Science Meets Policy Practice 1
Science Meets Policy Practice’
Peter S. Hovmand, Ph.D.” David F. Gillespie, Ph.D.
Washington University Washington University
Diane McFarland, MSW Enola Proctor, Ph.D.
Missouri Dept. of Mental Health Washington University
Benton Goon, MSW Barbara Levin Sally Haywood, MPA
Missouri Institute for Mental Health Washington University Washington University
Abstract
Recent discussions in system dynamics have raised questions about the
scientific acceptability of system dynamics and impact of system dynamics on
social problems. This paper describes two different projects in the area of mental
health services research—a research study recognized for its science and a
consulting project making an impact on the policy process—and the institutional
structures forming the university-agency-state partnerships. Together, the two
projects helped create a community with greater awareness of system dynamics,
interest in asking new questions from a system dynamics perspective, and
motivation to pursue additional system dynamics projects related to research and
policy. Key features of the individual projects, Center for Mental Health Services
Research, Alliance for Building Capacity, and the community are discussed. The
paper concludes with implications for future work in mental health services
research and system dynamics in the nonprofit sector.
Keywords: social science, public policy, health, mental health, group model building
' The projects described in this paper were supported by the National Science Foundation (Hovmand, PI; Gillespie,
Co-PI; SES-0724577); the Center for Mental Health Services Research, George Warren Brown School of Social
Work, Washington University through an award from the National Institute of Mental Health (Proctor, PI; P30
MH068579); and the Missouri Transformation Project, funded through a Substance Abuse Mental Health Services
Administration Mental Health Transformation State Incentive Grant (McFarland, Chair; Goon, Co-Chair; SM57474-
01).
2 George Warren Brown School of Social Work, Washington University in St. Louis, Campus box 1196, One
Brookings Drive, St. Louis, MO 63130, USA, (telephone) 314 935 7968, (email) phovmand@ wustl.edu
Science Meets Policy Practice 2
In the wake of the 50" anniversary of system dynamics, questions were raised about why
system dynamics has not had more of an impact with government or played a major role in
addressing major problems. Forrester (2007) raised some of these questions himself and
identified a number of causes including the careerism in academia, the lack of courage in
political leaders, the absence of popular works and movements that created political pressure for
change, the mistaken belief that having technical answers presented to political elites is enough,
and the misguided efforts to substitute systems thinking for simulation and elevate the role of
historical fit at the expense of deeper insights. It therefore seems important to present examples
of work that highlight the interchange between science and policy practice when it does happen.
In this paper, we set out to describe two projects that have created a bridge between
social science and policy practice and the institutional supports creating them. The first project
involves a research grant funded by the National Science Foundation. The second is a series of
contracts with the Missouri Department of Mental Health to support policy work on
transformation of mental health services. The projects were initiated by previous work and
collaborations facilitated by the Center for Mental Health Services Research and Alliance for
Building Capacity, both situated within the George Warren Brown School of Social Work at
Washington University in St. Louis. While each project has its weaknesses, together they
created a community that is both engaged in basic science and interested in developing
simulation models for policy design. The two projects now coexist within a reinforcing feedback
loop, providing a useful example of institutional supports can play a critical role in forming
research partnerships with nonprofits and state agencies that can advance science, impact the
policy process, and motivate more rigorous work in system dynamics and policy analysis.
The paper is organized as follows. After first describing each of the two projects
separately, we describe how the two projects reinforce each other through an overlapping
community of participants engaged in both projects. A variety of factors are discussed as critical
to the development of excitement within this community, along with some reflections on what
helped us along the way, and implications for future directions in mental health and health
services research and policy using system dynamics.
Institutional Supports
Two units within the George Warren Brown School of Social Work played instrumental
roles in facilitating university-state agency and university-nonprofit partnerships, supporting
pilot work developing those partnerships, and facilitating the ongoing collaborations. The Center
for Mental Health Services Center (CMHSR) is a National Institute of Mental Health research
center. The focus of the CMHSR has been on promoting mental health services research on
topics such as advancing research methods for services research, developing practice based
research partnerships, and advancing evidence based practices (EBP) and implementation
science in mental health services. Federally funded research centers provide invaluable
institutional supports that bring together national experts through a series of scientific meetings
and collaborations on research projects, helping to simultaneously set the research agenda and
build the capacity of researchers around the country to pursue that agenda through externally
funded research. As such, they bring together local and regional policy makers and practitioners
together with university based researchers, and opportunities for forming new research
partnerships on well focused substantive issues amenable to scientific research and evaluations.
Science Meets Policy Practice 3
In addition to the CMHSR, the Alliance for Building Capacity (ABC), an endowed
program at the George Warren Brown School of Social Work, seeks to work with small
nonprofit organizations in the St. Louis area. ABC has provided a variety of technical assistance
services and professional development for human service professionals working in nonprofits.
These include workshops and assistance on board development, strategic planning, budgets,
social entrepreneurship, developing business plans, and supervising students through field
practicum experiences with an interest in nonprofit management. ABC provides a unique entry
point for nonprofit organizations seeking to access the university and develop university-
nonprofit partnerships.
Science
The first project is a research grant to investigate the impact of innovation
implementation on organizational performance in the field of mental health, funded through the
National Science Foundation (NSF). An overarching aim of the study is to understand
something basic about organizational behavior using system dynamics simulation research. The
inspiration for this grant began in 2004 with the Diffusion of Innovative Practices study
(Hovmand and Proctor, Co-PI) funded through the Center for Mental Health Services Research
as a pilot. The aims of the pilot were to 1) identify the barriers and facilitators that
administrators and clinical directors faced with implementing evidence based practices, and 2)
develop collaborative relationships for a subsequent implementation study.
While the study team identified a number of facilitators and barriers, we also found that
administrators and clinical directors had expectations of evidence based practice that went
beyond the client. Specifically, managers wanted strategic outcomes for the organization as
well. For example, informants described how they expected implementation of EBP to
legitimate their services to stakeholders, anticipated that as consumers had more control over the
resources to pay for services they would be demanding EBP, or wanted to be known as regional
center of excellence. What struck us and our colleagues about this was how few people were
looking at the organizational outcomes from implementing EBP. So we (Hovmand and
Gillespie) decided to design and submit a grant application to the National Science Foundation to
study this.
Our initial work involved reviewing system dynamics models related to innovation
implementation and organizational change. For example, we drew on the early work by Levin
and Roberts (1976) and their dynamic theory of human service organizations, Samuel and
Jacobson’s (1997) work on planned organizational change, Sastry’s (1997) model of Tushman
and Romanelli’s (1985) theory of punctuated change, and Repenning’s (2002) work on
managerial commitment in implementation of Total Quality Management. We invested nearly a
year building, testing, integrating, and then retesting these models. Ambiguities arose between
different uses of organizational constructs that had to be ironed out, but as a consequence our
theory developed. We compared our constructs and mechanisms with the initial key informant
interviews and presented our results to colleagues for feedback (Hovmand and Gillespie 2006),
and then tested and refined our study questions pre-submission, validating via simulation the
logical consistency between our theory and the questions we were posing (Hovmand and
Gillespie 2007).
We submitted a grant proposal for a multiple case study design of 40 nonprofit
organizations providing mental health services to the National Science Foundation (Hovmand,
Science Meets Policy Practice 4
PI; Gillespie, Co-PI; SES 724577). Data sources would include key informant interviews,
longitudinal financial data from IRS 990 returns, organizational surveys, agency documents, and
group model building sessions. The three-year study would seek to build a system dynamics
simulation model of implementation and its effects on organizational performance. During the
preparation for this grant, we sought letters of commitment from the Department of Mental
Health to help us identify organizations. While they were supportive, they were not quite sure
what system dynamics was about or how this would work. We were asking questions about the
long term consequences of implementation, whereas most policy makers and researchers were
focused on simply getting the practices implemented. So our study seemed distant from the
immediate problems of implementation.
Figure 1 Overview of main feedback mechanisms
in the model of the implementation and organizational performance
Required Strategic
——
Adoption of EBP . Direction
+
_—— > Gap
Strategic Direction” *
Py Effectiveness
+
Desired increase in + Pressure to \
performance ~ > Change Ao)
Community
Reorientation Support
Perceived e 7
Performance Community
+ support
Resources
Efficacy By
Performance 7
+
5 Reliability
(mf Efficiency a
Quality
improvement
We wrote up the preliminary work and presented a talk about system dynamics later that
spring at the Missouri Institute of Mental Health. The paper and presentation focused on a brief
introduction to the problem, system dynamics as a research strategy, and discussion of the
preliminary simulation results. The story of implementation and its effects could be told in terms
of three feedback mechanisms: strategic reorientation, community support, and quality
improvement (see Figure 1). In showing this simple diagram and the simulation results, we were
able to convey in a relatively brief time some of the key issues organizations faced as they
sought to adapt to their environment and improve outcomes. The take-home point was that
implementation of innovations that produce better clinical outcomes do not always benefit the
organization in the long run, and that the outcomes depend on how the initial performance was
achieved.
Science Meets Policy Practice 5
What struck us, however, was how important a simulation was in illustrating the results
and made what seemed like a far-fetched idea three months ago become part of the practical
discussion for implementing policy. People were now referring to this simulation result in
conversations, not as a prediction of what would happen, but as a possibility that needed to be
considered as leaders planned various kinds of changes within their organization. By summer of
2007 we had received favorable scientific reviews from NSF that emphasized the argument for
using simulation, the focus on developing heuristics, use of triangulation, and using empirical
data to parameterize the model.
Missouri Transformation Project
In 2003, the President’s New Freedom Commission on Mental Health called for an
overhaul of the system of mental health services in the United States. The report urged states to
transform the mental health service system from one driven by bureaucratic and financial
incentives to one driven by the needs of consumers and families that focuses on facilitating
recovery and building resilience (DHHS 2006). In response to this challenge, the Substance
Abuse and Mental Health Services Administration (SAMHSA) proposed to invest $173 million
in the form of five-year state incentive grants to help transform mental health services. Nine
states have been awarded these grants, which provide support for strategic planning and capacity
building to transform mental health care to a system that 1) recognizes that mental health is
essential to overall health, 2) is driven by consumers and family members, 3) eliminates
disparities, 4) facilitates early mental health screening, assessment, and referrals to services, 5)
accelerates research, and 6) uses technology to access mental health care and information (New
Freedom Commission on Mental Health 2003). In 2006, Missouri was awarded one of the state
incentive grants from SAMHSA (McFarland, Chair; Goon, Co-Chair, SM57474-01).
The emphasis on eliminating disparities and making services consumer and family driven
meant that for changes to be sustainable, a broad based of participation was needed in the
planning process, from consumers and family members to providers and policy makers. This is
especially important given that quality of mental health care often depends on local support, and
that statewide solutions that ignore local needs are likely to exacerbate disparities.
Consequently, the leadership team developed a statewide planning process that would eventually
include approximately 240 individuals working in six groups, each group focused on developing
recommendations for one of the six areas identified by the New Freedom Commission Report.
The coordinators of the project designed the planning process in collaboration with a private
consulting group specializing in managing change in state agencies.”
Each workgroup consisted of approximately 40 members with a chair and co-chair
selected from the workgroup. Members were invited by the planning group to participate in the
process to ensure adequate geographic and stakeholder representation and include key
individuals who would later be important for implementing the recommendations. Participants
included consumers and family members of persons with mental illness, substance abuse
disorders, or developmental disabilities; advocates for persons with mental illness, substance
abuse disorders, or development disabilities; providers; representatives of hospital associations;
administrators from various state agencies including corrections, youth services, mental health,
3 Ken Miller and Blake Shaw of Change & Innovation Agency, Jefferson City, MO, USA.
Science Meets Policy Practice 6
and protective services; and researchers and academics. Each workgroup was expected to meet
every two weeks for three hours over a period of approximately three months.
One of the issues the planning group was facing was how to start the groups and motivate
a holistic view of the mental health transformation effort. Although participants in each
workgroup were familiar with some aspect of mental health, most had only a limited view of the
whole system. This was seen as both a consequence of system fragmentation and a barrier to
change. If participants were unable to move from their local views of the problem of
transforming mental health care to a global perspective, then the effort was likely to fail. In
particular, opportunities were sought to illustrate how efforts that might seem locally rational
were in fact contributing to a systems problem. The planners also wanted some way to
distinguish this transformation effort from earlier efforts to reengineer or change mental health
care.
Policy Practice
The planners were familiar with and drew on Peter Senge’s (2005) recent book Presence
with Otto Scharmer, Joseph Jaworksi and Betty Sue Flowers. They wanted a way to help people
suspend individual agendas and see the whole. The challenge was that while people refer to “the
mental health system” it was often experienced as a highly fragmented set of services spanning
multiple sectors.’ Initially, it was unclear if system dynamics could be of much use for this
project. However, the planners saw the potential after the first author shared some examples of
earlier simulation work and group model building from a technical assistance project with Save
the Children UK in Ulaanbaatar, Mongolia.°
The initial contract called for group model building exercises with each of the five
working groups in their respective areas: mental health is essential to health, evidence based
practice, reducing disparities, easy and early access, and consumer and family driven services.
Each session was to be conducted with approximately 40 people during the second half of their
first meeting. A four person modeling team was convened including a modeler/facilitator,
facilitator, and two recorders. Several scripts (Andersen and Richardson 1997) were piloted and
then a final set of scripts selected based on the practice sessions. All five sessions were
conducted in the same week with approximately 200 participants across the five groups. Figure
1 shows an example of the type of causal loop diagrams created. Word got out about the sessions
and toward the end of the week some participants were eager to know more about the exercise.
Follow-up work included identifying major themes that cut across all five sessions and
identifying possible next steps for a second phase of work and contract.
A second contract was drawn up to integrate the five diagrams into a single model, begin
development of a basic population model, and map and trace the approximately 180 policy
recommendations from the five working groups. This mainly involved the core working groups
consisting of the chairs and co-chairs from each of the five working groups. While not a
simulation model, mapping the policies and tracing their causes to the population flows proved
insightful and of great value to the core working group.
The core working group was faced with the problem of synthesizing the
recommendations whether they had the benefit of a system dynamics model or not. Identifying
the policies that were proximate to each population flow helped them see some of the challenges
“ This is similar to the point that Shaffer (1976) makes about the criminal justice system.
5 With Dr. Gautam Y adama, George Warren Brown School of Social Work, Washington University in St. Louis.
Science Meets Policy Practice
and more realistically consider the benefits of some ideas. They were also able to more clearly
see how a large set of policies were concentrated on specific areas such as coordination and
collaboration that had proved resistant to change in previous efforts. These results were
presented to the core working group in June 2007. One participate exclaimed that this was “total
value added” and the group decided that it would be beneficial for attendees at the July
Leadership Meeting see a similar presentation. The Leadership Meeting was the main decision
making body that would take the recommendations, develop the specific plan,
recommendations into legislative action.
Figure 2: Example of Causal Loop Diagram from
Consumer and Family Driven Services Workshop
Communication,
‘and marketing
eter outcomes services
feeording ©
a ‘murs. aoe “\
Having voice and
being heard
a Mentoring
Expansion offeer nd ’
tm peer services coaching S
f 7 More colfboration
‘aries a eadership SP eaten ‘ang coraton ~
a ay recovery
airs tent i.
~
Legislative
parameters hs
pT consumer and
Consumer
ie ‘amily den
thatguie taining of
professionals A Training of
Senices no
anger consumer
and family arven
consumer and
family diver
Senices nat
Money used in
Previous
enperence
RDO program
Dedication of
professionals
“Accessibilty of —~
people to services
Consumers and
families cont
money
ducaton
Stigma of
mental iiness
ri of Ee. PrScovery and
Contity win
/ “hs Dero parents
i y and cutie
eas ‘
‘pian = \
and tum the
'
Moving trom
management
proactive
Science Meets Policy Practice
Figure 3 Community Preparation (R3) and Stigma (R4)
Community Natural P opulations losing
preparation 7 supports “> appropriate services
Stigma ~
< R3 7 - R4
Public awareness of
mental illness
7+ Outcomes
Consumers + + according to
and Families Consumer Quality of fanillas hd
in Leadership Empowerment life caneuMers
Figure 4 Shortage of Mentors (B1), Losing Fidelity (B2), and
Slowing A doption and Implementation of EBP (B3)
Populations
Consumers
Losing
and families
Appropriate 7 in leadership
Services
2)
B1 Discontinue
je eer SY
B3 ) Caseload
Training of + + Mental Adopting and
Professionals, Mentors for Health implementing —
Support Staff, Implementing EBP. Workers EBP
and Supervisors Using EBP * 4
Sages
Science Meets Policy Practice 9
After the Leadership Meeting, a summary report based on the conversation during the
meeting was prepared. This report included a review of the overall process and description of
some of the key feedback mechanisms that participants had identified and discussed during the
previous 7 months and documents, and eventually included as an appendix into the actual plan.
Figures 3 and 4 below are two examples of the types of diagrams used to describe these feedback
mechanisms. These were based on a single integrated model of mental health transformation
that included a stock-flow population model with aging chains. Each of the population stocks
shown in Figures 3 and 4 were arrayed to include children, adults, and older adults. This tumed
out to raise some important issues as people noted, for example, that the greatest risks for
children are not necessarily the parents who have mental illness, but the parents who are at risk
of mental illness since these tend to occur earlier and there are more parents in this stock. Asa
consequence, this led to a more nuanced view of risk and prevention across the life-span since
adults at risk come from not just the adverse events they experience as adults, but also children
who are at risk and age into the stock of at-risk adults. The results were included as an appendix
to the state-federal cooperative agreement to transform mental health services in Missouri
(Hovmand 2008).
Science Meets Policy Practice
While each of these two projects is interesting in its own right and both continue, the
focus in this paper is on how the overlap in the community participants between the two projects
created unexpected opportunities and motivation to both participate in the NSF project and
pursue the development of a more rigorous system dynamics model of mental health
transformation for policy analysis. Figure 5 summarizes this relationship with the community of
participants coupling the two projects.
We first noticed this effect when we began our recruitment of nonprofit organizations in
the NSF study. Many of the people we met had attended one of the transformation workgroup
meetings and remembered the exercise or had a member of their executive team who had
participated. Most expressed enthusiasm for participating in more research based on this
experience and were eager to get more involved with the development of a simulation model and
learn more about system dynamics. And while they had been exposed to system dynamics, we
had also developed a better understanding of the relevant mental issues at the state and
community level. Importantly, our view was grounded in their collective experience as opposed
to a particular agenda. As a consequence, we had a much broader view of what might constitute
mental health services and supports, had a better idea of how services and supports were
organized, and understood some the barriers that they faced in their day to day operations.
Organizations also expressed a sense of trust that was reinforced when we discussed the
importance of the results being “open source” that could be used by them for a variety of
purposes without restriction. It was not that we had arrived at some significant insight, but that
we were able to effectively link questions about the dynamics of the problem we were studying
into their immediate context. When we did this, they not only saw the question, but its
immediate practical relevance to the questions they were facing. Some reasoned out loud that
they were already thinking about these questions, and so why not do it with the benefits and
insights that might come from working on a research project.
Science Meets Policy Practice 10
Figure 5
Model based insights
Recognizing challenges
Motivation to simulate transformation
Partnerships and collaboration
Community
Participants
Consumers, family r
members, providers, Transformation
policy makers, executive project
directors, advocates,
researchers, etc.
NSF
project
Public trust
Previous exposure to SD and GMB
Motivation to participate in research
Relevance to policy environment
The work on the NSF project and related smaller models also helped participants in the
transformation project put the qualitative models into context as well as motivate interest in
developing a simulation model of transformation. The ability to construct a simulation model of
mental health transformation depends greatly not only the skill of the modelers or motivation of
client, but also depends on the community of people who are interested in participating and can
effectively utilize the results and evaluate their impact. The NSF project has over 40
organizations and 60 people enrolled in the study. Some will be mainly involved as key
informants, but many have expressed interest in participating in group model building sessions
and subsequent research projects. In fact, we now have a nice problem of there being more
interest for more research projects than we could ever have imagined, and this has reshaped our
research agenda as organizations and state policy makers are now eager to use system dynamics
to address a variety of problems that they are facing.
Through this pair of projects, we have managed to simultaneously field a study looking at
basic questions of organizational behavior and impact the policy process. The critical feature of
driving the acceptance of both projects has been on developing a community that is motivated to
use system dynamics and simulation research as important tools for answering difficult questions
about how to improve the quality of services and supports in mental health. This community
was in effect able to create a new language for thinking about change made possible through the
large number of people exposed to system dynamics in both projects.
Equally important is the fact that as this community of participants developed, they began
to ask frame questions different and consider new projects. For example, people started to
explore how system dynamics might be used to think about the problem of implementing school
based mental health, or how the approach could be used to reduce disparities in mental health, or
how one could use the approach to support local communities in their planning of mental health
services.
Lastly, the two projects together were able to help people begin to question their
cherished policies and beliefs about how to deliver services. We helped create a different kind of
Science Meets Policy Practice 11
conversation—a framework—where someone could say “We know someone’s ox is going to get
gored” and begin to deal with an incredibly complex problem more honestly.
Conclusion
We attribute some of our success to always remembering a basic point about system
dynamics that it is fundamentally about improving the mental models of decision makers over
what they might have used otherwise. Seeing the whole might not seem impressive in
comparison to a running simulation model, but it is an essential step in the modeling process if
one is going to develop a model that will be supported by community leaders, consumers, and
their family members. Throughout all this work, we have continually retuned to the question of
what would decision makers have used otherwise.
Equally important to building this community has been a commitment to capacity
building. For example, being situated within a university, we have sought to create opportunities
for social workers to pursue intemships and take our system dynamics courses. We have
presented workshops to the community on system dynamics and talks to local institutes. In our
field work with the organizations, we have emphasized the multiple opportunities for them and
their staffs to learn more about system dynamics whether they want to participate or not in our
study. We have talked with schools and programs with at-risk teens to explore ways that system
dynamics could play a role. We have helped connect their causes and participated in their
movements to address some of the critical barriers, such as supporting petition drives for a local
children’s mental health tax levy. In short, we have not only sought them out as participants in
research, but been actively engaged in the community. They have in tum had higher
expectations of us, and impressed upon us the responsibilities we have to bring these
conversations forward to impact local communities.
This has created whole new sets of opportunities that we could not have imagined 18
months ago. Recruiting and partnering with organizations to work on implementation issues in
mental health has been recognized as one of the most formidable barriers to advancing
implementation research. We now work with a community of diverse organizations with
interests in pursuing more system dynamics. The first two projects with the Department of
Mental Health have created a desire to develop a third project. These are now in the process of
coming together in plans for new studies that aim to use system dynamics simulation modeling
for understanding and planning implementation strategies to reduce disparities and improve the
quality of mental health services. We are in a position to imagine that 10 years from now if not
sooner, community mental health planners may be drawing on a set of canonical simulation
models to design programs and organizations to address gaps in mental health services, or use
simulation models to design and conduct evaluations of changes to mental health services, or
draw on system insights derived from experience with these models to work with communities to
design and support key policy initiatives such as a school based mental health. We envision a
time when researchers proposing a grant to test a new implementation intervention in a
community would have conducted a simulation study prior to submission in much the same way
that we conduct a statistical power analysis today. Ultimately though, we foresee a fundamental
shift in how mental health services and supports are provided in the United States and globally,
and a welcomed role for system dynamics as a tool to facilitate that transformation and
developing research partnerships.
Science Meets Policy Practice 12
References
Andersen, David F., and George P. Richardson. 1997. Scripts for group model building. System
Dynamics Review 13 (2):107-129.
DHHS. 2006. Trends in mental health systems transformation: The states respond. Rockville,
MD: Center for Mental Health Services, Substance Abuse and Mental Health Serices
Administration.
Forrester, Jay W. 2007. System dynamics-the next fifty years. System Dynamics Review 23 (2-
3):359-370.
Hovmand, P. S. 2008. Appendix C: Causal mapping. In Cooperative agreements for mental
health transformation state incentive grants. Requests for applications no. SM-05-009.
State of Missouri Comprehensive Mental Health Plan, April 2008. Rockville, MD:
Substance A buse and Mental Health Services Administration, U. S. Department of
Health and Human Services.
Hovmand, P. S., and D. F. Gillespie. 2006. Dynamics of innovation implementation in social
service organizations. Paper read at Intemational System Dynamics Conference, at
Nijmegen, Netherlands.
——. 2007. Dynamics of innovation implementation and organizational performance in
mental health services. Paper read at 25th International Conference of the System
Dynamics Society and 50th Anniversary Celebration, at Boston, MA.
Levin, Gilbert, and Edward B. Roberts. 1976. The dynamics of human service delivery.
Cambridge, MA: Ballinger Publishing Company.
New Freedom Commission on Mental Health. 2003. Achieving the promise: Transforming
mental health care in America. Final Report. Rockville, MD.
Repenning, Nelson P. 2002. A simulation-based approach to understanding the dynamics of
innovation implementation. Organization Science 13 (2):109-127.
Samuel, Yitzhak, and Chanoch Jacobsen. 1997. A system dynamics model of planned
organizational change. Computational and mathematical organization theory 3 (3):151-
171.
Sastry, M. Anjali. 1997. Problems and paradoxes in a model of punctuated organizational
change. Administrative Science Quarterly 42 (2):237-275.
Senge, Peter M., C. Otto Scharmer, Joseph Jaworksi, and Betty Sue Flowers. 2005. Presence: an
exploration of profound change in people, organizations, and society. New Y ork, NY:
Currency.
Shaffer, William. 1976. Court management and the Massachusetts criminal justice system,
Alfred P. Sloan School of Management, Massachusetts Institute of Technology,
Cambridge, MA.
Tushman, Michael L., and Elaine Romanelli. 1985. Organizational evolution: A metamorphosis
and model of convergence and reorientation. In Research in Organizational Behavior,
edited by L. L. Cummings and B. M. Staw. Greenwich, CT: JAI Press.
