Towards a Theory of Interorganizational Collaboration: Generic Structures of
Cross-Boundary Requirements Analysis’?
Luis Felipe Luna-Reyes ‘ and David F. Andersen‘*
” Universidad de las Américas, Puebla, Business School, NE221J Santa Catarina Martir, Cholula, Puebla,
72820, Mexico, Phone: +52 (222) 229-2000 ext. 4536, Fax: +52 (222) 229-2726 email:
luna@udlay
” Rockefeller College of Public Affairs and Policy, University at Albany, 135 Western Avenue, Albany, NY,
12222,USA, Phone: +1 (518) 442-5280, Fax: +1 (518) 442-5298, email: david.andersen@albany.edu
Abstract
In this paper, we present a series of causal maps that constitute an initial effort in the
creation of a generic theory of interorganizational cross-boundary requirements analysis.
Such causal structures are the result of a simulation-based study in which we explored the
interactions and social processes associated with the development of trust and knowledge
sharing in the development of an interorganizational information system in New York
State: the Homeless Information Management System (HIMS). The paper includes the
main theoretical and practical implications of the modeling and simulation work, as well
as discussion of some paths to continue the exploration of collaboration in this specific
context. The causal maps are organized around three themes that emerged during the
modeling process. The first theme is related to trust development, and its recursive
interactions with knowledge sharing and learning. The second theme is related to the
importance of achieving stakeholder engagement by establishing a trusting environment
as opposed to the use of authority or coercive mechanisms. The last theme is associated
The research reported here is supported by National Science Foundation grant #SES-9979839. The
views and conclusions expressed in this paper are those of the authors alone and do not reflect the
views or policies of the National Science Foundation.
This paper is based on earlier work contributed to by Laura J. Black, Donna Canestraro, Meghan Cook,
Anthony M. Cresswell, Ignacio J. Martinez-Moyano, Theresa A. Pardo, George P. Richardson and
Fiona Thompson
with the understanding of requirement definition as a social process of learning and
knowledge transfer.
Introduction
Using information technologies to increase understanding of the impacts of social
programs, and to improve the level of service provided to clients or users of services, is
an important trend observed in both the public and the private sector. Interagency
collaboration to build such information systems to improve public programs or services is
appealing in many ways such as cost savings, resource sharing or improved efficiencies
(Bardach 1998). When public services are provided through networks of private service
providers or networks of geographically disperse and decentralized public offices,
developing these information systems in a collaborative way becomes a need. In spite of
the advantages of collaboration, collaborative approaches are not as common as they
should be, because of our lack of understanding about how to manage the collaboration
process (McCaffrey et al. 1995; Dawes and Pardo 2003).
The social processes associated with the development of trust and knowledge sharing are
particularly important in the success of such collaborative innovation initiatives.
Although interactions among stakeholders take place during the whole development
process, the interactions during the initial stages, where the requirements of the
technology innovation are analyzed and defined, appear to be particularly important to
the success of the entire process. The use of objects such as prototypes to facilitate the
definition of the innovation constitutes a viable alternative for improving the learning
process and for detecting overoptimistic estimates of costs and effort associated to the
full implementation of the innovation. The timing of the use of these objects, however,
has an impact on the effectiveness of the social process around the definition of
requirements and on the perception of the feasibility of the opportunity to collaborate.
In this paper, we present a series of causal maps that constitute an initial effort in the
creation of a generic theory of interorganizational collaboration. Such causal structures
are the result of a simulation-based study in which we explored the interactions of the
factors mentioned in the above paragraphs in the development of an interorganizational
information system in New York State: the Homeless Information Management System
(HIMS).
The paper includes the main learning and theoretical implications of the research, as well
as discussion of some paths to continue the exploration of collaboration in this specific
context. The paper is organized in four main sections after this brief introduction. The
first describes the methods and data used in the modeling process. The second includes
some of the theoretical implications of the work, in terms of three main themes that
emerged during the modeling process: trust development, stakeholder engagement, and
requirement definition as a social process. The third section discusses the practical
implications of the theory, and the last one outlines some of the paths for future research
and theory development.
Methods and Data
The causal maps reported in this paper constitute the main conclusions of a simulation-
based theory-building project grounded on two cases of interorganizational Information
Technology Projects. The first case deals with information resources for programs
serving the more than 29,000 homeless people receive emergency shelter and a diversity
of support services each day in New York State. Homeless services costs are estimated to
be $350 million each year, $130 of which are spent on service programs (CTG 2000).
The information needed to assess the effectiveness and impact of the services provided to
the homeless is distributed in several agencies and nonprofits, such as the Bureau of
Housing Services (BHS), and the New York City Department of Homeless Services
(DHS). The lack of integration of the data sources makes very difficult to assess them.
Starting in 1998, the Office of Temporary and Disability Assistance (OTDA), Bureau of
Housing Services (BHS) started a series of efforts to create an integrated decision support
system to help both government and nonprofit organizations to manage and assess
homeless services. The system would integrate information from a variety of sources.
Demographic data would be obtained from the individual shelters, payment information
would come from the state Welfare Management System (WMS), shelters’ information
would be gathered from the BHS’s providers certification database, medical information
from the State Department of Health, and data on substance abuse or other services from
other State Agencies. Although BHS is an oversight agency, which manages and
regulates temporary housing programs in New York State, it shares its regulatory
functions in New York City with the NYC Department of Homeless Services.
The second case deals with information resources for the Division of Municipal Affairs
(MA) from the New York Office of the State Comptroller (OSC), the agency responsible
for monitoring the financial operations of the 3,200 local governments in the State (CTG
2001). One of MA primary roles is to gather, organize, and distribute information from
and about local governments to a diversity of users. Some of the main sources and users
of information are government officials, MA staff, the media, taxpayers, the Governor’s
office, and the State Legislature. The information gathered can take many forms, such as
written correspondence, telephone calls, news articles or media reports, electronic
exchanges, and individual staff notes. The diversity of sources, forms and uses of the
information created a series of challenges for the new vision of MA’s work, which was
moving from a regulatory and auditing position towards promoting improvement in local
practices through a program of services. Also in 1998, MA started an initiative to
promote the creation of a “technology solution involving a widely accessible repository
of contact information [..] It would provide a ‘knowledge base’ of information about
municipalities and local officials, past services provided, and preferred modes of service
delivery” (CTG 2001).
This research builds on an earlier theory building effort that yielded a simulation model
for the HIMS project involving the collaboration of two actors: the Bureau of Housing
Services as the initiator of the innovation, and an aggregation of shelter providers and
local public agencies that represent the main stakeholders in the project. This initial
modeling effort, which took place on a series of Group Model Building sessions, was
continued and enriched through document analysis and individual interviews with project
participants.
A basic assumption of the model is that the processes of trust development and
knowledge sharing shape the collaboration patterns between the two actors in the
development of cross-boundary work. In this way, the key assets of each participant in
the development of the project are two different kinds of knowledge and their
accumulation of trust on the other actor in the project. By getting engaged in the project
work, each actor learns about his own role and the other actor’s role in the project.
Moreover, each of them develops a sense of the other actor’s trustworthiness.
Through the analysis of the case data, we identified three main cross-boundary activities
associated with the collaboration: the definition of requirements for the prototype, the
technical development of the HIMS prototype itself, and the development of a prototype
definition of services and evaluation model. The simulation model also recognizes that
besides the interchange of knowledge among the parties involved, new knowledge was
generated. Particularly relevant to the project was the creation of a shared vision about
the feasibility of the project from the providers’ perspective.
To assess the theory, both model structure and behavior were compared with the
experiences of HIMS participants, and the assumptions of the model were compared with
experiences of a group of people participating in different projects associated with the
development and implementation of the Multi-purpose Access for Customer Relations
and Operational Support (MACROS) at the NYS Office of the State Comptroller.
During the development of the model, three main themes emerged in the case stories and
the theories embedded in the model. These three themes appear to be particularly relevant
for a theory of collaboration in IT innovations because of the importance that they have
in the stories about the project, and because of the interest that researchers in the area
show in several related literatures. The first theme is related to trust development, and its
recursive interactions with knowledge sharing and learning. The second theme is related
to the importance of achieving stakeholder engagement by establishing a trusting
environment as opposed to the use of authority or coercive mechanisms. The last theme is
associated with the understanding of requirement definition as a social process of
learning and knowledge transfer. Thus, the causal maps summarizing the main learning,
as well as the implications of the model are organized around these three themes.
An Initial Theory of Cross-Boundary Collaboration
This work has as an immediate antecedent in the work of Black (2002). In her work, she
developed a theoretical framework to explain collaboration patterns in settings that
involve the collaboration of two actors. She demonstrated the utility of the framework in
two different settings; one involved the use of new scanning technologies between
technicians and doctors, and the second involved interdepartmental collaboration in
product development. Through this work, we found that the framework is also useful in
the analysis of interorganizational collaboration for the development of IT innovations,
particularly in the early stages of requirements identification and definition. More
specifically, the current theory considers the presence of an organization interested in
advancing the vision of an IT innovation, and its interactions with an aggregate of
stakeholder organizations whose involvement is necessary for the development of the
information system (see Figure 1).
An important difference between the theories developed in this work and the work of
Black is related to the way in which the collaboration developed. Her work analyzed
collaboration patterns as they emerged when two groups with different backgrounds
collaborate in a specific task or project. This research considers the presence of a
facilitation group that provides the necessary social processes and objects, as well as
coordinates the timing of each participant involvement in different project-related tasks.
Concretely, this theory reflects the way in which the Center for Technology in
government facilitates the definition of the requirements for IT innovations.
A common element in both theories is consideration of several accumulations of
knowledge as playing an important role in the collaboration process. These
accumulations of knowledge are related to the extent to which each actor understands his
own information needs, and the information needs of the other actors involved in the
project.
Finally, this research extends Black’s theory with the formal incorporation of trust and its
relationships with the knowledge sharing process. We present in the following sections
some simplified feedback structures that illustrate the main theoretical contributions of
this work.
Uses knowledge about
providers’ needs to
improve (or deteriorate)
collaboration experience
poceece enn e ene n eee +
H H
' H
' ‘
'
f
Hl
' BHS
' en
¥
\
Brings
' knowledge and
'
‘
Providers x
Hl
Hl
Hy
Provids Trust ; Learns effort to the work
seven) | enhances? —$. _about__ Leesan!
Tor timis) Defining CIG
knowledge needs and requirements
sharing his own
Providers’ f=
knowledge g CTG's
of BHS > engagement
wererneee---- aad § s
Learn about their needs, | “3 prototyping) | needs and
project feasibility, and 2 en
develop (or not) trust 3 ae
Providers! elon (OE ms os constraints
willingness 2
to engage 5
is)
Developing
service model
(interaction f) about project
o i objectives,
Providers’ providers and
knowledge BHS
of providers Bring knowledge and
effort to the work '
Provides project
management,
Providers’
assessment of
project
feasibility
fort
Allocation
Works with BHS to establish a risk free ins
framework for collaboration
ba------p|
Assesses providers’ willingness to engage and interest in the project
Figure 1. A Sector view of the trust-and-knowledge collaboration theory.
Trust Development
In terms of trust development, the research reported in this paper is not adding new
constructs to the trust literature. The a priori, calculative, and knowledge-based
components of trust, as well as the concepts of propensity to trust, trustworthiness or risk
are present in previous efforts to integrate this rich corpus of literature (Mayer et al.
1995; Rousseau et al. 1998). Previous research also suggests the existence of links
between knowledge and trust (Shapiro ef al. 1992; Levin et al. 2002a). However, the
discipline and rigor introduced by the use of mathematical modeling have created an
original organization of those concepts around two main feedback processes presented in
Figure 2 as a two-actor interaction.
As suggested in Figure 2, our trust in them can be conceptualized as a probability to act
that results from “averaging” two different components: a calculative component, and a
knowledge-based component. The knowledge-based component is in turn an “average”
of our a priori perception of their trustworthiness, and our perception of their
trustworthiness. Our knowledge of them weights the importance of each element in the
averaging process. The knowledge-based component of trust is also involved with two
feedback processes. A perception bias makes us reinforce our current perception of their
trustworthiness. Moreover, our trust in them enhances (or limits) their knowledge of our
needs, enhancing (or limiting) in turn their ability to build trust by using their knowledge
of our needs to enhance our collaboration experiences with them. Calculative trust is
conceptualized to be the result of our assessment of the risks and desirability of working
with them. Therefore, efforts to build institutional trust by providing structures or
regulations that penalize betrayal decrease our perception of risk, operating through the
calculative component of trust (Zucker 1986; Shapiro et al. 1992).
This trust conceptualization produces a nonlinear path dependence behavior documented
frequently in the literature on trust and collaboration (Burt and Knez 1996; Powell 1996;
Rousseau et al. 1998). Moreover, the structural elements are consistent with the
perception of HIMS and MACROS participants, as well as the literature on trust. The
structure is capable of reproducing the qualitative pattern of trust development in the
HIMS case.
Our colhboration
experiences with Their
them knowledge of
our needs
Perception
aes Their ability to
build trust
Our perception of
fiver ings
ore ey in
Gur apebil we, “hg
perception of their Our
trustworthiness knowledge
of them
Calculative
trust
Institutional
Trust
Figure 2. Feedback processes involved in the development of our trust in them.
Observed asymmetries between patterns of trust and distrust formation have led to
treating trust and distrust as two different constructs that can increase or decrease
independently, moving in different continua (Kramer et al. 1996; Lewicki et al. 1998).
Partial experiments with the theory of trust in Figure 2 suggest that the asymmetries can
exist in a single continuum, considering distrust as the lack of trust, and not a different
and conceptually independent construct.
Engaging Stakeholders
Trust and collaboration are frequently linked in the literature in a reinforcing process that
works as a trap or as a virtuous cycle of positive relations (Zand 1972; Kramer et al.
1996; Hardin 2001). However, most of these accounts are present in the literature in form
of a text, or at most in the form of a single feedback loop such as the one developed by
Vangen and Huxham (2003), in which they link the willingness to collaborate with
positive results obtained from working together. The theory development process
followed in this research points to the existence of two feedback processes that affect our
willingness to engage in collaboration that operates through the calculative and
knowledge-based components of trust (see Figure 3).
The development of trust is intertwined with the process of engagement, and depends
itself on several knowledge accumulations that result from doing work together. As
described in the previous section, our trust in them is a probability to act that comes from
the aggregation of an a priori component, a calculative component, and a history-based
component. Although the a priori component is considered to be more or less stable for
the duration of a project, the calculative and history-based components are considered to
be dynamic in nature. The calculative component is related to our perceptions of the risk
and desirability of the project, and the history-based component is related to the
accumulations of good and bad episodes in our collaboration experiences with them. The
weight that a particular actor assigns to each component depends on how much she
knows her collaborating counterpart, knowledge that is developed through repeated
interactions and activities developed together.
Partial experiments with this structure, suggest that managing the perception of risk by
establishing institutional trust has the potential to be a leverage point to start a
collaboration process. Moreover, showing results fast is an alternative way to leverage
the collaboration process.
10
Establishing
institutional
trust
Our perception
of their
trustworthiness.
Learnig about their
trustworthiness
trust
Learning about project \ Our collaboration
risks and desirability
; Ou knowkdge experiences with
Our willingness of our needs
to engage
NN Doing work.
together
Our trust in Calculative
them
Figure 3. Calculative trust, knowledge-based trust and the process of engagement.
Requirement Definition as a Social Process
Requirement definition is conceptualized as a learning-intensive process in which users,
analysts, and other stakeholders share their knowledge about information needs and uses.
The main assumption about the nature of knowledge is that it is distributed. It does not
only reside in the minds and bodies of individuals, but it is also socially and physically
distributed (Black 2002). That is to say, particular communities develop shared meanings,
frequently associated with the objects with which they interact on a regular basis (Brown
and Duguid 1991; Wenger 1998). The BHS team, for example, developed during their
interaction in the process particular terms to refer to the homeless services. Different
actors in the project shared meanings about data and business rules specific to each of
their particular practices (Figure 4).
The distributed nature of knowledge is highly consistent with the concept of knowing in
action (Cook and Brown 1999; Gherardi 2001). From this perspective differences
between individual, group, tacit, and explicit knowledge are elusive, and the modes of
knowledge creation proposed by Nonaka (1994) become less relevant for the discussion.
For example, making explicit the tacit knowledge of providers and BHS in the
development of HIMS data elements is not any more knowledge creation, but the creation
of an object that can be used to share knowledge across organizational boundaries.
Their knowledge of Their knowledge of
our needs their needs
ra Incorporating Learning about
. (or not) our themselves
Our trust in needs into
‘digi requirements
Knowledge-
based trust
Our perception Cross-boundary work
of their
trustworthiness
Facilitator's
Our perception knowledge about
of project risks caeuause project objectives,
and desirability ie and stakeholders!
needs
Learning about Spinning
ourselves plates/providing
objects
Our knowledge of
our needs
Effort
Allocation
Figure 4. The impact of trust in the process of knowledge sharing across boundaries.
A more relevant discussion about the nature of knowledge sharing in a cross-boundary
setting such as defining requirements for an information system can be organized in
terms of the three approaches to share knowledge across boundaries identified in the
literature: syntactic, semantic, and pragmatic (Saracevic 1999; Malhotra 2000; Carlile
2002a). These three approaches can be associated with three different process of
knowledge sharing: transferring, translating and transforming (Carlile 2002b). Knowledge
transfer occurs at the syntactic level, when groups of people share a basic syntax, as
providers and BHS shared the data dictionaries of each of their systems and databases.
Knowledge translation occurs when individuals or groups share meanings that promote
the appropriate interpretation of the shared syntax, as providers and BHS shared the
12
meanings of concepts such as recidivism, length of stay, first timers or repeaters.
Knowledge transformation occurs when groups of people understand the dependencies
and practical implications of their shared understandings to their day-to-day practice, like
the conversations between the prototyping team and the requirement definition team in
the HIMS project, which led to adaptations and modifications in the original design and
to a better understanding of the quality of the data and its implications for system
implementation.
As shown in Figure 4, the knowledge sharing processes is also intertwined with the
process of trust development (Levin et al. 2002a; Levin et al. 2002b; Levin et al. 2002c).
Again, the concept is not new for the literature, but the model provides a way to integrate
and explore the interactions between these two processes.
The view of requirement definition as a knowledge sharing process is consistent with
current theoretical and practical approaches existent in the literature. Moreover, problems
related to learning and sharing knowledge are recognized as the main obstacles to the
effective development of requirements (the WITHIN, BETWEEN, and AMONG
obstacles identified by Byrd (1992) are an example of it). However, and as pointed out by
Black (2002), the collaboration is not limited only to the adequate sharing of knowledge,
but it involves the creation of new knowledge. Providers in the HIMS project, for
example, developed a shared vision about the feasibility of the project, as they learned
about the differences and similarities of their practices solving the AMONG obstacles
described in the literature.
The social process associated with the creation of this shared vision appears to be related
with three feedback processes in the interaction (see Figure 5).
The first is related to a particular facilitation design that involves two main stages, one
divergent and the other convergent in nature. The convergent part of the interaction, in
which the participants look for emerging patterns in their information needs, is closely
related to the development of this shared vision. Providers in the HIMS project had the
opportunity to create 10 categories of services by grouping all services provided in their
facilities. Through this process, they learned about themselves and created a shared
13
picture of the homeless services in NYS. Participants in the MACROS project have also
made intensive use of this facilitation design throughout their collaboration in the project,
during the early stages of definition of the technical assistance function, and during the
meetings with the advisory committee. Other practitioners in IS development have found
the approach to be useful in reaching agreement and in creating shared visions in the
definition of requirements (Boehm et a/. 2001; Gottesdiener 2003).
Seeing is
believing,
ee
Prototyping
Learning from
project
progress :
Ow steaonest oF mde (ure)
project feasibility ~
Understanding
differences and
similarities
Our perception of
our diversity
Our trust in Developing trust
them
Our knowledge of
them
Figure 5. Knowledge creation and the social processes of requirement definition.
A second component in the social process of creating shared vision consists of the use of
artifacts or boundary objects that facilitate the process of sharing knowledge across
boundaries. A particularly important object in the HIMS project was the prototype itself,
which constituted a concrete artifact to prove the utility and feasibility of the system. In
the MACROS project, the contact repository prototype played a key role in the diffusion
process. As one of the project leaders mentioned in a presentation of the project, “seeing
is believing.”
14
The last component of the social process in the HIMS development model is the effect of
trust in the creation of this shared vision. Lower levels of trust will limit the effectiveness
of both the process facilitation and the use of boundary objects.
Finally, another important accumulation of knowledge involved in the model is related to
acquiring the experience of working together. HIMS participants acquired this experience
in their continued efforts to advance the vision of HIMS. When discussing this
assumption with participants of the MACROS project, for example, several of them made
reference to a common meeting design used in their collaborations composed by four
different stages, Forming, Storming, Norming, and Performing (Gottesdiener 2003).
Implications for practice
The theory development process—consisting of the case studies, modeling and
simulation—also suggests several practical implications for groups participating in IT
innovations similar to HIMS or MACROS. We will summarize the most important
implications in this section of the paper.
Managing complex projects is an activity that starts with a high level of uncertainty and
ambiguity. In spite of having an initial problem definition and a well-defined objective,
the BHS/CTG team members spent several months learning about the project and the
stakeholders, and refining their vision of the project. This learning process, however,
facilitated the process of sharing their vision with their main stakeholders. Similarly, the
MACROS group spent a long time drafting and redrafting their initial document to
transform it in an effective communication object with the rest of the divisions at OSC.
The apparently long period of inactivity in the model simulations corresponded to this
initial learning process.
Planning and reflecting is a continuous process. Plans in complex projects constitute
general guidelines to the development of specific project-related tasks. The initial
uncertainty prevents the creation of a specific plan that can be implemented step by step.
The CTG/BHS team adjusted and modified its initial plan to respond to specific
constraints in the project environment such as the existent technologies in the local sites.
15
MACROS project leaders have also adjusted their plans several times along the process.
This flexibility allowed both groups to push forward their IT innovations.
Partial simulations with the trust structure suggest that, although the a priori component
of trust has an important impact in trust development, the efforts to build trust in the day-
to-day interactions can overcome the initial weight of the a priori component. Moreover,
early efforts to develop trust are more effective than those that occur in later stages of the
interaction. Although the development of trust, because of the attention to the
relationship, is a gradual process, the lack of attention to the relation can revert the
process much faster. Finally, managing the institutional component of trust (i.e., reducing
risk) could be a strategy to break the initial trap of distrust.
There is a cost associated to maintain the relationships with stakeholders in the process.
In the theory for the HIMS case, the cumulative effort to contact providers can be
considered as a proxy for this cost. Model experiments suggest that this cost is higher in
those cases where there is no a previous interaction history, and those in which the
leading agency places little attention to the development of a trusting environment. That
is to say, placing little attention to relationship building will require from the project
leader more effort contacting stakeholders in the mid and long terms.
These experiments also suggest that managing the perceptions of risk and desirability of
the project can be effective ways to promote the initial engagement of stakeholders in the
project. Given the current formulation of the theory, managing the perception of risk
appears to be the most effective way to promote the initial work. If the project is
desirable enough, collaboration can continue even when the perception of risk increases
in later stages in the project. The strategy was effective for the HIMS project, and has
been used successfully also by the MACROS team.
The model suggests that is hard to create a history-based trust in as short a period of time
as the HIMS case. Some of the simulation experiments suggest that the most important
component of trust during the development of the prototype was the calculative one.
Given this situation, the team accomplishes its goals in a very similar period of time in
the situation in which there is little interest in fostering a trusting environment. However,
16
the experiments also suggest that the knowledge-based component will be more
important in subsequent project stages.
Stakeholders in interorganizational projects like HIMS or MACROS are actually
different; they have developed different business processes and languages in their day-to-
day work. Recognizing and reflecting these differences back to the group can be an
effective strategy for building a trusting environment. However, looking for patterns in
the differences to assess what is possible and what is not is an important ingredient of an
effective facilitation design. Therefore, facilitation designs that include this kind of
convergent process have the potential to be more effective to reach consensus compared
with those that do not consider convergent processes that look for patterns of
relationships, such as the grouping of services by the shelter providers in the HIMS
project, or the grouping of information needs by the MACROS Advisory Committee.
The use of prototypes and the development of a trusting environment contribute to the
creation of the shared vision about the feasibility of the project. Experiments with the
model suggest that although the presence of each of these elements is important in the
social process, the timing of each of them is also important. Lower levels of the
perception of feasibility of the HIMS are associated with those scenarios where the
intense part of the work with the providers took place in a different time frame than the
intense part of the work developing the prototype, or those in which the intense stage of
the work with the providers took place when the level of trust was low. Being sensitive to
the parallel development of these three processes could yield effective social processes in
the definition of requirements.
Future Research
This final section of the paper points to possible avenues to continue with the research
presented here.
Although we will point to additions and extensions to the current theory, we consider
that, before increasing the complexity of the theory, more experimentation with the
current model is needed. The process has been intense in the development of simulation
7
experiments, but to fully understand the operation of the feedback mechanisms involved
in the theory, we will need to keep experimenting for some period of time.
One interesting approach to conducting such explorations involves the design of partial
simulations isolating parts of the model to further analyze its behavior, contrasting it with
other innovation projects.
Although model assumptions were accepted as feasible by MACROS participants, we did
not try to use the same model structure to explain the development of the MACROS
project. In this way, an exercise could be to explore changes in the model parameters, to
assess the transferability of this particular structure to the second case or to different
cases. For example, similar collaboration processes can be observed in collaborative IT
initiatives in Mexico, which can be considered as additional places to test the theory.
Pieces of the model such as the trust structure or the providers work structure can be used
in individual studies to test the theory with other empirical evidence from collaboration
efforts, collecting the numeric data and observations to “calibrate” and test those dynamic
hypothesis.
There are several simplifying assumptions in the model that require further exploration of
situations of facilitated collaboration. For example, the model assumes that the facilitator
has the experience to guide a group process involving individuals and organizations with
different backgrounds and levels of technical skill. In this way, a natural way to expand
or refine the theory is to add a stock of facilitation knowledge or experience to the
facilitator sector.
Another area, maybe more interesting than the previous one, in which the model contains
important simplifying constraints is related to the idea that the objects used during the
conversations are effective enough to facilitate the translation process across practices.
Although the model does include the effects of the prototype as a concrete artifact to the
creation of a shared vision of the feasibility of the project, it does not include any
statement about the effectiveness of the rest of the objects used in the interaction (pieces
of paper, whiteboards, etc.). Thus, another interesting path of further development of the
theory involves the dynamic study of such kind of objects through additional case studies.
18
The current model can be also transformed into a learning laboratory, where project
managers can “manage” projects by assigning staff and making decisions about the
allocation of effort to the different streams of work in the project.
Finally, another path to continue with this project involves the development of a series of
critical success loops for the project manager of interorganizational IT projects. The work
will consist of refining and testing simplified versions of the key feedback processes
presented in this paper.
References
Bardach E. 1998. Getting agencies to work together: The practice and theory of
managerial craftmanship. Brookings Institution Press: Washington, DC.
Black L. 2002. Collaborating across boundaries: Theoretical, empirical, and simulated
explorations. Unpublished Unpublished Ph.D. Dissertation, Sloan School of
Management. MIT. Cambridge, MA.
Boehm B, Griinbacher P, Briggs R. 2001. Easywinwin: A groupware-supported
methodology for requirements negotiation. Proceedings of the 23rd International
Conference on Software Engineering, Toronto, Canada.
Brown JS, Duguid P. 1991. Organizational learning and communities of practice: Toward
a unified view of working, learning, and innovation. Organization Science
2(1):40-57.
Burt RS, Knez M. 1996. Trust and third-party gossip, in Trust in organizations: Frontiers
of theory and research, Tyler T, Kramer RM (eds.). Sage Publications: Thousand
Oaks, CA, 68-89.
Byrd TA, Cossick KL, Zmud RW. 1992. A synthesis of research on requirements
analysis and knowledge acquisition techniques. M/S Quarterly 16(1):117-138.
Carlile P. 2002a. A pragmatic view of knowledge and boundaries: Boundary objects in
new product development. Organization Science 13(4):442-455.
---. 2002b. Transferring, translating and transforming: An integrative framework for
managing knowledge across boundaries. Cambridge, MA.
Cook S, Brown J. 1999. Bridging epistemologies: The generative dance between
organizational knowledge and organizational knowing. Organization Science
10(4):381-400.
19
CTG. 2000. Building trust before building a system: The making of the homeless
information management system. Center for Technology in Government:
---. 2001. Using information for organizational change: The transition from regulation
to service in the division of municipal affairs. Center for Technology in
Government:
Dawes §, Pardo T. 2003. Building collaborative digital government systems: Systemic
constraints and effective practices, in Advances in digital government:
Technology, human factors, and policy, McIver WJ, Elmagarmid AK (eds.).
Kluwer Academic Publishers: Boston, MA, 259-273.
Gherardi S. 2001. From organizational learning to practice-based knowing. Human
Relations 54(1):131-139.
Gottesdiener E. 2003. Requirements by collaboration: Getting it right the first time. JEEE
Software 20(2):52-55.
Hardin R. 2001. Conceptions and explanations of trust, in Trust in society, Cook KS
(ed.). Russell Sage Fundation: New York, 3-39.
Kramer RM, Brewer MB, Hanna BA. 1996. Collective trust and collective action: The
decision to trust as a social decision, in Trust in organizations: Frontiers of theory
and research, Tyler T, Kramer RM (eds.). Sage Publications: Thousand Oaks,
CA, 357-389.
Levin D, Cross R, Abrams L. 2002a. Why should i trust you? Antecedents of trust in a
knowledge transfer context. Cambridge, MA.
---. 2002b. The strength of weak ties you can trust: The mediating role of trust in
effective knowledge transfer. Cambridge, MA.
Levin D, Cross R, Abrams L, Lesser E. 2002c. Trust and knowledge sharing: A critical
combination. Somers, NY.
Lewicki RJ, McAllister D, Bies R. 1998. Trust and distrust: New relationships and
realities. Academy of Mangement Review 23(3):438-458.
Malhotra Y. 2000. From information management to knowledge management: Beyond
the "hi-tech hidebound" systems, in Knowledge managment for the information
professional, Srikantaiah TK, Koenig MED (eds.). Information Today, Inc.:
Medford, NJ, 37-62.
Mayer RC, Davis JH, Schoorman FD. 1995. An integrative model of organizational trust.
The Academy of Management Review 20(3):709-734.
McCaffrey D, Faerman S, Hart D. 1995. The appeal and difficulties of participative
systems. Organization Science 6(6):603-627.
20
Nonaka I. 1994. A dynamic theory of organizational knowledge creation. Organization
Science 5(1):14-37.
Powell W. 1996. Trust-based forms of governance, in Trust in organizations: Frontiers
of theory and research, Tyler T, Kramer RM (eds.). Sage Publications: Thousand
Oaks, CA, 51-67.
Rousseau DM, Sitkin SB, Burt RS, Camerer C. 1998. Not so different after all: A cross-
discipline view of trust. Academy of Mangement Review 23(3):393-404.
Saracevic T. 1999. Information science. Journal of the American Society for Information
Science 50(12):1051-1063.
Shapiro DL, Sheppard BH, Cheraskin L. 1992. Business on a handshake. Negotiation
Journal 8(4):365-377.
‘Vangen S, Huxham C. 2003. Nurturing collaborative relations: Building trust in
interorganizational collaboration. The Journal of Applied Behavioral Science
39(1):5-31.
Wenger E. 1998. Communities of practice: Learning, meaning and identity. Cambridge
University Press: Cambridge.
Zand DE. 1972. Trust and managerial problem solving. Administrative Science Quarterly
17(2):229-239.
Zucker LG. 1986. Production of trust: Institutional sources of economic structure., in
Research in organizational behavior, Staw BM, Cummings LL (eds.). JAI Press
Inc.: Greenwich, CT, 53-111.
21
