.4 percent of these 52 episodes were followed by increased resistance, the remaining 9.6 percent being followed by no change in the resistance level. In all of the five episodes where the implementers response was ACKNOWLEDGMENT, this response was followed by an increase in the resistance level. In 63 percent of the 49 episodes where the implementers response was RECTIFICATION, the result was a decreased level of resistance, and in 32.7 percent of the episodes this was followed by increased resistance. Finally, when the implementers response was DISSUASION, resistance decreased in 64.5 percent of the 31 episodes and increased in 35.5 percent of these episodes. A Chi-square test found that these results are not due to randomness, suggesting that further analyses are warranted. We conducted a Mann-Whitney test to determine whether the identity of the implementers (i.e., business or functional managers or IT professionals) was related to the type of response provided. No significant difference across the two groups was found except for ACKNOWLEDGMENT, where the five episodes involved IT professionals only. Because our sample includes cases from 1974 to 2010, one could argue that implementers responses in 2010 might have been different from what they were 40 years earlier. To control for this effect, we compared the proportion of each type of response for three periods (19741990; 19902000; 2000 2010), but found no statistically significant difference. Set-Theoretic Analysis In its crisp-set version, set-theoretic analysis is based on Boolean logic. To conduct the analysis, the values of the conditions and the outcome of interest are dichotomized into values of 0 and 1, with 1 indicating that a given condition/ outcome is present and 0 indicating its absence. In our study, each episode of resistance represents a case in the vocabulary of set-theoretic analysis. We recoded each episode according to whether it belonged to a given set among the nine implementers response subcategories and the two outcomes of interest. This resulted in a 137 11 dichotomized data table (Rihoux and De Meur 2009), which was then synthe-
Determining Patterns
To answer our second research question on the effects of implementers responses on level of user resistance, we analyzed the 137 episodes of resistance. We first examined the descriptive statistics and then conducted set-theoretic analyses.
904
Table 2. Implementers Responses and Ex Post Levels of Resistance

Ex Post Levels of Resistance Action
INACTION ACKNOWLEDGMENT RECTIFICATION DISSUASION
N (%) 52 (38.0%) 5 (3.6%) 49 (35.8%) 31 (22.6%) 137 (100%)
Higher (%) 47 (90.4%) 5 (100% 16 (32.6%) 11 (35.5%) 79 (57.7%)
Lower (%) 0 (0%) 0 (0%) 31 (63.3%) 20 (64.5%) 51 (37.2%)
No Change (%) 5 (9.6%) 0 (0%) 2 (4.1%) 0 (0%) 7 (5.1%)
Total 52 (100%) 5 (100%) 49 (100%) 31 (100%) = 59.52 6df p < 0.000
sized in a truth table (Appendix C) that lists the different logically possible combinations of causal conditions along with the cases conforming to each combination (Ragin 1999, p. 230). Following our strategy of determining whether each response subcategory was a sufficient condition for either outcome, we assessed the extent to which the episodes that shared a given causal conditionimplementers responsedisplayed the same outcome. To this end, we used the truth table to construct subsets of episodes, as presented in Table 3. We then computed the two descriptive measures that help assess the strength of the empirical support for theoretical arguments describing set relations (Ragin 2006, p. 292). The first measure, set-theoretic consistency, is the proportion of episodes with a given cause or combination of causes that also display the outcome of interest. For example, for UNAWARENESS (and as reported in Table 3), 16 of the 18 episodes that were members of the UNAWARENESS subset were also members of the RESISTANCE INCREASE subset, for a consistency score of 0.889. Consistency scores should be high, with a minimum threshold of 0.75: With observed consistency scores below 0.75, it becomes increasingly difficult on substantive grounds to maintain that a subset relation exists, even a very rough one (Ragin 2006, p. 292). The second measure, set-theoretic coverage, is the proportion of memberships in the outcome explained by a given condition or combination of conditions. Assessing the value of coverage is relevant only when consistency is above the 0.75 threshold; indeed, it is pointless to compute the coverage of a cause or combination of causes that is not a consistent subset of the outcome (Ragin 2006, p. 299). The value of coverage is presented in Table 3 and can be derived from Tables 2 and 3. For instance, the 25 episodes that have memberships in both the DELIBERATE IGNORANCE and RESISTANCE INCREASE sets represent 31.6 percent of the 79 episodes with that outcome. There is no threshold for
coverage. If a condition or a combination of conditions explains only a very small proportion of memberships in the outcome, it is said that the researcher might deem this condition or combination less significant empirically (Ragin 2006). It is argued, however, that empirical significance is not equivalent to theoretical significance, and that some paths with a high coverage can be theoretically uninteresting or even trivial (Wagemann and Schneider 2007, p. 20). Because the objective of our study is to uncover the effects of a given condition or combination of conditions, we consider a result theoretically relevant when it stems from a condition or a combination of conditions that is associated with a given outcome at a high level of consistency. Evaluating Monocausal Arguments The results of the first round of analysis (presented in Table 3) are clear regarding the role of INACTION, ACKNOWLEDGMENT, and RECTIFICATION, as UNAWARENESS, DELIBERATE IGNORANCE, IMPOTENCE, ACKNOWLEDGMENT, and NONCONGRUENT RECTIFICATION are sufficient monocausal conditions for RESISTANCE INCREASE, and CONGRUENT RECTIFICATION is a sufficient monocausal condition for RESISTANCE DECREASE. We now revert to the content analysis of each episode with memberships in one of these subcategoies to inform the results of the set-theoretic analysis. Table 4 illustrates all three subcategories of INACTION responses and shows how they are all associated with an increased level of resistance. For the ACKNOWLEDGMENT responses (see Table 5), content analysis further reveals that although this response recognized resistance, it rarely involved a search for causes and never involved the identification of solutions. For RECTIFICATION, our qualitative content analysis shows the importance of congruence between the object of resistance and the object of rectification. Table 6 shows how congru-
905
906
Table 4. INACTION
Initial Resistance Level/Behavior UNAWARENESS Level 3 / Representatives from the Athens Plant generally hung back in these meetings (p. 206). Implementers Response Despite the fact that Athens representative hung back, the Capital City production control manager assumed project management responsibility. Working closely with a systems person from Capital City, he developed the project teams proposal (p. 206). The CEO shared the vision but didn't actively help us (p. 45). Ex Post Resistance Level/Behavior Level 4 / Athens continued to use its old WIP, claiming problems in the new system(p. 206). Source Markus (1981)
DELIBERATE
Level 2 / Business managers began pulling back, doubting that the technology would really work (p. 45). Level 2 / At installation, however, the wider ABC community refused to work with the financial management module (p. 515).
Level 4 / Business managers had gone back to using their original IS applications to manage their processes the way they had prior to BATON (p. 46). Level 4 / During the post implementation phase faculty formed a coalition and demanded a series of meetings with the provost (p. 515).
Wang and Paper (2005)
Why has central management become invested in changing, for an extremely large revenue stream, from a tried-and-true system, proven to be both efficient and well-accepted by the faculty, to one that has not been tested with these types of accounts, and is less efficient and useful for the faculty? My guess is that there is too much water under the bridge to go back (p. 515).
Wagner and Newell (2007)
Table 5. ACKNOWLEDGMENT
Initial Resistance Level/Behavior Level 3 / The sales team basically stuck their tongue out and said screw you; we aint going to do this unless you do it our way (p. 513). Level 3 / Everyone was calling for standards, but no one standard was acceptable (p. 98). Subject of Resistance Sales team Implementers Response In this period they spent time discussing these issues (p. 513). A series of roundtable meetings were held in to discuss the common problems and issues(p. 98). Ex Post Resistance Level/Behavior Level 5 / Both the marketing and sales modules were stalled (p. 513). Level 5 / A feud was brewing (p. 99). Source Wagner and Newell (2007)
IMPOTENCE
Division managers
DeSanctis et al. (1996)
ent rectification is associated with RESISTANCE DECREASE, whether the object of resistance is the system or system significance and the object of rectification is the system, system significance, or the users themselves. Conversely, Table 7 shows how, when NONCONGRUENT RECTIFICATION is the implementers response, an increase in the level of user resistance is linked to targeting the wrong object of resistance.
Evaluating Combinatorial Arguments The results of the set-theoretic analysis were not as straightforward for the DISSUASION responses, as two of the three responses within this category do not appear to be sufficient
monocausal conditions for either RESISTANCE INCREASE or Indeed, as shown in Table 3 and Appendix C, the outcomes of the responses were contradictory. Following our iterative analysis strategy, we focused on this subset of episodes during a second round of set-theoretic analysis, preceded and followed by qualitative content analysis. Because AUTHORITATIVE PERSUASION is a subcategory of DISSUASION, we included it in our second round, even though it was deemed a sufficient condition for RESISTANCE DECREASE.
RESISTANCE DECREASE.
We first went back to the cases that displayed episodes with dissuasive responses to perform content analysis. Our main objective was to identify additional candidate conditions that,
907
Table 6. CONGRUENT RECTIFICATION

Initial Resistance Level/Behavior Level 4 / The lawyers who were its potential usersrejected it (p. 275). Object of Resistance System: The computer system became more of a hindrance than an aid (p. 275). Object of Rectification/ Implementers Response System: The entire operation was overhauleda new computer system was built on two fundamental principles: simplicity and the integrity of the data base (p. 275). System significance: The administration created administrative assistant positions, which would operate the system and be funded with overhead costs (p. 312). Resistance Level/Behavior No resistance / The changes in the processing methods and in the computer system proved to be quite successful. In addition to improving morale, it led toa dramatic productivity increase (p. 276). No resistance / The researchers were pleased with the presence of the administrative assistant (p. 312). Source Alter (1980)
Level 4 / As frustration grew, the researchers told the administration that if the problems with the system were not going to be solved, they would go again to AHS (pp. 311-312).
System significance: Research project managers complained about how the administrative tasks were taking away time that could have been better spent on tasks directly related to the scientific nature of their projects(p. 311).
Silva and Backhouse (2003)
Table 7. NONCONGRUENT RECTIFICATION

Resistance Level/Behavior Level 2 / Despite the features, CONFIG was not widely used by sales reps.By 1989 [three years after implementation] fewer than 10 percent of reps continued employing it (M&K, p. 12). Object of Resistance System significance: First, sales reps were not motivated to do what the system enabled them to do. Second, using the system made it harder for them to do what they were motivated to do (M&K, p. 13). System significance: caused a lot of stress and anxiety among the staff nurses have feared legal troubles (p. 341). Object of Rectification / Rectification System: Althoughthese platform, interface, and training problems were not the root cause for not using CONFIG, they were problems that developers thought they could solve. So, in 1988, the developers initiated a major effort to deploy a more usable version of CONFIG (M&K, p. 13). System: small free-text areas were added at the end of each body system and even more space was provided for extensive comments at the end of the assessment (p. 340). Resistance Level/Behavior Abandonment / Despite all efforts to enhance the system, the level of usage remained disappointingly low until CONFIG was unplugged at the end of 1992 and all future development terminated (L&R, p. 90). Level 4 / Over time, ICU nurses have found and adopted a variety of ways to circumvent or bypass the system and hence reduce their level of stress (p. 341). Source Markus and Keil (1994); Lyytinen and Robey (1999)
Level 3 / Indeed, staff nurses complained about missing items in the menus. A lot of initial frustration was also expressed by nurses(p. 339).
Par (1995)
combined with the implementers response, would form a sufficient configuration of conditions for either an increase or decrease in resistance. A number of candidate conditions emerged. For instance, we observed that in some cases, system use was mandatory, yet in others it was voluntary. Also, in some cases, the implementers were from the business or functional side of the organization (top or middle managers, for instance), while in other cases they were IT professionals. This analysis revealed that some conditions, for example the mandatory/voluntary nature of system use, did not exhibit much variety (that is, 29 out of 32 episodes were instances of
mandatory use) and thus were not candidate conditions for resolving the contradictions. On the other hand, the identity of the implementers, the credibility of either the source or the message, and the power base of the implementers appeared potentially significant in resolving the contradictions. We initially focused on PERSUASION and the ensuing resistance level. We observed that in the case of AUTHORITATIVE PERSUASION, discourses that were credible (illustrated in Table 8) and were enacted by implementers whose power bases were greater than that of the users, were generally associated with a decrease in resistance.
908
Table 8. AUTHORITATIVE PERSUASION

Subject of Resistance/ Implementers Identity IT professionals/ Management
Initial Resistance Level/Behavior Level 2 / [Some] simply avoided it (p. 62).
Implementers Response Participants were urged to please log on The Communicator and run postman at least 2 or 3 times per day (p. 62). Management decided to tell that manager that they really did expect cases to be done though CAUSE insofar as possible (p. 223).
Credibility Management urging was the motivation for continued use (p. 63). Conformed with the directive from top management (p. 223).
Ex-post Resistance Level/Behavior No resistance / continued use (p. 63).
Source Carasik and Grantham (1988)
Level 1 / The manager of one of the divisions handling large cases was indifferent to CAUSE (p. 223).
Manager/ Top management
No resistance / That manager never became enthusiastic about CAUSE but conformed (p. 223).
Alter (1980)
Table 9. SUPPORTIVE PERSUASION (Decrease in the Level of Resistance)

Subject of Resistance/ Implementers Identity Supervisors and loan officers/ Operation heads and senior management
Initial Resistance Level/Behavior Level 3 / During this period, there was growing discontent especially among older supervisors and some loan officers (pp. 137-138).
Implementers Response Eventually, we had a campaign to explain. The process began with a presentation by senior management which explained the business strategy behind IPS and, in addition, provided a public statement of commitmentinformal meetings with small groups of employees to explain details of IPS, dispel rumors and discuss concerns (p. 138).
Credibility Informal meetings with small groups of employees to explain details of IPS, dispel rumors and discuss concerns (p.138).
Ex Post Resistance Level/Behavior No resistance / The education process was extremely effective, and led to a helping relationship across the whole organization. Old students became new teachers(p. 138).
Source Keen et al. (1982)
Level 2 / Their perception of an unfair balance of power (p. 30)
Repair agents/ IT consultants
Overcoming any suspicions that The persuasive approach was the repair agents had of the extremely system (p. 30). effective (p. 30).
No resistance / The success of this approach is demonstrated by the fact that over 1000 repair agents adopted the system within 18 months (p. 30).
Allen et al. (2000)
In the case of SUPPORTIVE PERSUASION, the notion of credibility seems to be paramount. As shown in Table 9, when the source was deemed credible and/or the arguments included a demonstration of the importance of the project or explained a rationale for the implementation of the IT, the level of resistance decreased. In contrast and as illustrated in Table 10, when the persuasive arguments were considered to be emanating from an untrustworthy source, were unclear, created confusion, or relied solely on reassurance, the imple-
menters arguments went unheeded and user resistance increased. Finally, we examined the six episodes that involved COERCION as the implementers response. In all of these instances, the power base of the implementers was formally greater than that of the resisters, and the implementers were all business or functional managers. As illustrated in Tables 11 and 12, when COERCION was associated with a threateither explicit
909
Table 10. SUPPORTIVE PERSUASION (Increase in the Level of User Resistance)

Subject of Resistance/ Implementers Identity Nurses/ Project team (project nurse)
Initial Resistance Level/behavior Level 2 / As far as many nurses on the ward were concerned, the training they had received was inadequate (p. 241).
Implementers Response Benefits sessions were also carried out by the Project Nurse, to up-date the information users [had] on the so-called benefits of the system (p. 241). Management reassurance that it would, once implemented, actually save time (p. 12).
Credibility The nurses had the impression to have been misled by the team (p. 242). Management reassurance went unheeded (p. 12).
Ex Post Resistance Level/behavior Level 3 / became disillusioned with its poor performance(p. 242).
Source Wilson and Howcroft (2002)
Level 3 / They (salespeople) saw it as adding time to their already busy schedules, and viewed it as unnecessary busy work (p. 12).
Sales people/ Management
Level 4 / They expressed their displeasure in meetings and were reluctant to provide the needed information for the new system. They also resisted using the system (p. 12).
(Joshi 2005)
or implicitthat was not deemed believable, the result was an increased level of user resistance. Conversely, in the episodes where the message included threats that were clearly plausible, the users ultimately complied with the demands of the implementers. Following this, we proceeded to perform a second round of set-theoretic analyses. We first combined each of the candidate conditions (implementers identity, power, and credibility) one by one with the implementers responses to determine whether this would resolve the contradictions. The resulting truth table is presented in Appendix D; Table 13 presents the corresponding subsets along with their consistency and coverage scores. Because implementers identity did not help resolve contradictions, we have not included it in Appendix D and Table 13. As can be seen in Appendix D and Table 13, with POWER and CREDIBILITY added, the data no longer contain contradictions. To reduce the complexity of the data sets that need to be interpreted, one additional step in set-theoretic analysis is to perform logical minimization (Rihoux and De Meur 2009). But before doing so, the researcher must ensure that the truth table no longer contains contradictions and that there is sufficient variety in the data. Although our data satisfy the first condition, we had to address the issue of limited diversity within the COERCION subset. Limited diversity results when there are logical remainders (i.e., when some logically possible combinations remain without empirical cases). As shown in Appendix D and Table 13, our data set did not
include a single episode in which implementers with less power than users enacted COERCION as a response. For this reason, we did not include the COERCION subsets in the minimization. For the other two subcategoies of responses, we used the principles of Boolean algebra to minimize the expressions that reflected the set relationships within each subcategory. We took two routes to minimize the data: we conducted the minimization by hand and we used specialized software, fsQCA2 2.0. Although minimization with specialized software is less prone to error, logical minimization by hand, feasible here because the number of Boolean expressions was relatively low (Wagemann and Schneider 2007), kept us closer to the data and more engaged in the interpretation of the results. The result of the minimization was similar for the two subsets containing persuasive responses. Indeed, all of the combinations of conditions found to be sufficient for RESISTANCE INCREASE included ~CREDIBILITY (i.e., an absence of CREDIBILITY), while all of the combinations of conditions that were found to be sufficient for RESISTANCE DECREASE included CREDIBILITY. Moreover, POWER was found to be a superfluous condition. Table 14 presents an overall summary of our results.
1999-2003, Charles Ragin and Kriss Drass; 2004-2008, Charles Ragin and Sean Dav. We used the truth table algorithm to conduct the minimization.
910
Table 11. COERCION (Increased Level of User Resistance)

Subject of Resistance/ Implementers Identity Repair Agents / TransLease
Initial Resistance Level/Behavior Level 2 / A dominant theme for repair agent complaints was their perception of an 'unfair' balance of power, which meant that they felt that lease companies would tie them into a system that would reinforce and amplify existing power structures (p. 30). Level 4 / Most of the physicians supported the demand for system withdrawal, and some even handed in their resignations (p. 107).
Implementers Response Next time a repairer tries to use the old process, tell them that they can't (p. 30).
Credibility They [TransLease] just use the rules to control us [repair agents] even more (p. 30).
Ex Post Resistance Level/behavior Level 5 / [This] degraded the relationship between stakeholders. It also increased the distrust the repair agents had of TransLease (p. 30). Level 5 / It means that they were then obliged to withdraw their Phase 2 because it just wasnt working anymore (p. 109).
Source Allen et al. (2000)
Physicians/ General management
The administration, supported by the hospitals Board of Directors, wanted to take hospitalization rights away from six physicians considered particularly stubborn (p. 107).
So the physicians, many physicians, said: If you want to laugh at us, then were out of here (p. 107).
Lapointe (1999)
Table 12. COERCION (Decreased Level of User Resistance)

Subject of Resistance/ Implementers Identity Nurses/ Hospital administrators
Initial Resistance Level/behavior Level 3 / The nurses were markedly more disillusioned.They had more problems with the system (p. 197).
Implementers Response The final word became: If you arent happy, go work elsewhere (p. 197).
Credibility It was clear that, for the hospitals GM and Dr. XX [physician responsible for the implementation], the professional services office and the Director of Nursing, it has to work this way, it will work this way and there would be no compromise(p. 231).
Resulting Resistance Level/behavior No resistance / No one refused to use the system. There was no choice. The nurses and all the other staff had been told that they didnt have a choice; that was it (p. 266). No resistance / Mr. Gall gathered together his administrative and supervisory staff and instructed them to make preparations to receive CODES personnel (p. 8).
Source Lapointe (1999)
Level 4 / Mr. Gall, after hearing the presentation of pros and cons of the system, made a seemingly unilateral decision to do everything in his power to put a stop to this project (p. 7).
Agency/ State administration
The state issued a Mr. Gall, seeing the initial battle lost (p. 8). regulation in 1977 mandating the conversion of all 21 counties to the CODES (the state system) (p. 8).
Miller (1983)
911
912
Table 14. Synthesis of Set-Theoretic Analyses

Action
UNAWARENESS DELIBERATE IGNORANCE IMPOTENCE ACKNOWLEDGMENT CONGRUENT RECTIFICATION NONCONGRUENT RECTIFICATION COERCION*POWER*CREDIBILITY COERCION*POWER*~CREDIBILITY AUTHORITATIVE PERSUASION*CREDIBILITY AUTHORITATIVE PERSUASION*~CREDIBILITY SUPPORTIVE PERSUASION*CREDIBILITY SUPPORTIVE PERSUASION*~CREDIBILITY
Sufficient for Outcome Increase Increase Increase Increase Decrease Increase Decrease Increase Decrease Increase Decrease Increase
An Explanation of the Effects of the Implementers Responses to User Resistance Behaviors

We propose an explanation of the effects of the IT implementers responses to user resistance behaviors. Our explanation is anchored in our set-theoretic analyses and enriched by the content analysis of our cases. It is also informed by the IS, management, strategy, and psychology literature, which we revisited to refine and expand our understanding of the patterns we observed in our data. Our explanation builds on extant models of resistance to IT, as we rely on the core elements of resistance common to these models: initial conditions, object of resistance, perceived threats, subject of resistance, and resistance behaviors. In particular, we borrow from Lapointe and Rivard (2005), who conceptualize resistance to IT as behaviors that occur following perceptions of threats associated with interactions between an object and initial conditions. They also posit that, over the course of an implementation, some triggerswhich may include the implementers responsesinfluence the antecedents and subsequently modify the resistance behaviors exhibited. Following this, our theoretical explanation posits that the implementers response will modify the object of resistance, the initial conditions, or the perception that users have of the threats resulting from the interaction between initial conditions and the object of resistance. We also suggest that users engage in an ongoing process in which they appraise their situation, and that changes in their resistance level occur following such appraisals. We will now explain the effect of
each type of modification brought about by the implementers responses.
Responses that Alter the Object or Initial Conditions

By definition, RECTIFICATION is an intended alteration of either the object of resistance or the initial conditions. We posit that INACTION and ACKNOWLEDGMENT also modify the initial conditions, even though this was not the implementers intention. The following discussion explains each category in turn. RECTIFICATION When they engage in RECTIFICATION, implementers explicit aim is to modify either the object of resistance or the initial conditions. The IS literature acknowledges some rectification-type actions in response to resistance behaviors. For instance, a handful of authors have mentioned that rectifying the systemredesigning or restructuring the system (Martinko et al. 1996), rebuilding the system and even reverting to the old system (Lauer and Rajagopalan 2002)may at times be an effective response to user resistance. Others have suggested altering the user/system environment, including hiring temporary staff, offering rewards, or introducing job reclassifications (Joshi 1991); these rectifications are aimed at altering the systems significance. Corrective responses aimed at the users have also been proposed, including training on the system to reduce learning effort and frustration (Joshi 1991) and imagery training to restore feelings of self-efficacy among users (Martinko et al. 1996).
913
Our taxonomy of implementers responses, however, highlights the role played by the degree of congruence between the object of resistance and the object of rectification. Indeed, CONGRUENT RECTIFICATION was found to be a sufficient condition for a decrease in resistance and NONCONGRUENT RECTIFICATION was found a sufficient condition for an increase in resistance. This implies that effective rectification requires a proper identification of the causes of resistance, suggesting that, notwithstanding the intrinsic value of a solution, what is critical is the accuracy of the diagnostic made of a given situation and that the solution indeed targets the source cause. The notion of Type III error (Mitroff 1998) is helpful in understanding this issue. A Type III error is said to exist when there is a mismatch between the problem and the solution. This does not mean that the solution is intrinsically wrong, but rather that it does not fit the problem at hand. In such situations, the solution will not solve the problem. Our explanation of the dynamics of the effect of rectification on the level of user resistance is as follows. After the RECTIFICATION response, users will appraise the interaction between the altered object and/or initial conditions. A congruent rectification will have altered the object and/or the conditions in such a way that their interaction will be perceived as less threatening and resistance will decrease. If the rectification was not congruent, the resulting object and/or the conditions will interact in such a way as to increase the perceived threat, hence leading to increased resistance. INACTION Our analyses reveal that INACTION on the part of the implementers is a sufficient condition for RESISTANCE INCREASE. Given the high percentage of implementers responses that fell into this category, it is somewhat surprising that this response has not been widely acknowledged in IS research. Only Lauer and Rajagopalan (2002) have hinted that implementers who fail to take action when confronted with user resistance risk long-term failure. The literature on strategy and management, however, sheds some light on the nature and effects of INACTION responses. First, UNAWARENESS, which refers to those instances where the implementers did not realize that the users were exhibiting resistance, was identified as a managerial response associated with problems that go unrecognized in an organization and that ultimately lead to damage (Shimizu and Hitt 2004). Second, DELIBERATE IGNORANCE, which includes instances where the implementers knowingly choose to ignore resistance, is considered akin to managerial hubris, an unrealistic, positive belief held by overconfident managers
that they will succeed no matter what (Duhaime and Schwenk 1985; Hayward and Hambrick 1997). Ultimately, managerial hubris is known to make the situation worse. Third, IMPOTENCE, which refers to instances where implementers believe that they do not have the competencies or the resources required to adequately respond to user resistance, has been linked to managers feelings of helplessness and powerlessness. Impotence can be associated with the sunk cost effect, which is a tendency by managers to escalate commitment to a course of action in which they have made considerable prior investments, despite consistently poor results (Roberto 2002). The sunk cost effect has also been instrumental in explaining escalation in software projects, albeit not in the context of user resistance (Keil 1995; Keil et al. 2000). We posit that even though implementers do not intend to modify the initial conditions when they do not react to user resistance behaviors, their very inaction indeed alters these conditions. One of the conditions that can be altered, for instance, is the degree of trust that users have in the implementers willingness or ability to address the issues they raise. Suppose that users engage in resistance behaviors because they perceive threats from the interaction between a system (object) and their level of computer efficacy (initial condition). They also trust the implementers ability and willingness to address the issue (initial condition). During a reappraisal of the situation, the implementers inaction will have altered (decreased) the users level of trust (initial condition). The users will perceive the interaction between the new set of values of the initial conditions and the object as more threatening, and the level of resistance will increase. A similar explanation can be proposed if, at the outset, the initial conditionlevel of user trust in the ability/willingness of the implementers to address the issues pinpointed by the resistance behaviorsis low. Indeed, inaction from the implementers will lower the level of trust even more, reinforcing distrust and intensifying the perceived threats, thereby leading to increased resistance. ACKNOWLEDGMENT The management literature has long recognized the importance of listening when resistance to change is detected (Lawrence 1954). Resistance has been compared to the symptoms of a bodily disease, leading to the suggestion that genuinely paying attention to resistance can help determine whether the problem is linked to technical imperfections or social issues, and hence pave the way to a solution. In the IS literature, a few authors have touched upon the importance of identifying and addressing the issues underlying resistance
914
(Lauer and Rajagopalan 2002; Markus 1983). The recommendations made by IS scholars often imply a need to unearth the causes of resistance in order to determine an appropriate corrective response, but this approach has never been empirically verified. Our analyses reveal that when implementers responded with ACKNOWLEDGMENT, which was found to be a sufficient condition for RESISTANCE INCREASE, they actually paid only limited attention to the causes of resistance. In management, it has been suggested that unless managers learn to truly listen to others, disputes will continue to flare up without notice (Maude 1971, p. 307). Also, in the context of change management, it is argued that when managers limit their response to offering lip servicewhen they hear rather than listen there can be no real responsiveness (Reynierse 1994). Much like our discussion of responses from the INACTION category, we conjecture that ACKNOWLEDGMENT alters the set of initial conditions, for instance, the trust that users have in the willingness/ability of the implementers to address the issues raised by their resistance behaviors. We posit that the dynamics and the effects will be similar to those of INACTION.
other hand, a lack of credible persuasive efforts helps reinforce the negative perceptions held by those resisting a message (Lydon et al. 1988; Tormala and Petty 2004). This research suggests that people tend to resist dissuasive efforts when they are forewarned of someones intent or when they do not believe that the message is trustworthy. It has also been suggested that the absence of credibility starts a process that ultimately will not only increase the level of resistance but even strengthen predictions of behavioral intentions to resist (Tormala and Petty 2002). Our explanation here is that a dissuasive response alters the perception of a threat that arises when users appraise the interaction between an object of resistance and initial conditions. When implementers revert to SUPPORTIVE PERSUASION, their message is often one of reassurance and benefits rationalization, and is aimed at convincing users that the situation is not actually threatening. We conjecture that messages of AUTHORITATIVE PERSUASION (e.g., reprimand or mandated use) are aimed at convincing users that their feeling of being threatened by the interaction between the object and the initial conditions is unwarranted. Finally, we conjecture that the goal behind COERCION (e.g., threatening users) is to position the perceived threats associated with the new IT as less threatening than the threats associated with not complying with the implementers message. The role of credibility is paramount in the following way. When users reappraise the situation, a credible dissuasion effort will have successfully altered their perception of the threats that result from the interaction between the object and the initial conditions. Indeed, even with an unaltered object and unmodified initial conditions, users will deem the interaction between the object and the initial conditions less threatening than they did before they heard the implementers message, due to the moderating effect of credibility. In the case of a supportive persuasion effort, the moderating effect of credibility will decrease the degree of perceived threat. For an authoritative persuasive effort, the moderating effect of credibility will make users conclude that there is no need to perceive a threat. The moderating effect of the credibility of a coercive message will diminish the threat associated with the interaction between the object and the initial conditions in comparison with the threat conveyed by the message. In all three cases, after users reappraise the situation, resistance will decrease. In contrast, when credibility is low, users will assess the situation as more threatening than before the implementers response, and this is likely to have a multiplicative effect on the level of perceived threats. Hence, resistance will increase.
Responses that Alter the Subjects Perception of Threats

We suggest that the responses within the DISSUASION category are aimed at modifying users perceptions of the threats originating in the object and the initial conditions. In our study, the credibility of the message was paramount in explaining the effect of a dissuasive response, as AUTHORITATIVE PERSUASION*CREDIBILITY and SUPPORTIVE PERSUASION* CREDIBILITY were sufficient conditions for RESISTANCE DECREASE, and AUTHORITATIVE PERSUASION*~CREDIBILITY and SUPPORTIVE PERSUASION*~CREDIBILITY were sufficient conditions for RESISTANCE INCREASE. Moreover, our data show that even when implementers have power, a noncredible coercive message was followed by increased resistance. Research in psychology shows that credibility plays an important role in persuasive efforts, as the judgment that the receiver makes of a messages credibility is a critical aspect of the messages persuasiveness (Petty and Cacioppo 1986). Message credibility depends on the credibility of both the source and the message itself, including its content, its plausibility, and its consistency (Wathen and Burkell 2002). A credible message is more likely to influence the receivers behavior and attitude than a non-credible message. On the
915
Conclusion
Our first research question was: What are implementers responses to user resistance to IT? We have answered this question with a taxonomy of implementers responses to user resistance, as identified in 137 episodes of resistance. This taxonomy comprises four categories: (1) INACTIONeither unawareness, deliberate ignorance, or impotence; (2) ACKNOWLEDGMENT; (3) RECTIFICATIONeither congruent or not congruent; and (4) DISSUASIONeither coercion, authoritative persuasion, or supportive persuasion. In answering our second research question (What are the effects of implementers responses on the intensity of resistance behaviors?), we developed a theoretical explanation of the patterns of relationships observed between implementers responses and user resistance. First, when implementers respond with INACTION, because they are simply ignorant of the situation, choose not to respond, or feel that they do not have the means to respond, resistance increases. When implementers actively respond, some of their actions are also linked to an increase in the level of user resistance. This is the case for ACKNOWLEDGMENT, NONCONGRUENT RECTIFICATION, and the instances where DISSUASION was not deemed credible. Second, some implementers reactions are associated with a decrease in user resistance. This is true when implementers respond with CONGRUENT RECTIFICATION or, if they can convey credibility, when they choose DISSUASION. We explain the influence of implementers responses on user resistance behaviors by their altering the object of resistance, the initial conditions, or the perception of threats. Our analysis suggests several avenues for future research. In this study, we adopted a crisp set approach to conduct settheoretic analyses. This implies that the level of resistance that followed an implementers response was assessed as being either higher or lower than the ex ante level. We opted for this approach in order to focus on the ultimate effect of implementers response, that is, on whether they contributed to increase or decrease resistance. Using our measure of resistance with six different levels of resistance, further research could adopt a fuzzy set approach to set-theoretic analysis (Ragin 2008). Because the fuzzy set approach requires researchers to establish meaningful thresholds for partial set membership, it could lead to finer-grained analyses of the data and more specific predictions. Another refinement would be to seek cases where several implementers responses follow the enactment of resistance behaviors and to determine the effect of such combinations of responses on the ensuing resistance behaviors. The combinatorial approach of set-theoretic analysis would be particularly useful here. Another avenue for future research would be to empirically
test our explanations of the dynamics of the influence of implementers responses on user resistance, either with longitudinal case studies or a survey. A fourth research avenue would be to examine the long-term effects of COERCION when it lowers user resistance. As reported here, when COERCION is exerted and is deemed credible by users, the result is a decrease in the level of resistance. It would be important to examine, on a longitudinal basis, whether this lowered resistance eventually morphs into resentment or covert resistance behaviors, which can become dysfunctional. Finally, investigation of the preferred choice of response in a given situation would be worthwhile. Indeed, while both DISSUASION (when the source and the message are deemed credible) and CONGRUENT RECTIFICATION are associated with reduced user resistance, it would be worthwhile to investigate the context under which each response is the better choice. Some limitations of our study must be acknowledged. The first limitation stems from the use of secondary data, meaning that the cases we selected had not been written with the specific purpose of studying implementers responses to user resistance behaviors and the ultimate effects of such responses. However, due to the care with which we selected and excludedcases, we are confident that the cases in our sample were indeed narrations of episodes of resistance, and that they provided sufficient data for appropriate coding and content analysis. A second limitation is that assigning numbers to qualitative data unduly simplifies the complex phenomena under investigation and may leave out some of the richness of case research (Larsson 1993, p. 1519). However, the use of a large number of cases (here, 89 cases with 137 episodes of resistance) compensates for such information loss (Larsson 1993). We also compensated for this weakness by using set-theoretic analysis, an approach that emphasizes the importance of in-depth analysis of case content. Our study makes a number of contributions. First, it fills a gap in the literature concerning implementers responses to user resistance. Although much has been written about user resistance to IT implementation, and despite the fact that some authors have made recommendations on how to prevent resistance from occurring, the literature is surprisingly mute on how implementers behave once user resistance has actually occurred. The taxonomy that we propose is based on a large number of episodes of resistance, taken from a wide variety of cases covering several decades of IT implementation studies. This taxonomy can serve as a basis for further explorations of implementers behaviors. The second contribution of our study is that it offers a theoretical explanation of the effects of implementers responses on user resistance behaviors. This explanation is anchored in our set-theoretic analysis of the 137 episodes of resistance that constituted our sample and
916
is enriched by our content analyses of the cases and the fact that we enfolded extant literature. Hence, our study not only helps predict the effect of implementers responses and user resistance, it also explains why those responses have the effect that they have. The third contribution of our study is methodological, in that we adopted a research methodthe case surveythat combines the benefits of analyzing a large quantity of data with those of conducting an in-depth analysis of qualitative data. Furthermore, we applied set-theoretic analysis to the data, an approach that allows for theory development and is seldom used in the IS field. Our research has significant implications for implementers. First, whenever implementers choose RECTIFICATION as their response, they should strive to identify the appropriate object of rectification, since in order to successfully reduce resistance, they must achieve congruence between the object of resistance and the object of rectification. Second, when implementers decide to use DISSUASION, they must ensure that they, and their message, are credible whenever they choose SUPPORTIVE PERSUASION, COERCION, or AUTHORITATIVE PERSUASION. Third, our research has revealed that implementers should avoid mere ACKNOWLEDGMENT of user resistance; rather, they should ensure that they truly listen to resistant users in order to identify the actual causes of their resistance behaviors before taking any action. Finally, we have observed that INACTION is one of the preferred responses of implementers. Since INACTION is systematically associated with an increase in resistance, we urge implementers to opt for more effective responses.
Acknowledgments
This research was supported by a grant from the Social Sciences and Humanities Research Council of Canada (SSHRCC). We gratefully acknowledge Thomas Allards assistance in the conduct of this study. We also thank Henri Barki and M. Lynne Markus for their insightful comments. We are most grateful to the senior editor, Ola Henfridsson, for his advice and strong support. We also thank the associate editor for the wise guidance provided on our work. A heartfelt thank you to the reviewers who have made this experience a true developmental review process.
References
Aarts, J., Doorewaard, H., and Berg, M. 2004. Understanding Implementation: The Case of a Computerized Physician Order Entry System in a Large Dutch University Medical Center, Journal of the American Medical Informatics Association (11:3), pp. 207-216.
Allen, D. K., Colligan, D., Finnie, A., and Kern, T. 2000. Trust, Power and Interorganizational Information Systems: The Case of the Electronic Trading Community Translease, Information Systems Journal (10:1), pp. 21-40. Alter, S. 1980. Decision Support Systems: Current Practice and Continuing Challenge, Reading, MA: Addison-Wesley Publishing Company. Ang, J., and Pavri, F. 1994. A Survey and Critique of the Impacts of Information Technology, International Journal of Information Management (14), pp. 122-133. Berg-Schlosser, D., De Meur, G., Rihoux, B., and Ragin, C. C. 2009. Qualitative Comparative Analysis (QCA) as an Approach, in Configurational Comparative Methods, B. Rihoux and C. C. Ragin (ed.), Thousand Oaks, CA: Sage Publications, pp. 1-18. Bondarouk, T. V. 2004. Implementation of a Personnel Management System Beaufort: Successes and Failures at a Dutch Hospital, Annals of Cases on Information Technology (6), pp. 352-369. Boudreau, M.-C. 2000. Exploring the Transition to Enterprise Resource Planning: A Longitudinal Study of IT-Related Change, unpublished doctoral dissertation, Georgia State University, Atlanta, GA. Bullock, R. J., and Tubbs, M. E. 1987. The Case Meta-Analysis for OD, in Research in Organizational Change and Development, R. W. Woodman and W. A. Pasmore (eds.), Greenwich, CT: JAI Press, pp.171-228. Carasik, R. P., and Grantham, C. E. 1988. A Case Study of CSCW in a Dispersed Organization, in Proceedings of the Conference on Human Factors and Computing Systems, Washington, DC, June 15-19, pp. 61-66. Cronqvist, L. 2007. Tosmana: Tools for Small-N Analysis, Version 1.3 BetaUser Manual, (http://www.tosmana.net/). Day, D. L. 2000. Behavioral Effects of Attitudes toward Constraint in Case: The Impact of Development Task and Project Phase, Information Systems Journal (10:2), pp. 151-163. DeSanctis, G., Jackson, B. M., Scott Poole, M., and Dickson, G. W. 1996. Infrastructure for Telework: Electronic Communication at Texaco, ACM SIGCPR/SIGMIS Conference on Computer Personnel Research, Denver, CO, Aprill 11-13, pp. 94-102. Dickson, G. W., Simmons, J. K., and Anderson, J. C. 1974. Behavioral Reactions to the Introduction of a Management Information System at the US Post Office: Some Empirical Observations, in Computers and Management in a Changing Society, D. H. Sanders (ed.), New York: McGraw Hill, pp. 410-421. Dos Santos, B., and Sussman, L. 2000. Improving the Return on IT Investment: The Productivity Paradox, International Journal of Information Management (20:6), pp. 429-440. Duhaime, I., and Schwenk, C. 1985. Conjectures on Cognitive Simplification Processes in Acquisition and Divestment Decision Making, Academy of Management Review (10:2), pp. 287-295. Ferneley, E. H., and Sobreperez, P. 2006. Resist, Comply or Workaround? An Examination of Different Facets of User Engagement with Information Systems, European Journal of Information Systems (15:4), pp. 345-356.
917
Fiss, P. C. 2007. A Set-Theoretical Approach to Organizational Configurations, Academy of Management Review (32:4), pp. 1180-1198. Gottschalk, P. 1999. Strategic Information Systems Planning the IT Strategy Implementation Matrix, European Journal of Information Systems (8:2), pp. 107-189. Greckhamer, T., Misangyi, V. F., Elms, H., and Lacey, R. 2008. Using Qualitative Comparative Analysis in Strategic Management Research: An Examination of Combinations of Industry, Corporate and Business-Unit Effects, Organizational Research Methods (11), pp. 695-726. Gregor, S. 2006. The Nature of Theory in Information Systems, MIS Quarterly (30:3), pp. 611-642. Gunawardane, G. 1985. Implementing a Management Information System in an Extremely Dynamic (and Somewhat Hostile) EnvironmentA Case Study, Interfaces (15:6), pp. 93-99. Hayward, M. L. A., and Hambrick, D. C. 1997. Explaining the Premium Paid for Large Acquisitions: Evidence of CEO Hubris, Administrative Science Quarterly (42:1), pp. 103-127. Hennen, J. M. 2002. Twilight or Rebirth? A Case Study of How New Technology Reconstructed Academic Counseling in a Community College, unpublished doctoral dissertation, University of St. Thomas, St. Paul, MN. Hirschheim, R., and Newman, M. 1988. Information Systems and User Resistance: Theory and Practice, The Computer Journal (31:5), pp. 398-408. Hsieh, H. F., and Shannon, S. 2005. Three Approaches to Qualitative Content Analysis, Qualitative Health Research (15), pp. 1277-1288. Jiang, J. J., Muhanna, W. A., and Klein, G. 2000. User Resistance and Strategies for Promoting Acceptance across System Types, Information & Management (37:1), pp. 25-36. Joshi, K. 1991. A Model of Users Perspective on Change: The Case of Information Systems Technology Implementation, MIS Quarterly (15:2), pp. 229-240. Joshi, K. 2005. Understanding User Resistance and Acceptance During the Implementation of an Order Management System: A Case Study Using the Equity Implementation Model, Journal of Information Technology Cases and Application Research (7:1), pp. 6-20. Keen, P. G. W. 1981. Information Systems and Organizational Change, Communications of the ACM (24:1), pp. 24-33. Keen, P. G. W., Bronsema, G. S., and Zuboff, S. 1982. Implementing Common Systems: One Organizations Experience, Systems, Objectives, Solutions (2:3), pp. 125-142. Keil, M. 1995. Pulling the Plug: Software Project Management and the Problem of Project Escalation, MIS Quarterly (19:4), pp. 421-447. Keil, M., Tan, B. C. Y., Wei, K. K., Saarinen, T., Tuunainen, V., and Wassenaar, A. 2000. A Cross-Cultural Study on Escalation of Commitment Behavior in Software Projects, MIS Quarterly (24:2), pp. 299-325. Kim, H. W., and Kankanhalli, A. 2009. Investigating User Resistance to Information Systems Implementation: A Status Quo Bias Perspective, MIS Quarterly (33:3), pp. 567-582.
Kitto, S., and Higgins, V. 2010. Working around ERPs in Technological Universities, Science, Technology and Human Values (35:1), pp. 29-56. Lapointe, L. 1999. Ladoption des systmes dinformation cliniques par les mdecins et les infirmires: une tude des variables individuelles, socio-politiques et organisationnelles, unpublished doctoral dissertation, HEC Montral, Montral. Lapointe, L., and Rivard, S. 2005. A Multilevel Model of Resistance to Information Technology Implementation, MIS Quarterly (29:3), pp. 461-491. Lapointe, L., and Rivard, S. 2006. Learning from Physicians Resistance to Information Technology Implementation, Canadian Medical Association Journal (174:11), pp. 1573-1578. Larsson, R. 1993. Case Survey Methodology: Quantitative Analysis of Patterns across Case Studies, Academy of Management Journal (36:6), pp. 1515-1546. Lauer, T., and Rajagopalan, B. 2002. Examining the Relationship between Acceptance and Resistance in System Implementation, in Proceedings of the 8th Americas Conferences on Information Systems, Dallas, TX, August 9-11, pp. 1297-1303. Lawrence, P. R. 1954. How to Deal with Resistance to Change, Harvard Business Review (32:3), pp. 49-57. Lederer, A. L., and Nath, R. 1991. Managing Organizational Issues in Information Systems Development, Journal of Systems Management (42:11), pp. 23-27, 39. Lee, H. G., and Clark, T. H. 1997. Market Process Reengineering through Electronic Market Systems: Opportunities and Challenges, Journal of Management Information Systems (13:3), pp. 113-137. Littler, C. R. 1982. The Development of the Labour Process in Capitalist Societies: A Comparative Study of the Transformation of Work Organization in Britain, Japan and the USA, London: Heineman Educational Books. Lucas, W. A. 1974. The Case Survey Method: Aggregating Case Experience, Santa Monica, CA: Rand Corporation (http://www. rand.org/pubs/reports/R1515). Lydon, J., Zanna, M. P., and Ross, M. 1988. Bolstering Attitudes by Autobiographical Recall: Attitude Persistence and Selective Memory, Personality and Social Psychology Bulletin (14), pp. 78-86. Lyytinen, K., and Robey, D. 1999. Learning Failure in Information System Development, Information Sytems Journal (9:2), pp. 85-101. Mann, F. C., and Williams, L. K. 1960. Observations on the Dynamics of a Change to Electronic Data-Processing Equipment, Administrative Science Quarterly (5:2), pp. 27-38. Marakas, G. M., and Hornik, S. 1996. Passive Resistance Misuse: Overt Support and Covert Recalcitrance in IS Implementation, European Journal of Information Systems (5:3), pp. 208-220. Markus, M. L. 1981. Implementation Politics: Top Management Support and User Involvement, Systems, Objectives, Solutions (1:4), pp. 203-215. Markus, M. L. 1983. Power, Politics, and MIS Implementation, Communications of the ACM (26:6), pp. 430-444.
918
Markus, M. L., and Keil, M. 1994. If We Build It, They Will Come: Designing Information Systems That People Want to Use, Sloan Management Review (35:4), pp. 11-25. Martinko, M. J., Henry, J. W., and Zmud, R. W. 1996. An Attributional Explanation of Individual Resistance to the Introduction of Information Technologies in the Workplace, Behaviour & Information Technology (15:5), pp. 313-330. Maude, B. 1971. Training Managers to Listen, Industrial and Commercial Training (3:7), pp. 305-307. Miles, M. B., and Huberman, A. M. 1994. Qualitative Data Analysis: An Expanded Sourcebook, Thousand Oaks, CA: Sage Publications. Miller, C. L. 1983. How to Successfully Resist a Computer System and Avoid its Benefits: A Victory for the Bureaucracy?, Systems, Objectives, Solutions (3:1), pp. 3-12. Mitroff, I. 1998. Smart Thinking for Crazy Times, San Francisco: Berret-Koehler Publishers. Moreno, V. J. 1999. On the Social Implications of Organizational Reengineering, Information Technology & People (12:4), pp. 359-388. Munkvold, B. E. 1999. Challenges of IT Implementation for Supporting Collaboration in Distributed Organizations, European Journal of Information Systems (8:4), pp. 260-272. Newig, J., and Fritsch, O. 2009. The Case Survey Method and Applications in Political Science, Paper presentation at the American Political Science Association Meeting, Toronto, September 3-6. Newman, M., and Robey, D. 1992. A Social Process Model of User-Analyst Relationships, MIS Quarterly (16:2), pp. 249-266. Par, G. 1995. Understanding the Dynamics of Information Technology Implementation: The Case of Clinical Information Systems, unpublished doctoral dissertation, Florida International University, Miami, FL. Petty, R. E., and Cacioppo, J. 1986. Elaboration Likelihood Model, in Advances in Experimental Social Psychology, L. Berkowitz (ed.), San Diego, CA: Academic Press, pp. 123-205. Poon, E. G., Blumenthal, D., Jaggi, T., Honour, M. M., Bates, D. W., and Kaushal, R. 2004. Overcoming Barriers to Adopting and Implementing Computerized Physician Order Entry Systems in U.S. Hospitals, Health Affairs (23:4), pp. 184-190. Ragin, C. C. 1999. Using Qualitative Comparative Analysis to Study Causal Complexity, Health Services Research (34:5), pp. 1225-1239. Ragin, C. C. 2006. Set Relations in Social Research: Evaluating Their Consistency and Coverage, Political Analysis (14:3), pp. 291-310. Ragin, C. C. 2008. Redesigning Social Inquiry: Fuzzy Sets and Beyond, Chicago: University of Chicago Press. Reuver, M. D., Bouwman, H., and MacInnes, I. 2009. Business Model Dynamics: A Case Survey, Journal of Theoretical and Applied Electronic Commerce Research (4:1), pp. 1-11. Reynierse, J. H. 1994. Ten Commandments for CEOs Seeking Organizational Change, Business Horizons (37:1), pp. 40-45.
Rihoux, B., and De Meur, G. 2009. Crisp-Set Comparative Analysis (CSQCA), in Configurational Comparative Methods, B. Rihoux and C. C. Ragin (eds.), Thousand Oaks: CA: Sage Publications, pp. 33-68. Rihoux, B., and Ragin, C. C. 2009. Introduction, in Configurational Comparative Methods, B. Rihoux and C. C. Ragin (eds.), Thousand Oaks, CA: Sage Publications, pp. xxvii-xxv. Roberto, M. A. 2002. Lessons from Everest: The Interaction of Cognitive Bias, Psychological, Safety, and System Complexity, California Management Review (45:1), pp. 136-158. Rowe, C. J. 1985. Identifying Causes of Failure : A Case Study in Computerized Stock Control, Behaviour & Information Technology (4:1), pp. 63-72. Roy, V. 1999 Le choix dun mode dapprovisionnement en systmes dinformation, une analyse par les comptences de base, unpublished doctoral dissertation, HEC Montral, Montral. Shaw, T. R., Pawlowski, S. D., and Davis, J. B. 2005. Building Theory About IT Professionals: Is a Taxonomy or Typology the Answer?, in Proceedings of the 2005 ACM SIGMIS CPR Conference on Computer Personnel Research, Atlanta, Georgia, pp. 9-11. Shimizu, K., and Hitt, M. A. 2004. Strategic Flexibility: Organizational Preparedness to Reverse Ineffective Strategic Decisions, Academy of Management Executive (18:4), pp. 44-59. Silva, L., and Backhouse, J. 2003. The Circuits-of-Power Framework for Studying Power in Institutionalization of Information Systems, Journal of the Association for Information Systems (4:6), pp. 294-336. Teng, J. T. C., Grover, V., and Fiedler, K. D. 1996. Developing Strategic Perspectives on Business Process Reengineering: From Process Reconfiguration to Organizational Change, Omega (24:3), pp. 271-294. Tormala, Z. L., and Petty, R. E. 2002. What Doesnt Kill Me Makes Me Stronger: The Effects of Resisting Persuasion on Attitude Certainty, Journal of Personality and Social Psychology (83), pp. 1288-1313. Tormala, Z. L., and Petty, R. E. 2004. Source Credibility and Attitude Certainty: A Metacognitive Analysis of Resistance to Persuasion, Journal of Consumer Psychology (14:4), pp. 427-442. Van Akkeren, J., and Rowlands, B. 2007. An Epidemic of Pain in an Australian Radiology Practice, European Journal of Information Systems (16:6), pp. 695-711. Wagemann, C., and Schneider, C. Q. 2007. Standards of Good Practice in Qualitative Comparative Analysis (QCA) and Fuzzy Sets, COMPASSS Working Paper (http://www.compasss.org/ pages/resources/wptheme.html). Wagner, E., and Newell, S. 2007. Exploring the Importance of Participation in the Post-Implementation Period of an Information System Project: A Neglected Area, Journal of the Association for Information Systems (8:10), pp. 508-524. Wang, B., and Paper, D. 2005. A Case of an IT-Enabled Organizational Change Intervention: The Missing Pieces, Journal of Cases on Information Technology (7:1), pp. 34-52.
919
Wathen, C. N., and Burkell, J. 2002. Believe It or Not: Factors Influencing Credibility on the Web, Journal of the American Society of Information Science and Technology (53:2), pp. 133-144. Wilson, M., and Howcroft, D. 2002. Re-Conceptualising Failure: Social Shaping Meets IS Research, European Journal of Information Systems (11:4), pp. 236-250. Yin, R. K., and Heald, L. A. 1975. Using the Case Survey Method to Analyze Policy Studies, Administrative Science Quarterly (20:3), pp. 371-381.
mation technology, outsourcing of information systems services, software project risk management, and strategic alignment. Her work has been published in such journals as Journal of Information Technology, Journal of the Association for Information Systems, Journal of Strategic Information Systems, MIS Quarterly, and Organization Science. Liette Lapointe is currently an associate professor, Information Systems Area coordinator, and director of the Business and Management Research Center at the Desautels Faculty of Management at McGill University. She holds an M.Sc. in Healthcare Administration from the Faculty of Medicine at Universit de Montral and a Ph.D. in Administration (Information Systems) from HEC Montral. Her research in information systems and healthcare management has been presented in conferences worldwide and published in scientific journals in management and medicine, such as MIS Quarterly, Journal of the Association for Information Systems, Organization Science, Implementation Science, International Journal of Medical Informatics, and Canadian Medical Association Journal.
About the Authors

Suzanne Rivard is a professor of Information Technology and holder of the Chair in Strategic Management of Information Technology at HEC Montral, Montral, QC. She received a Ph.D. from the Ivey School of Business, University of Western Ontario. Her research interests are in the areas of implementation of infor-
920
RESEARCH ARTICLE
INFORMATION TECHNOLOGY IMPLEMENTERS RESPONSES TO USER RESISTANCE: NATURE AND EFFECTS

Suzanne Rivard
HEC Montral, 3000 Chemin-de-la-Cte-Ste-Catherine, Montral H3T 2V7 CANADA {suzanne.rivard@hec.ca}
Liette Lapointe
Desautels Faculty of Management, McGill University, 1001 Sherbrooke Street West, Montral H3A 1G5 CANADA {liette.lapointe@mcgill.ca}
Appendix A
Case Sources1
Aarts, J., Doorewaard, H., and Berg, M. 2004. Understanding Implementation: The Case of a Computerized Physician Order Entry System in a Large Dutch University Medical Center, Journal of the American Medical Informatics Association (11:3), pp. 207-216. Allen, D. K., Colligan, D., Finnie, A., and Kern,T. 2000. Trust, Power and Interorganizational Information Systems: The Case of the Electronic Trading Community TransLease, Information Systems Journal (10:1), pp. 21-40. Alter, S. 1980. Decision Support Systems: Current Practice and Continuing Challenges, Reading, MA: Addison-Wesley Publishing Company. Bartis, E., and Mitev, N. 2008. A Multiple Narrative Approach to Information Systems Failure: A Successful System that Failed, European Journal of Information Systems (17:2), pp. 112-124. Bondarouk, T. V. 2004. Implementation of a Personnel Management System Beaufort: Successes and Failures at a Dutch Hospital, Annals of Cases on Information Technology (6), pp. 352-369. Boudreau, M.-C. 2000. Exploring the Transition to Enterprise Resource Planning: A Longitudinal Study of IT-Related Change, unpublished Ph.D. thesis, College of Business Administration, Georgia State University, Atlanta, GA. Box, T. L., McDonell, M., Helfrich, C. D., Jesse, R. L., Fihn, S. D, and Rumsfeld, J. S. 2010. Strategies from a Nationwide Health Information Technology Implementation: The VA CART STORY, Journal of General Internal Medicine (25:1), pp. 72-76. Carasik, R. P., and Grantham, C. E. 1988. A Case Study of CSCW in a Dispersed Organization, in Proceedings of SIGCHI Conference on Human Factors and Computing Systems, J. J. OHare (ed.), Washington, DC, May 15-19, pp. 61-66. Cornet, A. 1998. Reconfiguration d'un processus de commande/livraison : Le reengineering face ses contradictions, unpublished Ph.D. thesis, Facult d'conomie, de Gestion et de Sciences Sociales, Universit de Lige, Belgium. DeSanctis, G., Jackson, B. M., Poole, M. C., and Dickson, G. W. 1996. Infrastructure for Telework: Electronic Communication at Texaco, in Proceedings of the ACM SIGCPR/SIGMIS Conference on Computer Personnel Research, M. Igbaria (ed.), Denver, CO, April 11-13, pp. 94-102. Dickson, G. W., Simmons, J. K., and Anderson, J. C. 1974. Behavioral Reactions to the Introduction of a Management Information System at the US Post Office: Some Empirical Observations, in Computers and Management in a Changing Society, D. H. Sanders (ed.), New York: McGraw Hill Book Company, pp. 410-421.
Please note that some cases relate to multiple sources; some sources include multiple cases.
MIS Quarterly Vol. 36 No. 3Appendices/September 2012
A1
Doolin, B. 1999. Casemix Management in a New Zealand Hospital: Rationalisation and Resistance, Financial Accounting & Management (15:3-4), pp. 397-417. Farrell, K. T., and Lute, J. E. 2005. Document-Management Technology and Acquisitions Workflow: A Case Study in Invoice Processing, Information Technology and Libraries (24:3), pp. 117-122. Ferneley, E. H., and Sobreperez, P. 2006. Resist, Comply or Workaround? An Examination of Different Facets of User Engagement with Information Systems, European Journal of Information Systems (15:4), pp. 345-356. Gallivan, M. J. 2001. Organizational Adoption and Assimilation of Complex Technological Innovations: Development and Application of a New Framework, The Data Base for Advances in Information Systems (32:3), pp. 51-85. Gunawardane, G. 1985. Implementing a Management Information System in an Extremely Dynamic (and Somewhat Hostile) Environment: A Case Study, Interfaces (15:6), pp. 93-99. Hartmann, T., and Fischer, M. 2009. A Process View on End User Resistance During Construction IT Implementations, Journal of Information Technology in Construction (14), pp. 353-365. Hennen, J. M. 2002. Twilight or Rebirth? A Case Study of How New Technology Reconstructed Academic Counseling in a Community College, unpublished Ph.D. thesis, University of St. Thomas, St. Paul, MN. Hirschheim, R., and Newman, M. 1988. Information Systems and User Resistance: Theory and Practice, The Computer Journal (31:5), pp. 398-408. Jones, M. R. 2003. Computers Can Land People on Mars, Why Cant They Get Them to Work in a Hospital? Implementation of an Electronic Patient Record System in a UK Hospital, Methods of Information in Medicine (42:4), pp. 410-415. Joshi, K. 2005. Understanding User Resistance and Acceptance During the Implementation of an Order Management System: A Case Study Using the Equity Implementation Model, Journal of Information Technology Cases and Application Research (7:1), pp. 6-20. Keen, P. G. W., Bronsema, G. S., and Zuboff, S. 1982. Implementing Common Systems: One Organizations Experience, Systems, Objectives, Solutions (2:3), pp. 125-142. Kitto, S., and Higgins, V. 2010. Working Around ERPs in Technological Universities, Science, Technology & Human Values (35:1), pp. 29-54. Lapointe, L. 1999. Ladoption de systemes d'information cliniques par les medecins et les infirmieres: Une tude des variables individuelles, socio-politiques et organisationnelles, unpublished Ph.D. thesis, HEC Montral, Montral, QC. Lyytinen K., and Robey, D. 1999. Learning Failure in Information System Development, Information Systems Journal (9:2), pp. 85-101. Malato, L. A., and Kim, S. 2004. End-User Perceptions of a Computerized Medication System: Is There Resistance to Change?, Journal of Health and Human Services Administration (27:1), pp. 34-55. Markus, M. L. 1981. Implementation Politics: Top Management Support and User Involvement, Systems, Objectives, Solutions (1:4), pp. 203-215. Markus, M. L. 1983. Power, Politics, and MIS Implementation, Communications of the ACM (26:6), pp. 430-444. Markus, M. L., and Keil, M. 1994. If We Build It, They Will Come: Designing Information Systems That People Want to Use, MIT Sloan Management Review (35:4), pp. 11-25. Massaro, T. A. 1993. Introducing Physician Order Entry at a Major Academic Medical Center: I. Impact on Organizational Culture and Behavior, Academic Medicine (68:1), pp. 20-25. Massaro, T. A. 1993. Introducing Physician Order Entry at a Major Academic Medical Center: II. Impact on Education, Academic Medicine (68:1), pp. 25-30. Metcalf, C. J. 1999. Cultural Influences on the Adoption and Use of Information and Communication Technologies in Organizations, unpublished Doctoral thesis, Wayne State University, Detroit, MI. Miller, C. L. 1983. How to Successfully Resist a Computer System and Avoid its Benefits: A Victory for the Bureaucracy?, Systems, Objectives, Solutions (3:1), pp. 3-12. Nanji, K. C., Cina, J., Patel, N., Churchill, W., Gandhi, T. J., and Poon, E. G. 2009. Overcoming Barriers to the Implementation of a Pharmacy Bar Code Scanning System for Medication Dispensing: A Case Study, Journal of the American Medical Informatics Association (16:5), pp. 645-650. Newman, M., and Robey, D. 1992. A Social Process Model of User-Analyst Relationships, MIS Quarterly (16:2), pp. 249-266. Orlikowski, W. J. 1992. Learning from Notes: Organizational Issues in Groupware Implementation, in Proceedings of the 1992 ACM Conference on Computer-Supported Cooperative Work, Toronto, pp. 362-369. Orlikowski, W. J. 1993. CASE Tools as Organizational Change: Investigating Incremental and Radical Changes in Systems Development, MIS Quarterly (17:3), pp. 309-340. Par, G. 1995. Understanding the Dynamics of Information Technology Implementation: The Case of Clinical Information Systems, unpublished Ph.D. thesis, Florida International University, Miami, FL. Par, G., and Elam, J. J. 1998. Introducing Information Technology in the Clinical Setting: Lessons Learned in a Trauma Center, International Journal of Technology Assessment in Health Care (14:2), pp. 331-343.
A2
Powell, G. 1996 Microchips Versus Stethoscopes, Part 2: Revisiting OSCAR at the Foothills Hospital, Canadian Medical Association Journal (155:11), pp. 1601-1603. Prasad, P., and Prasad, A. 2000. Stretching the Iron Cage: The Constitution and Implications of Routine Workplace Resistance, Organization Science (11:4), pp. 387-403. Robey, D., and Rodriguez-Dias, A. 1989. The Organizational and Cultural Context of Systems Implementation: Case Experience from Latin America, Information & Management (17:4), pp. 229-239. Rowe, C. J. 1985. Identifying Causes of Failure: A Case Study in Computerized Stock Control, Behaviour and Information Technology (4:1), pp. 63-72. Roy, V. 1999. Le choix d'un mode d'approvisionnement en systmes d'information: Une analyse par les comptences de base, unpublished Ph.D. thesis, HEC Montral, Montral, QC. Silva, L., and Backhouse, J. 2003. The Circuits-of-Power Framework for Studying Power in Institutionalization of Information Systems, Journal of the Association for Information Systems (4:6), pp. 294-336. Sloane, S. B. 1991. The Use of Artificial Intelligence by the United States Navy: Case Study of a Failure, AI Magazine (12:1), pp. 80-92. Van Akkeren, J., and Rowlands, B. 2007. An Epidemic of Pain in an Australian Radiology Practice, European Journal of Information Systems (16:6), pp. 695-711. Wagner, E., and Newell, S. 2007. Exploring the Importance of Participation in the Post-Implementation Period of an ES Project: A Neglected Area, Journal of the Association for Information Systems (8:10), pp. 508-524. Wang, B., and Paper, D. 2005. A Case of an IT-Enabled Organizational Change Intervention: The Missing Pieces, Journal of Cases on Information Technology (7:1), pp. 34-52. Webster, J. 1998. Desktop Videoconferencing: Experiences of Complete Users, Wary Users, and Non-Users, MIS Quarterly (22:3), pp. 257-286. Whitley, E. A., and Pouloudi, A. 2001. Studying the Translations of NHSnet, Journal of End User Computing (13:3), pp. 30-41. Williams, L. S. 1992. Microchips Versus Stethoscopes: Calgary Hospital, MDs Face Off Over Controversial Computer System, Canadian Medical Association Journal (147:10), pp. 1534-1547. Wilson, M., and Howcroft, D. 2002. Re-conceptualising Failure: Social Shaping Meets IS Research, European Journal of Information Systems (11:4), pp. 236-250.
A3
Appendix B
Results of Delphi Study on Resistance Behaviors
Resistance Level
1 Being apathetic Being inactive Lacking interest in project activities Avoiding involvement in the project Doing as little as possible in the project context Gossiping about the implementers Non-key actors refusing to accept responsibility in the project Not utilizing the new system when utilization is voluntary Persisting in using a former system in parallel with the new system when usage is voluntary Using humor Voicing feelings of fear and concern toward the system/project Engaging in absenteeism behavior Complaining about the project/system Denying the need for the new system Gossiping about the project/system Making excuses for not participating in project-related activities Making excuses for not using the system Postponing actions/decisions in the project context Pretending to comply Refusing to engage in joint problem-solving Voicing opposing points of view regarding the system/project Withdrawing Asking others to intervene on project-related issues Attacking the credibility of the project Forming coalitions Key actors refusing to accept responsibility in the project Leaking information about the project Leaving the organization Leaving the department/business unit Not utilizing the new system when utilization is mandatory/expected Persisting in using the former system in parallel with the new system when usage is mandatory/expected Whistle-blowing about the system/project Attacking the credibility of the implementers Boycotting project activities Boycotting the system Engaging in contentious rivalry about project issues among members of the organization Making threats Posing ultimatums Rebelling Deliberately committing errors Engaging in destructive behaviors Going on strike Overtly blocking project activities Resorting to violence Sabotage Spoilage
Resistance Behavior
A4
Appendix C
Truth Table of Implementers Responses as a Sufficient Monocausal Condition
ACKNOWLEDGMENT
NONCONGRUENT RECTIFICATION
AUTHORITATIVE PERSUASION
UNAWARENESS
CONGRUENT RECTIFICATION
1 0 0 0 0 0 0 0 0
0 1 0 0 0 0 0 0 0
0 0 1 0 0 0 0 0 0
0 0 0 1 0 0 0 0 0
0 0 0 0 1 0 0 0 0
0 0 0 0 0 1 0 0 0
0 0 0 0 0 0 1 0 0
0 0 0 0 0 0 0 1 0
0 0 0 0 0 0 0 0 1
16 25 6 5 0 11 4 1 6
Appendix D
Truth Table of Dissuasive Responses, Implementers Power, and Message Credibility
NUMBER OF EPISODES COERCION AUTHORITATIVE PERSUASION SUPPORTIVE PERSUASION POWER CREDIBILITY RESISTANCE INCREASE RESISTANCE DECREASE
3 3 0 0 6 0 1 1 4 2 7 4
1 1 1 1 0 0 0 0 0 0 0 0
0 0 0 0 1 1 1 1 0 0 0 0
0 0 0 0 0 0 0 0 1 1 1 1
1 1 0 0 1 1 0 0 1 1 0 0
1 0 1 0 1 0 1 0 1 0 1 0
0 1 R R 0 R 0 1 0 1 0 1
1 0 R R 1 R 1 0 1 0 1 0
R indicates the presence of a logical remainderthat is, an unobserved combination of conditions (Rihoux, B., and De Meur, G. 2009. Crisp-Set Comparative Analysis (CSQCA), in Configurational Comparative Methods, B. Rihoux and C. C. Ragin (eds.), Thousand Oaks, CA: Sage Publications, pp. 33-68).
RESISTANCE DECREASE (N)
RESISTANCE INCREASE (N)
SUPPORTIVE PERSUASION
DELIBERATE IGNORANCE
IMPOTENCE
COERCION
0 0 0 0 31 0 2 7 11
A5
Copyright of MIS Quarterly is the property of MIS Quarterly & The Society for Information Management and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.

78168532

Caricato da

Informazioni sul documento

Descrizione originale:

Copyright

Formati disponibili

Condividi questo documento

Condividi o incorpora il documento

Opzioni di condivisione

Hai trovato utile questo documento?

Questo contenuto è inappropriato?

Copyright:

Formati disponibili

78168532

Caricato da

Copyright:

Formati disponibili

RESEARCH ARTICLE

INFORMATION TECHNOLOGY IMPLEMENTERS RESPONSES 1 TO USER RESISTANCE: NATURE AND EFFECTS

MIS Quarterly Vol. 36 No. 3 pp. 897-920/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

Table 1. A Taxonomy of Implementers Responses

N (%) 18 (13.1%) 27 (19.7%) 7 (5.1%) Doing nothing

DELIBERATE IGNORANCE IMPOTENCE

6 (4.4%) 8 (5.8%) 17 (12.4%)

AUTHORITATIVE PERSUASION SUPPORTIVE PERSUASION

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

Table 2. Implementers Responses and Ex Post Levels of Resistance

N (%) 52 (38.0%) 5 (3.6%) 49 (35.8%) 31 (22.6%) 137 (100%)

Higher (%) 47 (90.4%) 5 (100% 16 (32.6%) 11 (35.5%) 79 (57.7%)

Lower (%) 0 (0%) 0 (0%) 31 (63.3%) 20 (64.5%) 51 (37.2%)

No Change (%) 5 (9.6%) 0 (0%) 2 (4.1%) 0 (0%) 7 (5.1%)

Total 52 (100%) 5 (100%) 49 (100%) 31 (100%) = 59.52 6df p < 0.000

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

Wang and Paper (2005)

Wagner and Newell (2007)

DeSanctis et al. (1996)

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

Table 6. CONGRUENT RECTIFICATION

Silva and Backhouse (2003)

Table 7. NONCONGRUENT RECTIFICATION

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

Table 8. AUTHORITATIVE PERSUASION

Initial Resistance Level/Behavior Level 2 / [Some] simply avoided it (p. 62).

Ex-post Resistance Level/Behavior No resistance / continued use (p. 63).

Source Carasik and Grantham (1988)

Manager/ Top management

Table 9. SUPPORTIVE PERSUASION (Decrease in the Level of Resistance)

Source Keen et al. (1982)

Level 2 / Their perception of an unfair balance of power (p. 30)

Repair agents/ IT consultants

Allen et al. (2000)

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

Table 10. SUPPORTIVE PERSUASION (Increase in the Level of User Resistance)

Source Wilson and Howcroft (2002)

Sales people/ Management

MIS Quarterly Vol. 36 No. 3/September 2012

Rivard & Lapointe/IT Implementers Responses to User Resistance

Table 11. COERCION (Increased Level of User Resistance)

Source Allen et al. (2000)

Physicians/ General management