Information and Organization: Leiser Silva, H. Kevin Fulk

Information and Organization 22 (2012) 227–251
Contents lists available at SciVerse ScienceDirect
Information and Organization

journal homepage: www.elsevier.com/locate/
infoandorg
From disruptions to struggles: Theorizing power in ERP

implementation projects
Leiser Silva ⁎, H. Kevin Fulk
Bauer College of Business, University of Houston, 4800 Calhoun, Houston, TX 77204, United States
Southern University at New Orleans, 6801 Press Drive, New Orleans, LA 70126, United States
a r t i c l e i n f o a b s t r a c t
Article history: Enterprise resource planning (ERP) implementation projects often

Received 19 February 2008 fail to achieve their objectives. These often problematic projects are
Received in revised form 7 June 2012 frequently the setting for intense and growing power struggles.
Accepted 7 June 2012
Existing studies (e.g., Boonstra, 2006; Shepherd, Clegg, & Stride,
Available online 1 July 2012
2009) have provided researchers with insights about issues of power
in these projects. However, existing research has yet to provide a
comprehensive picture of power in these projects or insights on how
this picture changes with the passage of time. Clegg's (1989) circuits
of power framework provides a useful tool for developing this needed
comprehensive picture. We use the circuits of power framework as a
tool for categorizing existing literature on power in ERP implementa-
tions. More importantly, we draw on this framework to provide a
comprehensive view of power in the particular context of these
projects. Specifically, we analyze the power relations during the
implementation of an ERP in an organization. We do so by identifying
how disturbances to the circuits of power – power struggles – arise
and intensify during the implementation of the ERP. In this way, our
work makes both a theoretical and an empirical contribution to the
study of power in ERP implementation projects.
© 2012 Elsevier Ltd. All rights reserved.
1. Introduction
The implementation of ERP systems continues to garner the interest of business organizations. An ERP
system is a configurable off-the-shelf software package, an integrated set of systems and information
resources to coordinate a wide range of operational and management activities such as procurement and
accounting (Akkermans & van Helden, 2002; Al-Mudimigh, Zairi, & Al-Mashari, 2001; Daniel & White,
2005; Davenport, 1998). Purchases of ERP systems rank as the largest segment of enterprise software
⁎ Corresponding author. Tel.: + 1 713 743 4370; fax: + 1 713 743 4940.
E-mail addresses: lsilva@uh.edu (L. Silva), hkfulk@suno.edu (H.K. Fulk).
1471-7727/$ – see front matter © 2012 Elsevier Ltd. All rights reserved.
doi:10.1016/j.infoandorg.2012.06.001
228 L. Silva, H.K. Fulk / Information and Organization 22 (2012) 227–251
spending, which is projected to increase through 2015 (Gartner, Inc., 2011). ERP systems are expected to
provide benefits such as reducing development costs and giving real time access to business data
(Davenport, 1998; Palomino Murcia & Whitley, 2007; Seethamraju, 2008). However, ERP implementation
projects are often problematic endeavors. Estimates suggest that up to 75% of ERP implementations fail to
achieve their intended objectives (Robey, Ross, & Boudreau, 2002). Faced with the often problematic
outcomes of ERP implementations, IS researchers have sought to understand reasons for these projects'
poor performance.
Power struggles, often manifested in conflict and resistance, have been referred to as one of the main
reasons why ERP projects fail to achieve their goals (see for example Allen, Kern, & Havenhand, 2002;
Markus, Tanis, & van Fenema, 2000). Although, there have been many studies that concentrate on power
and ERP implementations (these will be discussed in the literature review section), their scope is narrow
as these tend to focus mainly on conflicting actions. However, as it has been demonstrated in sociology
(e.g., Clegg, 1989; Giddens, 1984; Lukes, 1974) and IS literature (see Jasperson, Carte, Saunders, Butler,
Croes, & Zheng, 2002), power operates beyond actions and behaviors; it does it also through the creation
of meanings and techniques of production. Hence the motivation of our paper; that is identifying what are
the antecedents of power struggles. We will argue that these struggles are rooted in rules of meaning and
techniques of production. Accordingly, our research objectives are: (1) to empirically demonstrate –
through a case study – how power operates through meanings and techniques of production and (2) to
theorize on the conditions that engender power struggles.
The remainder of this paper is organized as follows. The next section discusses our theoretical lens:
Clegg's Circuits of Power. 1 This is followed by a literature review of ERP studies which have power as their
focus. The section after this describes our research method. Next, we present the case study's narrative
along with its analysis through the lens of the Circuits of Power. We conclude by discussing the results of
our study, its implications and limitations, as well as possible directions for further research.
2. Circuits of power
The Circuits of Power framework is a legitimate theoretical lens for IS researchers to examine issues of
power. In our field it has been applied to examine the institutionalization of information systems (Silva &
Backhouse, 2003) and information security standards (Backhouse, Hsu, & Silva, 2006; Smith, Winchester,
Bunker, & Jamieson, 2010). With some limitations, which we discuss at the end of this section, we argue it
is an appropriate theoretical lens for examining power's role in ERP implementations because it
comprehensively integrates three different types of power: facilitative, dispositional and causal. Each type
of power corresponds to a circuit of power (see Fig. 1).
Facilitative power concerns a productive view of power that sees organizations and collective actions
as the result of power (Parsons, 1967: 345). Dispositional power sees power as a capability that
individuals have to produce outcomes as the result of their associations (Weber, 1999) as well as by
drawing on discursive resources (Foucault, 1977). Causal power concerns actions in which one individual
or a group (As) makes others, either a group or an individual, (Bs) do things the latter would not do
otherwise (Dahl, 1957). Thus, it is through this multi-perspective framework that Clegg argues power
should be understood. A more detailed discussion of the concepts of the framework is presented below.
2.1. Conceptual account of the circuits of power
For Clegg power is the force that keeps social systems coherent. Therefore, from this perspective, an
organization – understood as organized collective action – is the result of power. Hence, for Clegg, power
in organizations is manifested through the production of specific outcomes: tangible products or
1
The Circuits of Power framework is not the only approach used by IS researchers to consider multiple perspectives of power.
Examples include the well-known framework proposed by Markus and Björn-Andersen (1987) and the more recent approach
proposed by Sabherwal and Grover (2010). However, Markus and Björn-Andersen do not address how power can be embedded in
techniques of production and discipline. Sabherwal and Grover do not address the roles of rules of membership and meaning, or the
influence of exogenous contingencies. Given the attention that the Circuits of Power framework pays to these missing elements, we
selected it for our study.
L. Silva, H.K. Fulk / Information and Organization 22 (2012) 227–251 229
Fig. 1. Circuits of power.

Adapted from Clegg (1989).
institutionalized practices. From this perspective, organizational and social outcomes are possible as the
result of relations among persons, relations sustained by power. In the case of an ERP implementation, the
outcome of power would be the system being institutionalized in the organizational practices (Silva &
Backhouse, 2003). An organizational outcome originated in causal power can reproduce or transform
extant conditions of integration in both, systemic and social circuits (see Fig. 1). For example, an
implemented ERP can boost systemic integration by tightening controls and at the same time can reinforce
social integration by emphasizing the distinction between the roles played by management and
employees.
As depicted in Fig. 1, the circuit of episodic power is defined by the relationship between two types of
agents — As and Bs. The As make the Bs do something they would not do otherwise (Foucault, 1982). Clegg
warns that looking into the circuit of episodic power is not enough to understand the power phenomenon
entirely (i.e., concentrating on agents' actions is not enough to understand power completely). We need
instead to consider the conditions that secure access to and control of resources to ensure compliance, and
that allow As to have power over Bs. These are what Clegg calls standing conditions. Standing conditions
here are the symbolic and technological resources that facilitate or restrict As' and Bs' causal powers (this
is represented in Fig. 1 by the arrow stemming from the Standing Conditions Box). These are constituted
by the other two circuits of power, that is, social and systemic integration; for example, an actor's
authority based upon their position within an organizational hierarchy. In Fig. 1 the constitution of the
standing conditions as established by the other two circuits is represented by the arrow stemming from
both circuits.
The circuit of social integration is composed of two main concepts: rules of meaning and rules of
membership. This is closely related to the power of discourse as explained by Foucault, who indicates that
power is exercised by virtue of how we interpret events and by what we believe to be true (Foucault,
1980). The rules of meaning are also related to the concept of legitimacy as defined by institutional
theorists (Meyer & Rowan, 1991) (i.e., the power that establishes what is legitimate and distinguishes it
from what is not). Likewise, the rules of membership refer to power originated by individuals' association
with particular organizational groups. In this sense, social integration, that is an outcome of the circuit of
dispositional power, refers to legitimacy and to uncontested rules of meaning and membership. In the case
of an ERP implementation, social integration would be manifested when the system, as proposed by the
management group, is considered to be legitimate by the users group.
Although the interpretation of power provided by the two circuits mentioned above is incisive, our
understanding of the power relation between As and Bs should be complemented by consideration of the
systemic integration circuit. This circuit refers to the material conditions, such as means of production and
discipline, which afford an organization the production of services and goods, as well as its surveillance
practices. Systemic integration here is understood to be a high degree of cohesiveness in an organization
that permits the production of goods and services. Thus, Clegg distinguishes two concepts: techniques of
discipline and techniques of production. The techniques of production refer to material means and
procedures drawn on by organizations to generate goods and services, whereas the techniques of
discipline refer to the measures and methods to ensure compliance. In the case of an ERP implementation,
systemic integration is achieved when the system is in production, both increasing managerial controls on
operations and enhancing overall organizational performance.
The circular property of the circuits is manifested when the repeated actions of Bs' compliance become
institutionalized and integrated into the other two circuits reinforcing or transforming extant conditions
(in Fig. 1 this is depicted by the arrow stemming from the Outcomes Box). For example, in an organization,
after continuous repetition, the practice of Bs (subordinates) using an ERP proposed by As (managers) will
become an institutionalized technique of production and discipline. At the same time, the institutional-
ization of a practice such as this will also reinforce the social relations and working identities between As
and Bs. One of the limitations of the framework, though, is its orientation towards explaining social and
organizational stability.
Although organizational instability is defined by Clegg as the absence of integration in the social and
systemic circuits, the framework lacks some explanatory power when referring to the source of
disturbances in the circuits. Clegg ascribes disturbances and changes to the circuits to exogenous
contingencies in the same sense as institutional theorists explain changes to organizations (DiMaggio &
Powell, 1983; Meyer & Rowan, 1991). However, the Circuits of power framework does not go far enough in
explaining disintegration when this is related to internal organizational actors. Hence the relevance of our
research objectives; on the one hand, we will show empirically how power operates through the three
circuits of power, and on the other, we will highlight the sources of disturbances that hinder integration.
Moreover, in the discussion section, we will theorize on the disturbances leading to disintegration by
focusing on how these disturbances originate, the way they operate and their consequences.
3. ERP system implementation research addressing issues of power
In this section, we discuss the literature on ERP system implementation and draw on the circuits of
power to classify it. Each paper is categorized according to its focus in relation to power which could be on
the episodic, social or systemic circuit. Our objective for this classification is to show that an integrated
view of power – one that incorporates the three circuits – can help researchers to better understand power
and politics that occur within the context of ERP implementations. We acknowledge that it is possible to
classify some papers into more than one of these four categories. So, this classification is done for
analytical purposes only.
3.1. Episodic circuit
Most ERP implementation studies concentrate on the episodic circuit. These studies focus on actors'
power exercises (i.e., resistance, and standing conditions). Accordingly, we found in this literature that
power may be exercised by different types of actors such as managers and users (Fowler & Gilfillan, 2003;
Hislop, Newell, Scarbrough, & Swan, 2000; Pan & Sritharan, 2007; Sathish, 2006; Scott & Wagner, 2003;
Swan & Scarbrough, 2005; Volkoff, 1999; Wagner & Newell, 2006), vendors and consultants (Khoo &
Robey, 2007; Scott & Wagner, 2003; Wagner & Newell, 2006), and firms themselves (Seethamraju, 2008).
These actors exercise power in order to secure others' involvement in implementations (e.g., their use of
the new ERP system) (Pan & Sritharan, 2007; Sathish, 2006; Scott & Wagner, 2003), as well as to influence
decision-making regarding implementations (Khoo & Robey, 2007; Scott & Wagner, 2003; Seethamraju,
2008; Wagner & Newell, 2006).
The outcome of power exercises may also have consequences for the design of the resulting ERP
system. Through their exercises of power, managers and others championing the implementation may be
able to design the ERP in a way that will only reflect their interests (Bob-Jones, Newman, & Lyytinen, 2008;
Marabelli & Newell, 2009). However, for implementations in which multiple powerful actors are involved,
the implemented system may be reflective of the interests of these multiple actors. To ensure that their
interests are furthered by the implemented system, these actors have to engage in negotiations about the
configuration of the ERP system (Boonstra, 2006; Scott & Wagner, 2003; Wagner & Newell, 2006; Wagner,
Scott, & Galliers, 2006).
When no one powerful group is able to dominate these negotiations about the ERP system, multiple
actors are able to dictate aspects of the new system's configuration to a limited extent. The resulting ERP
system may be inscribed with a combination of these various actors' interests, rather than reflecting the
interests of only one individual or group in the organization (Boonstra, 2006; Scott & Wagner, 2003;
Wagner & Newell, 2006; Wagner, Newell, & Piccoli, 2010; Wagner et al., 2006). Despite these negotiations,
there may be groups whose interests are marginalized, producing an ERP system that fails to reflect their
interests (Marabelli & Newell, 2009; Scott & Wagner, 2003; Wagner & Newell, 2006; Wagner et al., 2006).
When this happens, resistance arises.
Standing conditions may be based on an actor's position within the organizational hierarchy (Fowler &
Gilfillan, 2003; Pan & Sritharan, 2007; Wagner et al., 2006), a position that may provide access to resources
valuable during the implementation process. In these implementations, actors' authority may also be
based upon the delegation of power from someone else (Pan & Sritharan, 2007), or the independence of
one's work practices from others' influence (Lim, Pan, & Tan, 2005). The network of relationships in which
an actor exists may also provide the basis of their standing conditions (Hislop et al., 2000; Pan & Sritharan,
2007). Such actors may seek to enhance their standing conditions by forming additional alliances in
support of the implementation with others inside or outside of the organization (Swan & Scarbrough,
2005; Wagner & Newell, 2006).
In short, these studies offer varied insights about exercises of power, resistance, and standing
conditions in the context of ERP implementations. An enhanced understanding of the episodic circuit,
though, requires a broader view of power, one that incorporates the other two circuits.
3.2. Social integration circuit
Studies that address issues of social integration provide insights about legitimacy, contradicting
interpretations, identity, and group membership in the context of ERP implementations. The literature
shows that actors championing an implementation may do so more effectively if organizational rules or
norms support these actors' activities as legitimate (Bagchi, Kanungo, & Dasgupta, 2003; Hislop et al.,
2000; Pan & Sritharan, 2007; Parr & Shanks, 2000; Volkoff, 1999). Those championing an implementation
may also employ discourses in an effort to legitimate an implementation (e.g., the new system will further
rational goals) (Shepherd, et al., 2009). However, these discourses may be challenged by others using
competing discourses (e.g., the new system will interfere with our work), threatening the implementa-
tion's legitimacy (Shepherd, et al., 2009).
Studies concerning the circuit of social integration offer insights about actors' antagonistic
interpretations in the context of ERP implementations. A common finding is that groups may interpret
the ERP system and its implementation differently (Allen et al., 2002; Harley, Wright, Hall, & Dery, 2006;
Lim et al., 2005; Pollock & Cornford, 2004; Scott & Wagner, 2003; Shepherd et al., 2009; Swan &
Scarbrough, 2005; Wagner et al., 2006; Whittaker & Beswick, 2008). Research in this category suggests
that actors, whose interpretation of the new system (e.g., the ERP system threatens their jobs), opposes
the one held by management, may turn to acts of resistance (Lim et al., 2005; Palomino Murcia & Whitley,
2007; Scott & Wagner, 2003; Whittaker & Beswick, 2008). For example, management may interpret the
system as a necessary tool to establish control. In contrast, workers may interpret the system as a means
for changing their work practices, a bother to use, and an instrument of domination from management.
Multiple factors may influence antagonistic interpretations of the new ERP system and the project
undertaken to implement it. For example, users may come to interpret the ERP system differently from
management, depending upon whether the system enhances their ability to perform their work or
threatens their ability to do their work (Harley et al., 2006; Scott & Wagner, 2003; Shepherd et al., 2009).
The literature also shows that users may alter their interpretations after actually using the new system,
replacing a more optimistic view of the system with a more critical view of it (Alvarez, 2008). Moreover,
the literature also shows that formal and informal organizational rules (Niehaves, Klose, & Becker, 2006),
as well as actors' cultural backgrounds (Palomino Murcia & Whitley, 2007), may also influence how the
system is interpreted.
We found studies that focus on issues of social integration but at the same time are related to the
episodic circuit of power. These studies typically show actors who champion the new system inducing
their opponents to develop favorable interpretations of the ERP (Adam & O'Doherty, 2000; Bohorquez
Lopez & Esteves, 2009; Dong, Neufeld, & Higgins, 2009; Swan & Scarbrough, 2005). Developing these
favorable interpretations is crucial for the organization to avoid resistance, as discussed in the section
addressing the episodic circuit, and to ensure smooth implementations. These actors supporting the
implementation may encourage such favorable views of the new system through involving the other
actors early in the implementation process (Bajwa, Garcia, & Mooney, 2004); maintaining effective
communication with them (Allen et al., 2002; Amoako-Gyampah & Salam, 2004); in the case of managers,
involving them in change management activities (Dong et al., 2009; Harley et al., 2006); training actors on
how to use the system (Amoako-Gyampah & Salam, 2004; Dong et al., 2009; Pan & Sritharan, 2007);
educating them on the benefits of the new system (Adam & O'Doherty, 2000; Bohorquez Lopez & Esteves,
2009) and its fit with the organization (Bohorquez Lopez & Esteves, 2009; Pan & Sritharan, 2007); or
through top management taking supportive actions that are adjusted in response to the dynamics of the
implementation process (Dong et al., 2009). In addition, the literature also shows that favorable
interpretations may be spread through communications with organizational actors and by forming groups
with these other actors (Alvarez, 2000; Swan & Scarbrough, 2005).
Those championing a particular interpretation of the ERP system or its implementation may seek to
silence competing interpretations (Scott & Wagner, 2003; Wagner et al., 2006). These actors may do so by
convincing other actors that their interests are aligned with this interpretation, while marginalizing the
interests of those holding a competing interpretation (e.g., users) (Howcroft & Light, 2002). However,
efforts to silence competing interpretations may not be successful. Those actors who promote competing
views may enjoy sufficiently strong standing conditions. The end result of a situation such as this is that no
one interpretation attains dominance in the implementing organization (Scott & Wagner, 2003; Wagner et
al., 2006). Lastly, one group's efforts to force another group to adopt its interpretation may instead breed
negative interpretations (Allen et al., 2002).
Additional research calls our attention to conflicts between groups in ERP implementations. ERP
systems are touted as allowing seamless integration of business processes in organizations and appear to
assume the existence of cooperative relationships among organizational actors (Elbanna, 2007). However,
groups in an organization may compete with one another (Elbanna, 2007). These divisions along group
lines in an organization may be reflected in the implemented system (requiring costly and elaborate
modifications to the system), compromising its ability to provide integration of current means of
production and work practices (Elbanna, 2007). In this body of studies, social integration appears to be an
uncommon occurrence. Instead, most of them examine implementations with conditions (e.g., actors'
conflicting interpretations) that may hinder the integration of this circuit. To complement these insights of
these studies, we must also consider research that addresses issues of the systemic integration circuit.
3.3. Systemic integration circuit
Studies that focus on ERPs being used as techniques of discipline and production were grouped under
the systemic circuit category. To begin, ERP systems can enhance the ability of upper management to
exercise disciplinary power over their employees (Caglio, 2003; Elmes, Strong, & Volkoff, 2005; Harley et
al., 2006). The ERP system can contribute to management's capacity to exercise discipline in two ways.
First, ERP systems can enhance organizational control by increasing the visibility of work performed by the
organization's employees (Berente, Yoo, & Lyytinen, 2008; Elmes et al., 2005).
The second way that an ERP system can enhance management's ability to exercise disciplinary power is
by carrying inscriptions that prescribe specific approaches to performing work tasks. These inscriptions
allow management to structure how employees perform their work and constrain their options for
performing tasks (Caglio, 2003; Elmes et al., 2005; Ignatiadis & Nandhakumar, 2007a,b). The ERP system
may also, by encouraging employees to be more mindful of how their work processes link to work
processes elsewhere in the organization, lead employees to regulate their own actions to align them with
interests inscribed in the ERP system (Elmes et al., 2005). In contrast with members of upper
management, low level managers directly supervising workers do not appear to enjoy these benefits.
Instead, the implemented ERP system appears to lead these managers to feel threats to their positions and
an intensification of their work (Harley et al., 2006).
As noted above, the implemented ERP may have inscribed onto it rules about how work is to be
performed (Berente et al., 2008; Bob-Jones et al., 2008; Caglio, 2003; Elmes et al., 2005; Ignatiadis &
Nandhakumar, 2007a,b; Marabelli & Newell, 2009; Pollock & Cornford, 2004; Scott & Wagner, 2003;
Wagner et al., 2006). Nevertheless, groups within the organization may view the ERP system's fit with the
work practices of the organization differently. Some actors (e.g., users) may perceive the new system as
fitting poorly with these practices (Boonstra, 2006; Ignatiadis & Nandhakumar, 2009; Scott & Wagner,
2003; Wagner et al., 2006, 2010). That can very well be a source of resistance and conflict. In addition,
these actors may find their workloads increased when using the new system (Scott & Wagner, 2003;
Shepherd et al., 2009).
Employees who find the ERP system inadequate to perform their work may turn to alternative
techniques known as workarounds and shadow systems to avoid using the ERP system or to use it in a way
(unintended by management) that still makes it possible for the employees to complete their work
(Alvarez, 2008; Berente et al., 2008; Bob-Jones et al., 2008; Ignatiadis & Nandhakumar, 2007a,b, 2009;
Marabelli & Newell, 2009; Pollock & Cornford, 2004; Scott & Wagner, 2003; Wagner et al., 2006). However,
in some instances, management may limit users' options for taking such actions by cutting off their access
to needed legacy systems (Bob-Jones et al., 2008). In addition, these actors may turn to the use of other
human actors in the organization, who will use the systems in their place (Berente et al., 2008).
Users may also interact with the system in a way that does not contribute to organizational
productivity, but is meant to give the outward appearance of complying with the requirements of the
system (Berente, Gal, & Yoo, 2010). For example, researchers working at a government agency may enter
data into the ERP system in a way that demonstrates compliance with the dictates inscribed in the system
(Berente, et al., 2010). However, the entered data may not accurately reflect how these researchers
actually accomplish their work (Berente, et al., 2010). Thus, these studies point to management's use of
the implemented ERP system to constrain the productive activities of their organizations. However, these
managers are faced with the possibility that employees will avoid use of the ERP system or use it in
unintended ways, further altering how the work of the organization is performed and management's
capacity to exercise discipline.
While most of the studies referred to in this section enhance our understanding of power, they also
illustrate the limited focus of most ERP research on power. As we have shown, most research does not
focus on all three types of power. Instead, these studies mainly focus on limited subsets of issues of power
from the three circuits. This gap in existing work provides the foundation for our research objectives. On
the one hand, we will show empirically how the three circuits of power operate during the
implementation of an ERP. On the other, it is through the case's analysis that we are able to identify the
sources of disturbances (i.e., conflict and resistance) in the circuits of power of ERPs.
4. Research approach
To achieve our empirical and theoretical objectives, and to be consistent with the emphasis placed by
the circuits framework on meanings and intentions, we conducted an interpretive case study (Silva, 2007;
Taylor, 1971; Walsham, 1995, 2006). To validate our research we drew upon the set of principles proposed
by Klein and Myers (1999) (see Appendix A). The case's unit of analysis is the implementation of the ERP
system, ERPlus. 2 ERPlus is an information system designed to combine the different managerial functions
of the organization under study. We provide a more in-depth description of the functionality of ERPlus in
the case narrative and analysis below.
2
All names pertaining to the implementation are pseudonyms created as part of an assurance of confidentiality.
4.1. Research sites
The sites for this research were Alpha Campus and Beta Campus, two campuses within the Gulf Coast
University (GCU) system where ERPlus was implemented. We selected these sites for two reasons. First,
gaining access to research sites to conduct research on power relations is often problematic. For this
reason, Buchanan, Boddy, and McCalman (1988) recommend that researchers take a pragmatic approach
to securing field research sites. Access to the implementation sites was facilitated by the open offer by
members of management to allow access to the sites for research. Second, managers and users, in our
initial conversations with them, expressed their strong belief that power relations had an important
influence on the project's outcomes.
In addition, the GCU system's governance structure makes it a rich environment for examining power
relations related to this implementation. The richness of this environment is present in the complex
relationships among the organization's actors. For example, the board of regents is charged with, and
granted authority by state law, for making high level administrative decisions (e.g., approving continued
funding for an ERP implementation project). The chancellor, reporting to the board of regents, decides all
professional and administrative actions and is president of Alpha Campus as well. Vice chancellors manage
system-wide activities. Each of the remaining campuses is headed by a president who reports directly to
the chancellor. These presidents are largely independent in managing the affairs of their campuses, subject
to GCU system policies.
4.2. Data collection
Data collection took place at Alpha Campus and Beta Campus between September 2003 and May
2008. We conducted 42 semi-structured interviews with implementation stakeholders. The selection of
interviewees followed a snowballing approach with the intent to interview a wide variety of stakeholders.
When conducting the initial interviews, we used a list of topics derived from the Circuits framework as a
guide (see Appendix B for a list of the topics). We also conducted 28 follow-up interviews in order to
clarify responses from the initial interviews or to obtain answers to questions stemming from those earlier
interviews. On average, each initial interview lasted one hour and each follow-up interview lasted twenty
minutes.
Whenever the interviewees agreed, the interviews were tape recorded. When this was not possible,
written notes were taken. Transcripts were produced from both the tape recordings and the written notes.
In order to understand other actors' interpretations of events (Klein & Myers, 1999), we also obtained
relevant documents (e.g., project documentation) from informants (see Appendix C for a list of the
documents). After completing data collection, we proceeded on to the data analysis phase.
4.3. Data analysis
The analysis of data proceeded in four steps. The first step involved reading the interview transcripts
and documents in order to develop an initial understanding of the events, and the roles played by the
interviewees and other actors. This step also facilitated the incorporation of different interpretations of
these issues, enhancing the validity of the insights developed in later steps (Klein & Myers, 1999).
The second step consisted in developing an initial interpretation of the data. The Circuits framework
guided this interpretation process. Specifically, the framework facilitated the identification and
classification of the power struggles in the data. This interpretation allowed the development of tables
corresponding to each circuit with examples of supportive case data. These tables also provided a
foundation for performing the next step of the case analysis and are provided in summary form in
Appendix D.
Next, we prepared a case narrative, based upon our initial interpretation. In doing so, we emphasized
the sequence of critical events related to our case's unit of analysis. Our theoretical assumptions derived
from the Circuits of Power framework helped us to organize events and other narrative components. In
this case narrative, quotations from our informants not only provided factual support for our assertions
but also served as the basis for interpreting the case through Clegg's theoretical lens.
We presented this narrative to our informants to obtain their feedback. We revised the narrative,
incorporating the informants' feedback when appropriate, to clarify and validate our initial interpretation.
This allowed us to complete the Hermeneutic Circle, as advised by Klein and Myers. This approach, when
coupled with our reflective application of Klein and Myers' criteria, including especially those of suspicion
and multiple interpretations, also helped us to minimize the effects of our participation in the generation
of our findings. The problematic effects of such involvement stem from power relationships between
researchers and research participants that may favor the desired objectives of the former. This problem is
not unique to qualitative research but is a pervasive property of most empirical research involving humans
(Denzin & Lincoln, 1994).
The fourth step consisted of preparing the final case narrative and analysis of the case. While preparing
the final narrative, we used an iterative process to interpret it. This process entailed going back and forth
from theory to data, resulting in our analysis of the data in terms of three Circuits. We proffer the case
narrative and its analysis in the next section. The explanation that we provide of power's roles in ERP
implementations is presented mostly in the form of analytical generalizations, as they constitute
generalizations to theory (Lee & Baskerville, 2003; Walsham, 1995, 2006; Yin, 1994).
5. Case narrative and analysis
In this section we present the case narrative interwoven with its analysis. This approach allows us to
make clear the linkages between the case data and their interpretation. This section is organized
chronologically into three distinct episodes: (1) planning, (2) starting the implementation, and (3)
conflict and its attempted resolution. Each episode is followed by an analysis through the circuits of power.
Doing so permitted us to examine the disturbances as they developed in each of the circuits over time. As
suggested earlier, by disturbances, we mean incidents or conditions that make a circuit fail to attain
integration.
To reduce the complexity of the narrative and its analysis, we focus on the most salient power relation.
The two groups of actors in this power relation are the management group (comprised of the GCU
chancellor, executive sponsors for the project, the project team, the project manager, as well as GCU
employees and middle managers participating in the implementation) and the users group (comprised of
GCU employees in staff positions, their middle managers, and faculty members). They are the respective As
and Bs of the power relation that concerns us.
5.1. Episode 1: planning
The management group in charge of implementing ERPlus was assembled in 1997. Three vice
chancellors were selected to be executive sponsors. Each executive sponsor was a member of GCU's upper
management reporting directly to the chancellor and was granted authority by GCU bylaws to oversee
such projects. They were charged with devising and enacting the project's implementation strategy. The
executive sponsors turned their attention to staffing the project. To do so, they formed a project team. This
team was composed of GCU employees, middle managers and users. Team members with pre-existing
duties in the system (e.g., working in the Beta Campus registrar's office) were expected to continue to
perform these duties while devoting sufficient time for this project. A Beta Campus manager described the
effect of this practice on his department: “…I was basically working two full time jobs and we had four or
five people in the area who were doing the same thing.” This was a source of tension among users required
to work on the new project plus on their regular duties. With the formation of the new management
group, the executive sponsors turned their attention to the actual implementation of ERPlus.
The first step that the executive sponsors took was to define the implementation's strategy. The project
would follow a centrally-controlled approach to implement three ERPlus applications: Student
Administration, Human Resource Management System (HRMS), and Finance. The student administration
application would encompass modules to support student administration functions, including admissions,
student records, student financials, financial aid, and academic advisement. The HRMS application would
encompass modules supporting the functions of human resources, benefits administration, payroll, and
time and labor. Lastly, the finance application would consist of modules for general ledger, accounts
payable, purchasing, project costing, assets management, accounts receivable, billing, and inventory. The
Finance and HRMS applications of ERPlus would be implemented across all campuses while the student
administration application would initially be implemented only at Beta Campus. This campus was targeted
due to its small size, as well as due to concerns that this campus' existing system could not be modified to
address the Year 2000 (Y2K) problem.
The management and users groups varied in their views of ERPlus. For the management group, ERPlus
was critical to complying with the board of regents' mandate, enhancing management's control of the GCU
system, and addressing GCU's future needs. The board had charged the GCU system administration with
implementing an integrated information system linking together GCU's business processes to enable more
complete and accurate reporting on the status of the GCU campuses. These capabilities would allow GCU
to better respond to reporting requirements of government agencies. To management, these capabilities
would also allow more accurate control over GCU's business processes and make it possible to meet the
GCU system's future needs. By contrast, members of the users group did not share the management
group's view. One Beta Campus faculty member noted, “To me it's [ERPlus] pretty senseless…the system
that we had was working fine.” These differing views of ERPlus would be a continuing source of conflict
between these two groups during the project.
The implementation project was to be completed under tight budgetary and time restrictions. A key
tactic intended to comply with these restrictions entailed following a “plain vanilla” implementation
approach. The leading executive sponsor described this approach: “The principle that we have applied
from the very beginning is we want to have a vanilla, as vanilla as possible system. In other words, don't go
out and just make all sorts of modifications.” In this case, modifications meant making actual changes to
code comprising the software. Following this approach, such customization of the software was to be
minimized. The executive sponsors favored this approach as they believed that it would facilitate upgrades
to ERPlus and, more importantly, it would lower implementation and maintenance costs. Nevertheless,
management could not adhere completely to the plain vanilla approach.
An example of an exception to the plain vanilla approach addressed the inability of ERPlus to meet
state regulatory requirements. ERPlus was designed by the vendor, ERPCorp, to meet federal regulations.
State regulatory requirements were another matter. A member of the management group described an
instance of this issue:
…we turn in a series of several, but three main reports, every semester to [our state]…Those are
mandated reports. They are the data files off of which our funding is based. So any funds that we
appropriate from the state are based on those. The amounts are calculated based on data in those
files. They really need to be right. ERPlus doesn't deliver that because it's [our state's] thing.
Regulations of GCU's home state could not be met by a vanilla ERPlus. Thus, the management group
had to make modifications to the software to comply with these requirements.
Prior to starting the implementation, the management group conducted fit-gap sessions. The purpose
of these sessions was two-fold. First, as described by a project team member, “The fit-gap sessions were to
get as many future users as familiar with ERPlus as possible.” Second, the sessions were intended to
compare the functionality of ERPlus with that of GCU's legacy information systems. Through the fit-gap
process, a list of the ways that ERPlus matched and failed to match the existing systems' functionality
would be developed. These sessions revealed many gaps, as described by a retired Alpha Campus manager
who participated: “…there's about 80% of the functions that we had that [those conducting the sessions]
could not say, yes, we can deliver it [with ERPlus].”
This inability of ERPlus to provide much of the existing systems' functionality concerned users. Users
believed that these gaps, if not addressed, would interfere with their ability to do their work. In response,
the users called for the gaps to be addressed by modifying ERPlus. The management group did not share
the users' view of the situation. The management group rejected the idea of addressing all gaps with
changes to ERPlus. The leading executive sponsor argued “If you go out there with all of those gaps and say
we're going to meet every one of them, that's where you're going to make a mistake.” Instead of making a
large number of modifications to address gaps, the management group encouraged users to change their
business processes. The outcome of the fit-gap sessions, coupled with management's response, served to
undermine users' confidence in the capabilities of ERPlus. The completion of the fit gap sessions marked
the end of the preliminary steps to be taken before implementing ERP-Plus.
5.1.1. Analysis of the circuits during episode 1

In the first episode, initial conditions were set that contributed to later disturbances in each circuit.
From the perspective of the episodic circuit we can appreciate that early in the planning stages, the
management and users groups' interests stood opposed to one another. Management sought to enhance
its control over GCU's operations, respond to external reporting demands, and accomplish these goals in a
timely, cost-effective manner. In contrast, the users were opposed to being forced to be involved in
projects that would result in them losing capabilities to do their work as faculty and staff. It was these
interests that provided the directions for the actions that members of these groups took during the
implementation.
Aspects of power from the social integration circuit (i.e., rules of meaning and membership)
influenced the overt elements of power during the first episode. Management and users interpreted the
state of affairs and one another through particular rules of meaning. The management group viewed the
situation through its concern for enhancing control, reporting to others, and meeting future needs. The
perspective of the users was quite different. How the users interpreted the situation was shaped by two
factors. The first was users' roles in GCU's mission as a higher educational system. The second was users'
belief that participation in projects should be voluntary and that they should have a voice in these
projects. Thanks to these rules, management and users viewed ERPlus and its implementation differently
from the beginning. For management, ERPlus was necessary to meet GCU's needs. The users saw ERPlus
as unnecessary. This was because the higher educational mission of GCU was served adequately by
existing information systems. For management, the involvement of users was necessary and could be
mandated. Users objected to their forced involvement. Over time, these groups' interpretations would
grow more divergent.
The disciplinary and productive aspects of power from the systemic integration circuit were also
salient during this first episode. From the beginning, the management employed disciplinary techniques
to constrain users' actions. The management group controlled the implementation centrally. In this way,
they limited other actors' discretion to make decisions about implementing ERPlus. Furthermore, ERPlus
was itself being implemented as a disciplinary tool. This new system was implemented, in part, so that
management could enhance its control over GCU's operations. Control was to be enhanced by constraining
how work was done and by enabling enhanced monitoring of operations.
While management sought to enhance their control over GCU, users wanted to be able to do their
work. Users were concerned about the ability of ERPlus to support their work processes. This concern
began during this first episode, when users attended fit gap sessions. These sessions exposed users to the
capabilities of ERPlus. Based upon these sessions, users regarded ERPlus to be limited in its ability to meet
the requirements of their work. Thus, before the implementation, users already had concerns about the fit
of ERPlus with GCU work processes. These disciplinary and productive aspects of power, together with
those considered previously, provided the foundation for changes in the circuits during the next episode.
We consider that episode next.
5.2. Episode 2: starting the implementation
After the initial preparations were completed, the implementation began. Management used multiple
techniques during the project. First, they used a participation program to obtain the users' cooperation.
Management believed that completing the implementation would depend heavily on GCU employees'
involvement and cooperation, with an executive sponsor noting: “Well, no system is going to be successful
unless you have user buy-in…if the customers are not somehow supportive, you are going to have major
problems.” From management's perspective, users needed to be involved both in working on the project
and by using ERPlus. However, the communication program did not go as planned. Users objected strongly
to what they saw as their forced participation in the project. One manager from Beta Campus voiced his
objection thusly: “GCU has never done anything but shove it down the throat of their current employees.
That's the only way that they've ever done a conversion…It just sort of fell on the backs of the current
people employed.”
To complement the program for user participation, the management group undertook a program of
communication. One member of management described this program:
We had a ‘traveling road show’ to make sure everyone was on the same sheet of music. This meant
going to each of the different campuses to talk with people. Employees have to be brought up to
speed on the peculiarities of each campus. It was to keep harmony and keep everybody informed…
Our team met with senior department heads on each campus to give them briefings, goals,
timelines, etc.
Although users were permitted to give their input at meetings, management was unwilling to negotiate
on many system features. This led to two sharply different views of the communication between users and
management. Management suggested that this program was one of open, two-way communication with the
users to gain their cooperation. In sharp contrast, members of the users group believed that the
communication was one-way, in which they were told what would happen and their input was ignored.
Training of users was an additional technique used by management. Members of the users group objected to
the training on two grounds. First, they objected to the training's timing. One Beta Campus dean contended:
“The training was done before the system was online; by the time we went live, we forgot things….” Second,
users objected to the policy requiring that they first take courses on basic features of ERPlus. Only after
completing this training could users take training more closely related to their jobs. These users wanted to be
able to take only job-related training, arguing that no more was necessary. A Beta Campus faculty member
noted “There's about 3 to 4 things a faculty member needs to know. And that's all.” For these reasons, users
believed that management understood their training needs poorly.
In contrast, management argued that problems with training were based upon users' misunderstanding.
They claimed that some users needed far more training than they had initially believed. The project manager
contended that “Later they [the users] found that the training they deem unnecessary for the jobs was
needed.” Basic training courses, argued members of management, were necessary to prevent misuse of ERPlus
stemming from a lack of understanding of the new system. In addition, some users had also misunderstood the
training policy. This misunderstanding had led these users to take unnecessary training courses.

Disturbances in the circuit of episodic power that started during the planning phase grew during the
next phase of the implementation. Management continued to draw upon its standing conditions to further
its interests and continued to exercise causal power to force users to work on the implementation.
Likewise, the management group also sought to gain users' support for the implementation and to get
them to use ERPlus. These ends were to be accomplished through the programs of participation,
communication and training. These were clear instances of episodic power to make employees to use the
system; that impacted the other circuits of power.
In relation to the circuit of social integration (rules of meaning and membership), users and
management interpreted the techniques used by the other group differently. Management saw the
techniques that it used as necessary to gain users' support and/or to prepare them to use ERPlus. Consider
the example of users' training. For users, this training was poorly timed and led them to believe that
management did not understand their training needs. In this way, users interpreted this and other
programs as contrary to their interests. Management dismissed users' views as being based on their
misunderstanding of what training was actually needed. Thus, the disturbances in the circuit of social
integration arose from contradicting interpretations of discourses.
Disturbances in the circuit of systemic integration (techniques of production and discipline) were
prominent in the discrepancies about the training program. Training was key for the systemic integration
of the ERP; however, both groups, managers and users, came to believe that the other group understood
poorly what the training should accomplish and their roles in it. In this way, each group saw the others'
interests as different from their own and the boundary between them grew more pronounced. These
rising differences between the two groups continued during the final episode. In these ways, conditions
for unresolved power struggles continued to develop during the second episode.
5.3. Episode 3: conflict and its attempted resolution
The implementation entered a period of increasing conflicts. Users' dissatisfaction with management
and ERPlus grew, reaching a peak in April 2001. With the support of their campus president and vital
knowledge of their campus business processes, users at Beta Campus brought a halt to the student
administration application implementation. They issued an ultimatum to management that the project
would not continue on their campus until their list of concerns was addressed. These concerns included a
set of incomplete items relating to the prospects/admissions component of the student administration
application, the lack of an adequate project plan, problems with management, quality assurance, sign-off
practices, relationships with the management group and an implementation schedule that was
incompatible with student registration periods.
Faced with the indefinite stoppage of the student administration application implementation, the
project team took the list of concerns and addressed them. Management's response to the users' concerns
also carried with it a warning. A project team member recalled a letter sent to the Beta Campus users: “…
the chancellor sent them a letter saying that ‘The go-live will be rescheduled, but that we will go forward.
Also, don't do that again’.” In addition to contributing to rising tensions between management and users,
the ultimatum was costly. The implementation was delayed by five months to November 2001, and
consumed two thirds of the original budget for the system-wide student administration application
implementation. For the users, the ultimatum was necessary to address serious problems that had to be
resolved if the implementation was to continue. From the management group's perspective, the
ultimatum was a costly, and unnecessary, interruption.
Relations between management and users did not improve after the ultimatum. By late 2001, the first
versions of ERPlus had been implemented. Many staff and managers had opportunities to use ERPlus.
ERPlus placed constraints on work processes that users found to be cumbersome. A Beta Campus manager
described his difficulty: “It's not as easy as the old system. It takes longer to do everything. You can't do
anything the way that we used to.” A particularly problematic element of ERPlus was a feature for
canceling students' enrollment. A Beta Campus employee described this feature's inconsistent functioning:
“One day the enrollment cancelation process works. The next day it doesn't…We never know from one
day to the next if it will even run — much less run correctly.” Problems with reports produced by this
feature required Beta Campus staff to spend considerable time correcting errors manually.
Users responded to what they saw as the new system's inadequacies in varied ways. Some users found
they had to develop “creative” approaches for using ERPlus. One faculty member noted: “The big problem
with ERPlus…is [that it is] so rigid that it makes liars out of us all. We kind of had to figure out how to
schedule our classes.” These faculty members had to provide ERPlus with numbers of students greater
than the actual enrollment numbers. Otherwise, ERPlus would not allow them to reserve classrooms with
facilities required for teaching. Users also responded to the new systems' limitations by making requests
for additional customization and turning to programmers within their departments to write queries or
‘workarounds’ to obtain information that they could not otherwise get using ERPlus. Workarounds were
used so frequently that one Beta Campus manager remarked: “Workarounds have become our life.”
Instead of responding to ERPlus in these ways, some users chose to resist it differently. Some resistance
was open. For example, the project manager recalled: “One woman in human resources quit.” Other users,
in middle management positions, appear to have attempted to shape their subordinates' beliefs about the
capabilities of ERPlus. A project team member described these attempts:
…you had people, some people who were trainers and some people who were even directors would
tell the users that it wasn't a good system and that it doesn't work…When people who are in a
position of influence say things such as this, they can poison the opinions of other users…
Users also resisted in more subtle ways. Some users employed shadow systems (i.e., using the old
processes and other stores of needed data) to avoid using ERPlus, apparently in the belief that they could
only perform their job functions if they avoided using it. Management interpreted this resistance as the
result of the users' dislike of changes that they were called upon to make in their work tasks.
Management group members responded to users' actions that they believed threatened the project.
Members of the users group argued that management group members unsuccessfully attempted to paint
more a favorable picture of the implementation than was the case in reality. A retired Alpha Campus
manager pointed to a practice used by the project manager when making reports to project stakeholders.
The project manager grouped issues together to try to make the number of problems with ERPlus seem
smaller. Still seeking to follow the plain vanilla implementation approach, the executive sponsors used an
administrative systems steering committee chaired by one of them to limit additional, extensive
customization to ERPlus. Granted authority by the executive sponsors, this committee decided whether to
approve customization requests. Users who had their requests denied and could not obtain solutions
through other means (e.g., workarounds) had little choice but to change their business processes.
This response by management divided users into two groups. Users who had their requests for
customization granted viewed ERPlus more favorably. Those who had their requests denied grew more
displeased. Faced with mounting resistance, the management group retaliated. Some users were re-
assigned to new positions. Others were fired. User actions considered threatening by management and the
latter group's responses to them greatly increased the rift between the management and users groups
during the implementation.
By late 2002, the board of regents had grown concerned about the project status. The executive
sponsors informed the board about the financial status of the project. Many more millions of dollars would
be required to complete the project, due to the budget overrun from the student application
implementation. In addition, the board was concerned about the new system's ability to deliver its
promised benefits. The Alpha Campus faculty senate president, after attending a board meeting, noted
“The regents expressed a certain concern and a desire to clearly see the benefits of the new system.”
Moved by this concern, the board ordered an in-depth review of the project to determine whether it
should continue to receive funding. The management group was ordered to make a detailed report on the
project's status.
Comments by users suggest that members of management responded to the board's order in a way
intended to ensure the board's continued support of the implementation. A Beta Campus manager
required to participate in making this report described its planned purpose: “The decision to stay with
ERPlus had already been made [by the chancellor]. No one was going to bash it. Too much money had
already been spent.” Accomplishing this purpose entailed silencing users and excluding technology that
might influence the regents to develop an unfavorable view of the implementation. Any users with
negative views of ERPlus and its implementation were prevented from voicing them. One Beta Campus
faculty member stated:
…apparently, they came down one time to do a kind of a hearing…and we were invited and then
we were uninvited. And so, I don't know the politics of all of that… So I don't think there was ever,
as to my knowledge, a time at which we could all talk about the problems that we have had.
Through this means, users critical of the project were prevented from voicing their displeasure.
Elements of ERPlus that might interfere with the management group's efforts to garner the board's
continued support were also prevented from doing this. Management's report to the board was supposed
to highlight the benefits of ERPlus through demonstrating the applications. At that time, there were
elements of ERPlus (e.g., the enrollment cancelation feature described previously) that could not be
counted on to work reliably. These elements of ERPlus would not be demonstrated to the board. In their
place, elements of the new system that worked reliably (e.g., a feature allowing faculty members to post
grades) were scheduled to be included in the presentation. In this way, unreliable elements of ERPlus were
prevented from compromising the management group's presentation.
In January 2003, what one executive sponsor described as a “cast” of over 30 individuals made a report
to the board of regents. Those making the presentation included both members of the management group,
as well as members of the users group who were required to participate. This presentation was
overwhelmingly positive in its tone. The presentation proclaimed the benefits brought by the
implemented ERPlus applications. Some presenters made references to problems with ERPlus and its
implementation. However, these issues were reported as having been addressed or were treated as
relatively unimportant to the implementation's success.
This presentation appeared to have two effects for the implementation project. First, it altered the
board members' interpretation of the implementation. The nature of this change in the board's
interpretation was contested. The leading executive sponsor noted that: “…they came away, saying:
‘Thank you. This was very helpful.’ So, they [the board of regents] had a more complete understanding of
this [the implementation] than they had received in the past…” Some members of the users group,
though, did not share this view that the board developed a more balanced view of the project status.
In contrast, these users emphasized the carefully-engineered nature of the presentation to the board. A
Beta Campus manager required to participate in the presentation recalled how management edited the
presentation to focus on positive aspects of the implementation: “…basically, my [part of the] presentation
was screened before I presented it.” Additional users, who did not attend the event, developed similar
views of the presentation's nature. A retired Alpha Campus manager claimed that: “…it was a ‘dog and
pony show,’ one in which the regents were given the clear impression that all was right and good with
ERPlus.”
While the effect of the presentation on the board of regents' understanding of the implementation was
disputed, the second effect for the project was not. Having a more favorable interpretation of the
implementation, the board of regents approved the continuation of funding for the project. This key point
in the ERPlus implementation marks the end of our narrative. Our analysis of this third episode that
follows concludes this section.

Power struggles manifested in the episodic circuit culminated during the third phase, remaining
largely unresolved at the end. To begin, conflict grew more intense during this stage than it had been
previously. Management continued to take actions to achieve their objectives. Disturbances also grew
more intense along the lines of the circuit of social integration during this episode. As noted previously,
both members of the management and users groups attempted to engineer other actors' interpretations of
the implementation during this period. These attempts were successful in terms of meeting their
objectives in some instances. As was the case before, these actions were interpreted differently by these
groups, serving to increase divisions between them.
In terms of the circuit of systemic integration, use of disciplinary techniques and effects on work
processes also figured prominently during this episode. Both groups employed disciplinary techniques.
Concerned about problems with the implementation, users issued an ultimatum that served to halt
continued work on the implementation at Beta Campus until management took steps to address these
users' concerns. An interesting example of discipline occurred at a pivotal point of the implementation in
2003. As noted above, management was required by the board of regents to account for the status of the
implementation. The executive sponsors were faced with the possibility of losing the board's support for
the project.
Preparing a presentation sufficiently persuasive to regain the regents' support required that both
humans and technology be restrained. Disciplining users entailed preventing some from expressing their
negative views of ERPlus. Other users were required to participate in making the report. Their
presentations were carefully screened to ensure that they conveyed a message favorable to the
implementation. Technology also had to be disciplined. Elements of ERPlus that could not be counted on
to convey a positive picture of the new system by working reliably were also excluded from the
presentation. In their place, components that functioned reliably were demonstrated to the regents. Thus,
regaining the support of the board of regents required that the management group extend its use of
discipline by preventing the use of technology that, by malfunctioning, could interfere with efforts to
influence the former group.
As noted previously, management implemented ERP-Plus as a disciplinary tool for increasing control
over GCU operations, but users were concerned about the threat to their ability to do their work that
ERPlus posed. Users' concerns grew after they gained first-hand experience with using ERPlus. As
management intended, ERPlus required users to change how they did their work. Users found these
constraints imposed by ERPlus to be poorly aligned with their work. Adding to the problem was the
unreliable functioning of some ERPlus components that required users to clean up outputs of the new
system. Overall, these users found ERPlus to be an unsatisfactory replacement to existing information
systems for helping them to do their work. In response to the inadequate performance of ERPlus, users
turned to other means to do their work.
Disturbances in the circuit of systemic integration, in the form of disturbances to workflow, impacted
social integration as unfavorable interpretations of the ERP grew. To address disturbances in workflow and
production (systemic integration), users drew upon varied means to do their work. Users frequently asked
programmers in their departments to develop workarounds to gain needed information, provided ERPlus
with false information to get the desired results, and employed shadow systems to avoid using the new
system. Other users requested additional customizations to ERPlus to meet their work requirements.
Management used the administrative systems steering committee as a disciplinary tool to limit what it
saw as unnecessary customizations of ERPlus. This practice had the side effect of dividing users between
satisfied users who were able to gain changes that allowed them to do their work, and dissatisfied users
who were forced to change their work processes. This way, the disturbances in systemic integration
affected the circuit of social integration by forming two groups; with the latter being prone to exhibit more
resistance toward the ERP.
At the end of the third episode of the ERPlus implementation, disturbances in the circuits remained
largely unresolved. Overt conflict, in the form of power exercises by management and resistance by users,
did not come to an end. Social integration also failed to occur, with the two groups growing increasingly
divided along the lines of membership and their views of the implementation. Lastly, systemic integration
also did not occur. The use of discipline, and the implementation of an ERP system that many users saw as
poorly-aligned with work at GCU, continued.
6. Summary of results and discussion
In this section we take a more general view of the disturbances in each of the circuits (i.e., incidents or
conditions that make a circuit fail to attain integration) and discuss their relation to previous research on
ERPs. This type of analysis allowed us to develop a further theorization of the power struggles that may
arise while implementing an ERP system (see Table 1). The identification of disturbances is relevant from
the organization's point of view as they would prevent the integration of the circuits, and therefore would
hinder implementation.
6.1. Disturbances in the circuit of social integration
Disturbances in the circuit of social integration are characterized by antagonistic interpretations. The
acknowledgement of instances of antagonistic interpretations (e.g., when managers and users held
opposing and even hostile views towards an information system) is not new in the studies of power and IS
implementation (see for example, Franz & Robey, 1984; Markus, 1983; Markus & Björn-Andersen, 1987).
In our case, we found a clear and intense instance of antagonistic interpretations held by managers and
users. While the management group interpreted the system as fundamental for controlling operations and
improving overall organizational performance, the users group qualified the system as an unnecessary tool
that was pushed to them by the management group.
Instances of antagonistic interpretations between users and managers have been widely documented
in the ERP literature (see for example, Allen et al., 2002; Harley et al., 2006; Lim et al., 2005; Pollock &
Cornford, 2004; Scott & Wagner, 2003; Shepherd et al., 2009; Swan & Scarbrough, 2005; Wagner et al.,
2006; Whittaker & Beswick, 2008). In addition to confirming those studies, our research demonstrates that
antagonistic interpretations may occur not only towards the ERP itself but also regarding other groups'
actions. In this sense, our data shows that while the users group disapproved of management's actions, the
management group regarded those by the users as uncooperative and obstructive. Moreover, we found
that antagonistic interpretations intensify with time. This occurs when the ongoing actions of a group
reinforce the opposing group's beliefs toward the system. In our case, antagonistic interpretations got
reinforced by almost every action taken by the opposite group. Typical examples of these situations were
the way training was offered by the management group and the maneuvers made by the users group to air
their discontent, namely those that led to producing the ultimatum. We found that, once exacerbated,
Table 1
Disturbances in the three circuits.
Circuit of power Conditions for disturbance Consequences for disturbance Manifestation in episodic circuit
Social integration Contradicting interpretations Disposition to not cooperate Bypassing

Systemic integration Lack of fit of system with practices Increased workloads Workarounds
antagonistic interpretations engender one group's disposition to not cooperate with the other, which in
turn translates into resistance.
We found that the antagonistic interpretations can be recognized in a specific discourse drawn on by
both managers and users, called discursive valuations. Discursive valuations are statements that qualify
the value and utility of the system. In our case, we found that management's valuation consisted mainly
in minimizing the problems of the system and at the same time in maximizing its advantages. By
contrast, the users group held an opposite valuation; one that maximized problems and minimized
virtues. These valuations had a reciprocal relation as our data shows how actions and discourses of one
group would reinforce the opposite valuation of the other. Moreover, as suggested above, we found that
antagonistic discursive valuations lead to a group's disposition to not cooperate with the other. This was
more evident in the users' group. Thus, antagonistic discursive valuations were a major source of
disturbances in the circuit of social integration as they contributed to polarizing the groups and
intensifying overall hostility.
The ERP literature confirms this relationship between antagonistic interpretations and resistance. After
reviewing the literature we identify at least five types of contrary interpretations that are related to
resistance. Accordingly, we found that users may tend to resist ERPs when they: (a) perceive as threats to
their interests (Allen et al., 2002; Markus et al., 2000), (b) differ with management in their understanding
of the new system's benefits (Lim et al., 2005), (c) believe that the new system will require that they
perform their work differently (Legare, 2002; Lim et al., 2005; Markus et al., 2000; Pollock & Cornford,
2004; Sheu, Chae, & Yang, 2004; Wagner & Newell, 2006); (d) believe the system will engender losing
their jobs (Lim et al., 2005; Scott, 1999), or (e) believe that management does not consider the
implementation important (Newman & Zhao, 2008). This typology is not intended to be exhaustive as we
could add antagonistic discursive valuations to it. However, it shows empirically – as we confirmed in our
case as well – that in general contrary interpretations tend to be followed by resistance. Our contribution
to this body of knowledge consists in providing a theoretical explanation that integrates all these studies.
We do that by conceptually linking resistance, manifested in the circuit of episodic power, as originated
from disturbances in the social circuit of power.
6.2. Disturbances in the circuit of systemic integration
Disturbances in this circuit are characterized by changes in techniques of production and discipline that
affect the efficiency of the organization. As mentioned in the literature section, the lack of fit between
procedures inscribed onto the system and current organizational practices is a major source of disruptions
in this circuit (see for example Boonstra, 2006; Ignatiadis & Nandhakumar, 2009; Scott & Wagner, 2003;
Wagner et al., 2006, 2010). In addition to confirming those studies, we found that a disruption in the
systemic circuit is followed by an upsurge in the users' workload. This coincides with what Scott and
Wagner (2003) and Shepherd et al. (2009) found in their studies. An increase in users' workload can be
worsened, as it was the case in the ERPlus implementation, when improper functioning of the system
resulted in additional work by users to clean up system errors. Our data show that these disruptions in the
systemic circuit not only engendered resistance from users but also reinforced antagonistic
interpretations.
We argue, then, that disturbances in the systemic circuit tend to generate conditions for disturbances
in the circuit of social integration as antagonistic interpretations are reinforced. We found that among the
main sources of antagonistic interpretations were the lack of fit between actual work tasks and those
inscribed in the system, as well as increased workloads. Once the lack of fit was realized and workloads
were increased, the previous uncooperative disposition that users had toward the system got
strengthened. The circular theorization of power as proposed by Clegg and our own data show that the
disposition to not cooperate – as part of the circuit of social integration – influences the users' belief that
the system does not fit with current work practices (see Fig. 2). The dotted line in the figure denotes that
the disposition to not cooperate is not a direct reason for the lack of fit (the specific design of the system
would be a direct reason) but suggests a predisposition on behalf of the users to be less tolerant with
perceived lack of fitness. In other words, users with a non-cooperative disposition will tend to realize and
complain more about the lack of fit of the system than those that would have a more cooperative attitude.
Fig. 2. The interplay between social and systemic integration.
Thus, disturbances in the circuits of social and systemic integration will impact each other and eventually
will be manifested in the episodic circuit as instances of resistance.
As mentioned above, extant ERP implementation research has already considered issues related to
poor fit of the ERP system (e.g., Boonstra, 2006; Ignatiadis & Nandhakumar, 2009), increased workloads
(e.g., Scott & Wagner, 2003; Shepherd et al., 2009), and use of workarounds (e.g., Alvarez, 2008; Ignatiadis
& Nandhakumar, 2007b). Existing research, however, stops short of providing a theoretical explanation of
the relationship between these issues and changes in the social integration and episodic circuit. Our
contribution here consists in theoretically relating technical disruptions with social ones.
6.3. Disturbances in the episodic circuit
Disturbances in this circuit are grounded in the other two and are manifested by actions. These actions
are often related to what IS researchers would refer to as resistance. The ERP literature identifies different
ways in which resistance is manifested. For example, users may resist the implementation of an ERP
system by failing to do their jobs (Scott, 1999), avoiding using the ERP system or using it minimally
(Ignatiadis & Nandhakumar, 2007a,b; Lim et al., 2005; Newman & Zhao, 2008; Pollock & Cornford, 2004;
Scott & Wagner, 2003; Wagner & Newell, 2006), damaging property belonging to their employer (Scott,
1999), failing to provide those implementing the ERP system with useful information (Volkoff, 1999), or
threatening to resign from their positions (Sheu et al., 2004). In these ways, users can impede the progress
of the implementation. In our study, we found two tactics related to the circuit of episodic power:
bypassing and workarounds. The first could be classified as overt resistance while the second aims at fixing
the circuit of systemic power.
Bypassing may occur when the Bs of the power relation find themselves in a situation of intense
antagonism. Our case shows that in those situations the weak group (the Bs) will bypass the authority of
the strong group and communicate their interpretations of the system to higher authorities. This is done
with the purpose of undermining the standing conditions of the more powerful group while enhancing
their own. In the ERPlus implementation, the weak group (i.e., the users) communicated with their
campus president, bypassing the authority of the strong group, management. This was done with the
purpose of undermining management's authority. Our findings here are consistent with those of
previous studies (Allen et al., 2002; Howcroft & Light, 2002; Scott & Wagner, 2003; Wagner et al., 2006).
Our contribution, though, consists in conceptualizing a specific type of resistance, bypassing, and by
identifying the context in which it may occur as well as its relationship to social integration. We argue,
then, that bypassing will tend to occur in a context similar to the one of our case in which there are higher
authorities and there is a climate in which dialogue and communication among different hierarchical
levels are allowed.
Unlike bypassing, which aims at social integration (meanings and membership), workarounds are
directed to systemic integration. Workarounds are episodic power instances that disturb systemic
integration as they constitute a divergence with one of management's main objectives, imposing tight
controls on operations. Extant research (e.g., Ignatiadis & Nandhakumar, 2007a,b) finds the use of
workarounds as a means of resisting the implementation of an ERP system. Workarounds are often
unproductive for the organization as the use of the ERP system in ways unintended by management can
compromise the capability of the system as a tool for production and discipline (Berente et al., 2008, 2010;
Ignatiadis & Nandhakumar, 2007a,b, 2009). This capability may also be compromised when users are able
to evade monitoring of their work through the ERP system (Elmes et al., 2005). In addition, we found that
workarounds reinforce unfavorable beliefs about the ERP system, which as we discussed before, brings
about resistance. Thus, our contribution here consists in conceptualizing workarounds as a response to
disturbances in the circuit of systemic integration and identifying their relation to interpretations and
group membership (social integration). Table 1 below summarizes the effects that disturbances in the
circuits of social and systemic integration will have on the episodic circuit.
7. Conclusion
Our study's findings have implications for both research and practice. As noted previously, our work
demonstrates the value of the Circuits of Power framework both as a tool for categorizing and integrating
the diverse findings of existing research on power and as a tool for analyzing power issues. Nevertheless, it
may be worth reflecting on the limitations of our theoretical lens. One of the main issues is the Circuits of
Power focus. The Circuits framework concentrates on explaining integration as a result of power, yet the
system we studied did not reach integration. However, although not for explaining integration, the
Circuits' conceptual richness helped us in structuring and conceptualizing the case. It was fundamental in
identifying the circumstances and context in which power disruptions occurred; which was one of the
objectives of our research.
Even though it may seem dissonant to apply a theoretical lens that focuses on integration to explain
disruptions, it is important to emphasize that we do not deem integration and lack of it as a dichotomy but
as a spectrum, with low integration on one extreme and high integration on the other. In order to
understand such a continuum we need to acknowledge not only factors that favor integration but also
those that hinder it; hence our contribution in theorizing about disruptions. We do not claim, however,
that we have made a general contribution to the theory, but a very specific one, which circumscribes itself
to ERP implementations. It may be a matter of further research to establish whether the proposed
conditions for disruptions hold for other types of systems and in other contexts.
It is not only researchers studying power can who benefit from our research, as our findings also carry
practical implications. Both managers seeking to implement ERP systems and other stakeholders may
benefit from our findings. By reading this study, project managers and others championing an ERP
implementation may be able to gain insights into the nature of power struggles in their own
implementations. Our analysis of disturbances in the circuits may be particularly valuable in this regard.
This is because considering how disturbances arise may allow these managers to better anticipate how
users may react to the implementation and how relations with users may worsen over time. Future users
of a new ERP system may also benefit. By considering our findings, these future users may be able to
better plan for effects of the new system on their work and the consequences of resisting the new system.
In these ways, our findings may prove to be valuable guides for stakeholders of ERP implementations.
Though this research contributes to researchers studying power in ERP implementations, it also
possesses limitations that offer opportunities for further study. One limitation of this research is that the
case's timeline ended with the board of regents' decision to continue supporting the ERPlus
implementation. It would have been interesting to consider the long term effects of events in the case,
such as the users' ultimatum, on the still ongoing implementation. In addition, we limited ourselves to one
type of organization because the scope of our research: a large public university. As suggested above, one
area of further research would be to investigate the nature of disturbances in the Circuits of Power in other
kinds of organizations, such as multinational corporations. Such research may lead to additional, useful
applications of Clegg's framework for understanding the complexity of power in these other settings.
Researchers interested in this issue can build future research upon the foundation that we have
strengthened with our contributions to this framework in the context of ERPs. All in all, researchers,
stakeholders of these projects, including managers and users, can draw upon our work to get a better
appreciation for how power struggles can emerge and intensify during ERP implementations.
Appendix A. Application of Klein and Myers' validation criteria to this case
Klein and Myers' criteria Application to this study
1. The fundamental principle of the hermeneutic circle: The researchers developed an initial narrative and presented it to
Researchers conducting case research in the informants. The first researcher revised this initial narrative by
hermeneutic interpretive tradition should engage in a incorporating informant feedback when appropriate. This allowed
process based on the principle that all human the researchers to clarify and validate the relationship between the
understanding is achieved by the act of iterating initial theoretical framework and the data, through revising language
between considering the interdependent parts of a and names and eliminating confusion about concepts that were
whole and that whole. emerging from the analysis.
2. Contextualization: The historical and social contexts, including those relating to the
For interpretations to make sense, they must account emergence of the implementation project, are incorporated into the
for the social and historical context of the case. study's case narrative and analysis.
The researchers were not actively involved in the implementation
project, but are a faculty member at and a graduate of Gulf Coast
University's Alpha Campus.
3. Interaction between the researcher and subjects: The researchers selected documentary materials that they believed
The informants play an active role by offering their would complement the data of the interview transcripts.
interpretations of events. The questions asked by the researchers may have introduced the
The researcher must critically reflect on the way in informants to power and political concepts, shaping to some degree
which the data were socially constructed as a result the responses they provided.
of the informants' interactions with the researcher.
4. Abstraction and generalization: Theoretical concepts were utilized in the field work through the use
This involves generalizing particulars to abstract of an interview guide. They are also reflected in the analysis and
categories and social theories. discussion.
These general concepts describe human The results of the fieldwork provide illustrations of the theoretical
understanding and social action. concepts.
The application of the framework facilitated the development of a
theoretical explanation of the roles of power and in the context
of an
ERP implementation.
5. Dialogical reasoning: The researchers revised their conceptions of the three forms of power,
This requires attention to possible contradictions relationships between them, and the ways in which power relations
between the theoretical concepts employed in the allow actors to obtain desired objectives. This enabled the researchers
study and the actual findings. to develop an explanation of the roles of disturbances in the circuits in
the rise and growth of power struggles in an ERP implementation.
6. Multiple interpretations: The framework used by this study provides explanations for why
This requires that the researcher be sensitive to stakeholders have different interpretations. They are posited to result
differences in interpretations of events by informants. from differences in stakeholder identities, group membership and
vested interests.
The present study utilized findings from different classes of informants,
complementing the interviews with supplementary data, including
official documents produced by the administrators at the
member universities.
7. Suspicion: This research examines the different interpretations of informants
The researcher must be sensitive to informants' that reflected their particular desired objectives in the implementation
biases and systematic distortions of events stemming (e.g., those associated with being a member of the implementation
from their desired objectives and positions within team).
the organizational context.
Appendix B. Themes for semi-structured interviews
- A (or B)'s identity in the power relations.

- A (or B)'s desired objectives
- Disagreement over implementation objectives
- A (or B)'s strategies and tactics in the implementation
- Resistance
- A (or B)'s standing condition in the organization
- Resources
- The organizational roles and norms that determine A (or B)'s position
- The meanings of the implementation
- Rules and modalities used to make sense of the system and/or implementation
- Techniques to assure compliance

- Work tasks affected by the implementation
- Resources to instill discipline
- Use of persuasion
- External events or other forces influencing decision making related to the implementation
Appendix C. List of documents used as secondary case data
1. ERPlus Project History.

2. GCU-Beta Campus 2003–2004 Catalog.
3. Bylaws, board of regents, Gulf Coast University system.
4. Profile of the Gulf Coast University system universities, downloaded from the GCU system web site.
5. GCU ERPlus implementation documentation, November 20, 2006.
6. Memorandum on ERPlus HRMS implementation outstanding issues, November 5, 2001.
7. GCU-Beta Campus training plans.
8. Gulf Coast University ERPlus implementation project charter and work plan.
9. GCU ERPlus implementation organization charts.
10. GCU-Beta Campus ERPlus student administration system, a PowerPoint presentation to the GCU
system board of regents, January 29, 2003.
11. ERPlus student administration system: GCU-Beta Campus' position, prepared by Beta Campus operations
middle managers, April 5, 2001.
12. Alpha Campus and Beta Campus Faculty Senate Minutes.
13. E-mail messages sent by implementation stakeholders.
Appendix D. Initial interpretation of the data

Table D.1
The episodic circuit at GCU.
Circuit Research issues The episodic circuit at GCU

(form of (adapted from Silva &
power) Backhouse, 2003)
Episodic Who are the As and Bs? This allows The members of the management group are the As and the
(causal power) identification of those who users group are the Bs.
promote the information system
and those who challenge it
What are the desired objectives of the The As sought to implement ERPlus on time, under budget and with
As and Bs as they struggle about the minimal modifications. Further, they sought to gain accurate
information system? accounting control across the GCU campuses.
The Bs sought to perform their work tasks in ways that are consistent
with the business processes of a university.
What strategies and tactics have the The As used their control over financial and programming resources in
As and Bs employed to achieve their their efforts to keep the implementation on schedule and minimize
desired objectives? customization of the information system. In addition, they instituted
other implementation techniques (e.g., the program of communicating
with users) aimed at furthering their objectives. Lastly, they used their
control over access to the board of regents to limit the Bs' ability to
communicate any negative information or opinions about ERPlus.
The Bs used their critical knowledge of work processes and Beta
Campus' independent operation to issue an ultimatum to the As that
the implementation would not proceed until the As met their
requirements.
They communicated negative views about ERPlus to their
subordinates to convince them not to use it. The Bs used workarounds
(continued on next page)

Table D.1 (continued)
Circuit Research issues The episodic circuit at GCU

to adapt ERPlus to their needs and shadow systems to avoid using the
system. Finally, they imputed false information into the system to
meet their needs.
What are the As' and Bs' standing The standing conditions of the As are determined by the organizational
conditions? These should be rules and structure of GCU, as well as by a grant of authority and
understood in terms of the positions control of resources from the board of regents.
that the As and Bs hold within the The standing conditions of the Bs are determined by the
organizational structure. organizational rules and structure of GCU.
What resources do the As and Bs The As have control over financial and programming resources, as
have access to and utilize in their well as authority granted by the board of regents.
efforts to achieve their desired The Bs have the support of the leadership of Beta Campus, which is
objectives? relatively independent in its control over the operations of this
campus and its participation in the implementation. Some Bs also
have formal authority over their subordinates.
Table D.2
The social integration circuit at GCU.
Circuit Research issues The social integration circuit at GCU

Social integration What organizational rules and norms The organizational rules and structure of GCU place the As
(dispositional power) determine the and Bs in their respective positions.
As' and Bs' respective positions?
What are the meanings that are The As viewed ERPlus as a means to obtain accurate
assigned to the information system accounting control of GCU's campuses and meet the
and implementation techniques during university system's future needs. In addition, the As saw the
the implementation process? implementation techniques that they employed as
appropriate means for accomplishing their objectives.
The Bs believed that ERPlus was inconsistent with their work
practices and the business processes of a university.
Furthermore, they viewed the implementation techniques
used by the As as a hindrance to their efforts to perform
their work tasks.
What rules and modalities The As interpreted ERPlus in terms of its ability to provide
of interpretation do accurate accounting control of the GCU system's campuses.
the organizational In addition, they interpreted the implementation techniques
members draw upon to make that they used in terms of their ability to further the
sense of the information objectives that they set for the project. They interpreted the
system and the implementation Bs' resistance to the implementation as being the result of
techniques? their dislike of changes to their work tasks.
The Bs interpreted ERPlus and the techniques used to
implement it in terms of their conflict with their work
practices and the business practices of a university. The Bs
viewed their resistance to the implementation as a
reasonable response to the information system's
inconsistency with these work practices and the
business processes.
Table D.3
The systemic integration circuit at GCU.
Circuit Research issues (adapted The systemic integration circuit at GCU

(form of power) from Silva & Backhouse, 2003)
Systemic What techniques do the As and Bs utilize to The As controlled the Bs' access to resources needed to
integration compel and monitor one another's compliance? modify ERPlus and limited their ability to communicate
(facilitative negative views or opinions about ERPlus or its
power) implementation to the board of regents. They required
Table D.3 (continued)
Circuit Research issues (adapted The systemic integration circuit at GCU

(form of power) from Silva & Backhouse, 2003)
the Bs to use the system by retiring the legacy information

systems.
The Bs used the ultimatum to force a stop to the student
administration implementation and compel the As to take
actions more closely aligned with the former group's
understanding of GCU's business processes.
What work tasks are affected by the new The work tasks that ERPlus and the implementation
information system and the techniques used techniques affected included those related to student
to implement it? administration, finance, and human resource
management activities.
How can the information system be set up The As attempted to use ERPlus as a means to discipline
as a tool for instilling discipline? the Bs by requiring them to perform their work tasks in
a standardized fashion that followed the designs for work
tasks and business processes embedded within ERPlus.
References
Adam, F., & O'Doherty, P. (2000). Lessons from enterprise resource planning implementations in Ireland — Towards small and
shorter ERP projects. Journal of Information Technology, 15(4), 305–316.
Akkermans, H., & van Helden, K. (2002). Vicious and virtuous cycles in ERP implementation: A case study of interrelations between
critical success factors. European Journal of Information Systems, 11(1), 35–46.
Allen, D., Kern, T., & Havenhand, M. (2002). ERP critical success factors: An exploration of the contextual factors in public sector
institutions. Proceedings Of The 35th Hawaii International Conference On System Sciences, Waikoloa, Hawaii (pp. 227–236).
Al-Mudimigh, A., Zairi, M., & Al-Mashari, M. (2001). ERP software implementation: An integrative framework. European Journal of
Information Systems, 10(4), 216–226.
Alvarez, R. (2000). Examining an ERP implementation through myths: A case study of a large public organization. AMCIS 2000
Proceedings, Paper 203.
Alvarez, R. (2008). Examining technology, structure and identity during an enterprise system implementation. Information Systems
Journal, 18(2), 203–224.
Amoako-Gyampah, K., & Salam, A. F. (2004). An extension of the technology acceptance model in an ERP implementation
environment. Information Management, 41(6), 731–745.
Backhouse, J., Hsu, C. W., & Silva, L. (2006). Circuits of power in creating de jure standards: Shaping an international information
security standard. MIS Quarterly, 30, 413–438 (Special Issue).
Bagchi, S., Kanungo, S., & Dasgupta, S. (2003). Modeling use of enterprise resource systems: A path analytic study. European Journal of
Information Systems, 12(2), 142–158.
Bajwa, D. S., Garcia, J. E., & Mooney, T. (2004). An integrative framework for the assimilation of enterprise resource planning systems:
Phases, antecedents, and outcomes. The Journal of Computer Information Systems, 44(3), 81–90.
Berente, N., Gal, U., & Yoo, Y. (2010). Dressage, control, and enterprise systems: The case of NASA's full cost initiative. European
Journal of Information Systems, 19(1), 21–34.
Berente, N., Yoo, Y., & Lyytinen, K. (2008). Alignment or drift? Loose coupling over time in NASA's ERP implementation. ICIS 2008
Proceedings, Paper 180.
Bob-Jones, B., Newman, M., & Lyytinen, K. (2008). Picking up the pieces after a “successful” implementation: Networks, coalitions
and ERP systems. AMCIS 2008 Proceedings, Paper 373.
Bohorquez Lopez, V. W., & Esteves, J. (2009). Overcoming knowledge integration barriers in ERP implementation using action
research approach. AMCIS 2009 Proceedings, Paper 392.
Boonstra, A. (2006). Interpreting an ERP-implementation project from a stakeholder perspective. International Journal of Project
Management, 24(1), 38–52.
Buchanan, D., Boddy, D., & McCalman, J. (1988). Getting in, getting on, getting out, and getting back. In A. Bryman (Ed.), Doing
research in organizations (pp. 53–67). London: Routledge.
Caglio, A. (2003). Enterprise resource planning systems and accountants: Towards hybridization? The European Accounting Review,
12(1), 123–153.
Clegg, S. R. (1989). Frameworks of power. London: SAGE Publications.
Dahl, R. (1957). The concept of power. Behavioural Science, 2, 201–215.
Daniel, E. M., & White, A. (2005). The future of inter-organizational system linkages: Findings of an international Delphi study.
European Journal of Information Systems, 14(2), 188–203.
Davenport, T. H. (1998). Putting the enterprise into the enterprise system. Harvard Business Review, 76(4), 121–131.
Denzin, N. K., & Lincoln, Y. K. (Eds.). (1994). Handbook of qualitative research. London: SAGE Publications.
DiMaggio, P., & Powell, W. (1983). Institutional isomorphism and collective rationality in organization fields. American Sociological
Review, 48(2), 147–160.
Dong, L., Neufeld, D., & Higgins, C. (2009). Top management support of enterprise systems implementations. Journal of Information
Technology, 24(1), 55–80.
Elbanna, A. (2007). Implementing an integrated system in a socially dis-integrated enterprise: A critical view of ERP enabled
integration. Information Technology & People, 20(2), 121–139.
Elmes, M. B., Strong, D. M., & Volkoff, O. (2005). Panoptic empowerment and reflective conformity in enterprise systems-enabled
organizations. Information and Organization, 15(1), 1–37.
Foucault, M. (1977). Discipline and punish. New York: Vintage Books.
Foucault, M. (1980). Power/knowledge: Selected interviews and other writings 1972–77. Brighton: Harvester Press.
Foucault, M. (1982). Afterward. The subject and power. In H. L. Dreyfus, & P. Rainbow (Eds.), Michel Foucault: Beyond structuralism
and Hermeneutics (pp. 208–226). London: Harvester Wheatsheaf.
Fowler, A., & Gilfillan, M. (2003). A framework for stakeholder integration in higher education information systems projects.
Technology Analysis & Strategic Management, 15(4), 467–489.
Franz, C. R., & Robey, D. (1984). An investigation of user-led system design: Rational and political perspectives. Communications of the
ACM, 27(12), 1202–1209.
Gartner, Inc. (2011). Gartner says worldwide enterprise software revenue to grow 9.5. percent in 2011. [Press release]. Retrieved on
July 4, 2011 from http://www.gartner.com/it/page.jsp?id=1728615
Giddens, A. (1984). The constitution of society. Cambridge: Polity Press.
Harley, B., Wright, C., Hall, R., & Dery, K. (2006). Management reactions to technological change: The example of enterprise resource
planning. The Journal of Applied Behavioral Science, 42(1), 58–75.
Hislop, D., Newell, S., Scarbrough, H., & Swan, J. (2000). Networks, knowledge, and power: Decision making, politics, and the process
of innovation. Technology Analysis & Strategic Management, 12(3), 399–411.
Howcroft, D., & Light, B. (2002). A study of user involvement in packaged software selection. ICIS 2002 Proceeding, Paper 7.
Ignatiadis, I., & Nandhakumar, J. (2007a). The impact of enterprise systems on organizational resilience. Journal of Information
Technology, 22(1), 36–43.
Ignatiadis, I., & Nandhakumar, J. (2007b). The impact of enterprise systems on organizational control and drift: A human–machine
agency perspective. International Journal of Enterprise Information Systems, 3(3), 36–51.
Ignatiadis, I., & Nandhakumar, J. (2009). The effect of ERP system workarounds on organizational control: An interpretivist case
study. Scandinavian Journal of Information Systems, 21(2), 59–90.
Jasperson, J., Carte, T. A., Saunders, C. S., Butler, B. S., Croes, H. J. P., & Zheng, W. (2002). Review: Power and information technology
research: A metatriangulation approach. MIS Quarterly, 26(4), 397–459.
Khoo, H. M., & Robey, D. (2007). Deciding to upgrade packaged software: A comparative case study of motives, contingencies and
dependencies. European Journal of Information Systems, 16(5), 555–567.
Klein, H. K., & Myers, M. D. (1999). A set of principles for conducting and evaluating interpretive field studies in information systems.
MIS Quarterly, 23(1), 67–94.
Lee, A. S., & Baskerville, R. L. (2003). Generalizing generalizability in information systems research. Information Systems Research,
14(3), 221–243.
Legare, T. L. (2002). The role of organizational factors in realizing ERP benefits. Information Systems Management, 19(4), 21–42.
Lim, E. T. K., Pan, S. L., & Tan, C. W. (2005). Managing user acceptance towards enterprise resource planning (ERP) systems:
Understanding the dissonance between user expectations and managerial policies. European Journal of Information Systems,
14(2), 135–149.
Lukes, S. (1974). Power: A radical view. London: The Macmillan Press Ltd.
Marabelli, M., & Newell, S. (2009). Organizational learning and absorptive capacity in managing ERP implementation projects. ICIS
2009 Proceedings, Paper 136.
Markus, M. L. (1983). Power, politics, and MIS implementation. Communications of the ACM, 26(6), 430–444.
Markus, M. L., & Björn-Andersen, N. (1987). Power over users: Its exercise by system professionals. Communications of the ACM,
30(6), 498–504.
Markus, M. L., Tanis, C., & van Fenema, P. C. (2000). Multisite ERP implementations. Communications of the ACM, 43(4), 42–46.
Meyer, J., & Rowan, B. (1991). Institutionalized Organizations: Formal Structure as Myth and Ceremony. In W. W. Powell, & P. J.
DiMaggio (Eds.), The New Institutionalism in Organizational Analysis (pp. 41–62). London: The University Press of Chicago.
Newman, M., & Zhao, Y. (2008). The process of enterprise resource planning implementation and business process re-engineering:
Tales from two Chinese small and medium-sized enterprises. Information Systems Journal, 18(4), 405–426.
Niehaves, B., Klose, K., & Becker, J. (2006). Governance theory perspectives on IT consulting projects: The case of ERP
implementation. e-Service Journal, 5(1), 5–26.
Palomino Murcia, M. A., & Whitley, E. A. (2007). The effects of national culture on ERP implementation: A study of Colombia and
Switzerland. Enterprise Information Systems, 1(3), 301–325.
Pan, S., & Sritharan, S. (2007). Managing a feature-function-stakeholder (FFS) fit process in an enterprise system implementation.
ICIS 2007 Proceedings, Paper 95.
Parr, A., & Shanks, G. (2000). A model of ERP project implementation. Journal of Information Technology, 15(4), 289–303.
Parsons, T. (1967). Sociological theory and modern society. New York: Free Press.
Pollock, N., & Cornford, J. (2004). ERP systems and the university as a ‘unique’ organization. Information Technology & People, 17(1), 31–52.
Robey, D., Ross, J. W., & Boudreau, M. -C. (2002). Learning to implement information systems: An exploratory study of the dialectics
of change. Journal of Management Information Systems, 19(1), 17–46.
Sabherwal, R., & Grover, V. (2010). A taxonomy of political processes in systems development. Information Systems Journal, 20(5),
419–447.
Sathish, S. (2006). Tripartite stakeholder management during an enterprise system implementation: A processual perspective. ICIS
2006 Proceedings, Paper 74.
Scott, J. E. (1999). The FoxMeyer drugs' bankruptcy: Was it a failure of ERP? AMICS 1999 Proceedings, Paper 80.
Scott, S. V., & Wagner, E. L. (2003). Networks, negotiations, and new times: The implementation of enterprise resource planning into
an academic administration. Information and Organization, 13(4), 285–313.
Seethamraju, R. (2008). Role of enterprise systems in achieving supply chain integration. International Journal of Business Insights &
Transformation, 1(2), 1–7.
Shepherd, C., Clegg, C., & Stride, C. (2009). Opening the black box: A multi-method analysis of an enterprise resource planning
implementation. Journal of Information Technology, 24(1), 81–102.
Sheu, C., Chae, B., & Yang, C. -L. (2004). National differences and ERP implementation. Omega, 32(5), 361–371.
Silva, L. (2007). Epistemological and theoretical challenges for studying power and politics in information systems. Information
Systems Journal, 17(2), 165–183.
Silva, L., & 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), 294–336.
Smith, S., Winchester, D., Bunker, D., & Jamieson, R. (2010). Circuits of power: A study of mandated compliance to an information
systems security de jure standard in a government organization. MIS Quarterly, 34(3), 463–486.
Swan, J., & Scarbrough, H. (2005). The politics of networked innovation. Human Relations, 58(7), 913–943.
Taylor, C. (1971). Interpretation and the sciences of man. The Review of Metaphysics, 25(71), 3–51.
Volkoff, O. (1999). Using the structurational model of technology to analyze an ERP implementation. AMCIS 1999 Proceedings, Paper
84.
Wagner, E. L., & Newell, S. (2006). Repairing ERP: Producing social order to create a working information system. The Journal of
Applied Behavioral Science, 42(1), 40–57.
Wagner, E. L., Newell, S., & Piccoli, G. (2010). Understanding project survival in an ES environment: A sociomaterial practice
perspective. Journal of the Association for Information Systems, 11(5), 276–297.
Wagner, E. L., Scott, S. V., & Galliers, R. D. (2006). The creation of ‘best practice’ software: Myth, reality and ethics. Information and
Organization, 16(3), 251–275.
Walsham, G. (1995). Interpretive case studies in IS research: Nature and method. European Journal of Information Systems, 4(2),
74–81.
Walsham, G. (2006). Doing interpretive research. European Journal of Information Systems, 15(3), 320–330.
Weber, M. (1999). The three types of legitimate domination. In R. Swedberg (Ed.), Essays In Economic Sociology (pp. 99–108).
Princeton, NJ: Princeton University Press.
Whittaker, L., & Beswick, C. (2008). When a sure thing goes awry: Implementing an enterprise system at IST Ltd. Management
Dynamics, 17(2), 27–39.
Yin, R. K. (1994). Case study research: Design and methods. London: SAGE Publications.

Information and Organization: Leiser Silva, H. Kevin Fulk

Diunggah oleh

Informasi Dokumen

Judul Asli

Hak Cipta

Format Tersedia

Bagikan dokumen Ini

Bagikan atau Tanam Dokumen

Opsi Berbagi

Apakah menurut Anda dokumen ini bermanfaat?

Apakah konten ini tidak pantas?

Hak Cipta:

Format Tersedia

Information and Organization: Leiser Silva, H. Kevin Fulk

Diunggah oleh

Hak Cipta:

Format Tersedia

Information and Organization 22 (2012) 227–251

Contents lists available at SciVerse ScienceDirect

Information and Organization

From disruptions to struggles: Theorizing power in ERP

Article history: Enterprise resource planning (ERP) implementation projects often

2.1. Conceptual account of the circuits of power

Fig. 1. Circuits of power.

3. ERP system implementation research addressing issues of power

3.1. Episodic circuit

3.2. Social integration circuit

3.3. Systemic integration circuit

4.1. Research sites

4.2. Data collection

4.3. Data analysis

5. Case narrative and analysis

5.1. Episode 1: planning

5.1.1. Analysis of the circuits during episode 1

5.2. Episode 2: starting the implementation

5.2.1. Analysis of the circuits during episode 2

5.3. Episode 3: conﬂict and its attempted resolution

5.3.1. Analysis of the circuits during episode 3

6. Summary of results and discussion

6.1. Disturbances in the circuit of social integration

Social integration Contradicting interpretations Disposition to not cooperate Bypassing

6.2. Disturbances in the circuit of systemic integration

Fig. 2. The interplay between social and systemic integration.

6.3. Disturbances in the episodic circuit

Appendix A. Application of Klein and Myers' validation criteria to this case

Klein and Myers' criteria Application to this study

Appendix B. Themes for semi-structured interviews

- A (or B)'s identity in the power relations.

- Techniques to assure compliance

Appendix C. List of documents used as secondary case data

1. ERPlus Project History.

Appendix D. Initial interpretation of the data

Circuit Research issues The episodic circuit at GCU

(continued on next page)

Table D.1 (continued)

Circuit Research issues The episodic circuit at GCU

Circuit Research issues The social integration circuit at GCU

Circuit Research issues (adapted The systemic integration circuit at GCU

Table D.3 (continued)

Circuit Research issues (adapted The systemic integration circuit at GCU

the Bs to use the system by retiring the legacy information

Anda mungkin juga menyukai