212
Chapter XII
Evolutionary Diffusion Theory
Linda Wilkins
RMIT University, Australia
Paula Swatman
University of South Australia, Australia
Duncan Holt
RAYTHEON, Australia
AbsTRAcT
Improved understanding of issues affecting uptake of innovative technology is important for the further
development of e-business and its integration into mainstream business activities. An explanatory theory
that can provide a more effective instrument for determining acceptance levels should therefore be of
interest to IS practitioners and researchers alike. The authors aimed to establish whether evolutionary
diffusion theory (EDT) could offer such an instrument, developing a set of axioms derived from the EDT
literature and applying these to an in-depth review of two e-business implementations: a G2B document
delivery system introduced by the Australian Quarantine Inspection Service (AQIS) across a number of
industry sectors; and an enterprise-wide system implementation in a local government instrumentality.
The authors found EDT offered remarkable explanatory depth, applicable not only to analysing uptake
of complex, multi-user technologies in organisational settings but to any e-business investigation requiring a system-wide perspective.
InTRODucTIOn
A variety of theoretical frameworks and approaches have been used to study Information
Systems (IS) diffusion processes (see, for example,
Holbrook and Salazar, 2004; Baskerville and
Pries-Heje, 2001; Edquist, 1997). Investigations
of a number of these theories and models of Innovative Technology Uptake (ITU) have found
that each has only a narrow perspective which
tends to capture ‘just one part of the story’ and
only highlights particular areas of interest. No
Copyright © 2009, IGI Global, distributing in print or electronic forms without written permission of IGI Global is prohibited.
Evolutionary Diffusion Theory
single theory appears uniquely able to explain
the circumstances of any particular case (Jones
and Myers, 2001 p.1018).
Despite these limitations, influences on uptake
and diffusion of IT innovations are of perennial interest to IS researchers. Those attempting to make
progress in identifying key issues affecting uptake
have had to grapple with the limited explanatory
power of recognised diffusion theories over some
four decades. The most commonly cited diffusion
theory in the IS literature is Rogers’ Classical Diffusion of Innovation (DoI) theory, first published
in 1961 (Clarke 1999). Rogers originally focused
attention on the shape of the diffusion curve,
describing innovation as a process that moves
through an initial phase of generating variety
in technology, to selecting across that variety to
produce patterns of change resulting in feedback
from the selection process, to the development of
further variation (Rogers, 1995).
As a pioneering contribution to conceptualising adoption and diffusion, Classical DoI theory
appears to have maintained its iconic status over
time and continues to be cited in the IS literature,
despite the fact that interest in innovation studies
has moved on from the shape of the diffusion curve
to a focus on articulating underlying dynamic
mechanisms (Lissoni and Metcalfe, 1994; Nelson,
2002). The innovation ‘journey’ now appears
to be more readily understood as a non-linear
dynamic system, far less predictable and stable
than staged models based on Classical DoI theory
represented it to be (see for example Van de Ven
et al., 1999). The static orientation of Classical
DoI theory, its focus on individual firms and a
‘single innovation’ perspective has diminished its
relevance to the IS field and to the development
of online technologies in particular.
The limited explanatory power of Classical DoI
theory is well documented in the literature (see,
for example, Downes and Mohr, 1976; Moore and
Benbasat, 1991; Damsgaard and Lyytinen, 1996;
Galliers and Swan, 1999; Clarke, 2002). Seminal
work in the IS field (for example: Orlikowski and
Hofman, 1997; Boudreau and Robey, 1999; Reich
and Benbasat, 2000) has also clearly established a
need for analytical theory in this field which:
•
•
•
•
Aligns more closely with the way beliefs,
attitudes and understanding of plans and
structures are known to influence organisational decision-making
Can articulate underlying dynamic mechanisms intrinsic to adoption and diffusion
processes
Addresses how complex and networked
technologies diffuse
Acknowledges the uncertainty and surprises
that mark the ITU process
The IS field needs a unified theory to identify
key influences on uptake of innovative technology
that is: appropriate to reviewing open-ended and
customisable innovations associated with uptake
of e-business technologies; takes into account issues of discontinuing practice or slowing uptake
of inappropriate technologies; acknowledges
the active role users can play in the innovation
process; and allows for changes in an innovation
during the adoption and implementation process.
Such a theory must also be readily applicable in
organisational settings featuring the adoption of
complex, multi-user technologies – where the
majority of potential applications of diffusion of
innovation now occur. An analytical instrument
which appears well suited to these requirements
is Evolutionary Diffusion Theory.
In this chapter, we begin by exploring the
origins and principles underpinning Evolutionary
Diffusion Theory, before turning to examine its
explanatory strengths. We will make close reference to two case studies that illustrate the relevance
of EDT to the field of Information Systems (IS)
and, in particular, to the explanation of innovative
technology uptake.
213
Evolutionary Diffusion Theory
eVOluTIOnARy DIffusIOn
TheORy (eDT)
The limitations of standard theoretical approaches
to analysis of the influences on ITU in the field of
Information Systems research have been increased
by the fragmentation and minimal ‘diffusion’ of
diffusion research itself. The lack of awareness
which the various diffusion research traditions
have of one another’s work has hindered researchers in their attempts to grasp and frame the ITU
problem (Rogers, 1995 p.38). Such communication
gaps across the disciplines help to explain why
Evolutionary Diffusion Theory does not feature
among the analytical instruments referred to in
the mainstream IS diffusion literature.
evolutionary Diffusion Theory:
Development and evolution
Evolutionary Diffusion Theory (EDT) emerged
from Evolutionary Economics, a discipline which
describes economic phenomena and deals, in
particular, with situations of change, open systems and innovation processes (Nelson 1995). As
a discipline it has had a major impact on more
recent studies of firms, industries, and technical
change. The idea of technological advance as
an evolutionary process has been developed by
scholars operating independently in a variety of
different disciplines. These disciplines include,
but are not confined to: sociology (Constant, 1980;
Bijker, 1995); technological history (Rosenberg,
1969; Mokyr, 1996); and economic modelling
(Nelson and Winter, 1982; Saviotti, 1996; Metcalfe, 1994).
Evolutionary Economics—and Evolutionary Diffusion Theory (EDT) in particular—are
relatively recent developments, with the bulk of
the EDT literature having been published only
since the early 1980’s. The Nelson-Winter classic
evolutionary model of technological change (1982)
pioneered a relatively simple conceptual model of
Evolutionary Economics which was crucial to the
214
development of Evolutionary Diffusion Theory.
The model played a prominent role in defining
a paradigm for further research into the conditions which determine industrial concentrations,
dynamic competition in alternative technological
regimes and the relationship between innovators and imitators (Andersen, 1996). The model
demonstrated the possibility of collating a wide
diversity of elements and integrating them into
an evolutionary process which could then be applied to understanding the uptake of innovative
technology. Key elements include: the processes
of transmission; variety creation; and selection
(Nelson and Winter, 1982, Chs. 4-5).
Nelson and Winter’s classic evolutionary model
of technological change (1982) was followed by
several overviews of Evolutionary Economics
(see, for example, key publications by Dosi et
al, 1988; Saviotti and Metcalfe, 1991; Dosi and
Nelson, 1994; Andersen, 1994; Nelson, 1995).
These contributions to the Evolutionary Economics literature share a concept of innovation
as a process that moves through an initial phase
of generating variety in technology, to selecting
across that variety to produce patterns of change
resulting in feedback from the selection process,
to the development of further variation under
continual injections of novelty (Dopfer, 2001).
Researchers within this paradigm now apply their
attention to enduring issues in innovation studies;
and to finding reasons for unexplained outcomes
from technology adoption and diffusion.
Instead of the more traditional application of
DoI theory to an individual firm, EDT reviews
the impact of diffusion theory when applied to the
more complex environment of a market (Lissoni
and Metcalfe, 1994). Placing theory in a market
context raises questions of particular interest to
e-business practitioners, such as why all potential
users do not immediately adopt innovations which
appear to be advantageous compared to existing
technology (instead, some firms as potential users
adopt later or not at all); and why some agents who
can be identified by their spatial location always
Evolutionary Diffusion Theory
Table 1. Features of evolutionary diffusion theory
Evolutionary Diffusion Theory
Explains the innovation process as non-linear and rarely predictable
Explains innovation and diffusion in market environments
Presents the policy maker’s role as one of stimulating the building of innovative infrastructure
cooperatively with local institutions
Accepts the possibility of human intervention in the process of technology development
Describes the natural trajectory of an unpredictable original selection where the outcome is not
always the best one
Stresses the gradualism of internal adoption
Defines a clear role for Government as a policy maker coordinating institutions in innovative systems
and seeking solutions which are context specific and sensitive to local path dependencies
Presents adoption of single innovations as part of a greater process of change impacting on organisations and their culture
adopt later than others (Lissoni and Metcalfe,
1994, pp.106, 127).
Table 1 sets out features of EDT frequently
cited in the literature and selected for their clear
relevance and application to IS research on issues
affecting uptake of innovative technology.
We have now outlined the origins and background of Evolutionary Diffusion Theory and
provided a synthesis of its key features. The
following section of this chapter attends to the
explanatory strength of EDT. We will show that
EDT is particularly well suited to analysing issues
affecting uptake of innovative technology in the
e-business context.
evolutionary Diffusion Theory:
explanatory strength
Evolutionary Diffusion Theory offers the IS researcher a basis for reviewing innovative technology uptake (Kowol and Kueppers, 2003; Lambooy
and Boschma, 2001; Norgren and Hauknes, 1999;
Amin, 1998; Lissoni and Metcalfe, 1994; Saviotti
and Metcalfe, 1991) which includes:
•
•
A focus on unexplained outcomes
An emphasis on gradualism of internal
adoption
•
•
•
•
A concern with development and diffusion
of new variety in market environments
Acceptance of input from a variety of disciplines
Acceptance of human intervention in technology outcomes
A clearly defined role for policy makers
fostering innovative technology uptake
EDT provides an analytical instrument to examine issues in the uptake of innovative technology in specific cases. The first of two case studies
analysed in this chapter features the introduction
of EXDOC, an innovative online technology.
EXDOC was implemented by the Australian
government agency AQIS (Australian Quarantine
and Inspection Service) to support the needs of
food producers across a number of sectors for
access to export documentation. These producers
subsequently became the EXDOC community
and the population of the case study. Once key
features of EDT had been synthesised from the
literature and their explanatory strength for the
IS/e-commerce context established, it became
a matter of deciding how best to apply EDT as
the instrument to analyse ITU issues within the
case study. The second case study used EDT to
review the implementation of an enterprisewide
215
Evolutionary Diffusion Theory
electronic records management solution for a
local government authority – an instance of this
theoretical approach’s relevance to individual
organisational analysis, despite its published focus
on market-wide innovation.
Four axioms derived from the literature of
Evolutionary Diffusion Theory provided the
required analytical instrument. Key features of
each of the four EDT axioms is described in detail
in the following section of this chapter.
An eDT-based Axiomatic model
According to the EDT literature, key features of
Innovation Diffusion Theory are: the rejection of
optimisation or the feasibility of determining one
‘best’ policy; a focus on systems and markets rather
than individual firms; the acceptance of human
agency in technology development; and the role
of government as policy maker. These features
are described and explained in some detail as
the basis of the four EDT axioms which formed
the analytical instrument used to examine the
case study data. Elements related to each of the
four axioms were then drawn from the EXDOC
case study.
Figure 1 sets out the four axioms derived from
Evolutionary Diffusion theory. The elements
describe the features associated with each axiom
which are pertinent to understanding influences
on the uptake of innovative technology.
Rejection of Optimisation: Innovation Diffusion theory rejects the idea of optimisation or
implementing one ‘best’ policy at both macro and
micro levels. At the macro level, rejection of the
possibilities of optimising is one of the strongest
points of differentiation in the conceptualisations
that have emerged from Evolutionary Economics
(Metcalfe, 1994). ‘If one wants a model in which
it is presumed that the actors fully understand the
context ... then the formidable challenge facing the
‘rational’ models let alone a supposedly ‘rational’
actor is what it means to ‘ fully understand’ the
context, whenever the latter depends in some
216
complex, non linear ways on the distribution of
micro decisions and on chance and is always full of
surprises’ (Dosi and Nelson 1994, pp.163-164). Instead of optimisation, the Evolutionary Diffusion
model presents the idea that a diversity of policies
is necessary to allow for a variety of development
paths (Amin, 1998). Economists argue that path
dependency1 means that systemic technological
and innovative capabilities can only be enhanced
by openness of competition, lowering barriers to
innovative entry; and nurturing interactive learning as a source of innovation (McKelvey, 1997).
Evolutionary Diffusion Theory also rejects the
assumptions that: individual firms behave optimally at the micro level; that adoption is a one-off
decision; and that uptake of new technology is
more or less instantaneous. Innovation processes
are described instead as: multi-referential; nonlinear; depending on various framing conditions;
and rarely predictable (Kowol and Kueppers,
2003). Once diffusion is viewed as a selection
process it becomes evident there is no guarantee
that more optimistic or better-informed firms will
adopt earlier. Successful innovations represent the
outcome of multiple and contingent variables and
do not always have to be the best ones (Saviotti
and Metcalfe, 1991). Models based on Evolutionary Diffusion Theory stress the gradualism of
internal adoption between firms and over time.
‘Firms called ‘adopters’ at a given time actually
differ in the extent of their commitment to the new
technology ... some of them may subsequently
reverse their adoption decision’ (Lissoni and
Metcalfe, 1994, p.108).
Market Focus: Evolutionary Diffusion Theory
is primarily concerned with the development
and diffusion of new variety or innovation in
an economic system. Evolutionary models serve
to extend traditional DoI theory from studies of
individual firms to explaining the impact and
effects of innovation and patterns of diffusion
in the more complex environment of a market
(Lissoni and Metcalfe, 1994). A focus on firms
alone misses the contributions and investments
Evolutionary Diffusion Theory
Figure 1. Axioms and constituent elements of evolutionary diffusion theory
EDT AXIOM:
Rejection of
Optimisation
EDT AXIOM:
Whole of Market
Perspective
EDT AXIOM:
Holistic View of
Contributions to
Knowledge Devlpt
EDT AXIOM:
Government as
Policy-Maker
Elements:
Uptake of Innovation as a non-linear and rarely
predictable process
Highly variable selection outcomes from uptake
Outcome of original random selection is not always
the best possible one
Elements:
Gradualism of adoption between industry sectors
Gradualism of adoption between firms in a sector
Focus on aggregate behaviour of a sample of firms
Attends to impact of external conditions on sectoral
responses
Information sharing and communication underpin
dissemination
Elements:
Knowledge development is influenced by
management style
Knowledge development is influenced by network
externalities
Elements:
Policymaking requires coordination of institutions
in innovation systems
Policy makers seek solutions that are contextspecific and sensitive to local path dependencies
Policy has the strongest impact when intervention
occurs early
of a wider population of stakeholders into relevant knowledge development. The literature of
Evolutionary Diffusion acknowledges the need to
encompass these additional sources of information
in explaining the diffusion process and thus shifts
attention to the aggregate behaviour of a sample
of firms, without necessarily relying on explicit
modelling of a single firm’s decision processes
(Nelson and Winter, 1982; Nelson, 1995).
Human Intervention in Economic Processes:
Implicit in Evolutionary Diffusion Theory is the
assumption that there is a possibility of intervening in the process of technology development;
and that the selection of a theory can influence
the design of policy. Evolutionary Economics
accepts the possibility of human intervention in
economic processes where ‘users are not exclusively selectors but also involved in the shaping
of innovations’ (Tushman and Rosenkopf, 1992).
Once the soft components of technology innovation are recognised, actors must be understood
as capable of consciously attempting to change
their environment (Nelson, 1995). Evolutionary
models draw on a number of reference disciplines
to expand the definition of technology to include
organisational and cultural elements as well as
human artefacts such as machinery and materials.
(Lissoni and Metcalfe, 1994.
Government as Policy Maker: The role policy
intervention can play and the institutional pressure
government can exert to stimulate ITU have been
explored by Lissoni and Metcalfe (1994). More
recently, the application of EDT to understanding
the role of government as policy maker has been
extended in work by Lambooy and Boschma
(2001) who emphasise the need for ITU solutions
217
Evolutionary Diffusion Theory
to be context-specific and sensitive to local path
dependencies. At a regional level, government can
use policy measures to stimulate technological
and innovative capabilities and minimise adjustment problems.
Clearly, there is no limit to the number of areas to which Evolutionary Diffusion Theory can
be applied. It would not be possible, however, to
discuss all—or even a sizable sub-set—within the
constraints of a single chapter. We have therefore
limited our examples of EDT applications to two
case studies of public sector implementations of
e-business technologies.
cAse sTuDy One:
ImplemenTATIOn Of A g2b
DOcumenTATIOn sysTem
(exDOc)
The Australian Quarantine and Inspection Service
(AQIS) developed EXDOC2 to support the preparation of food export documentation by Australian
primary producers, who require a health and/
or phytosanitary export certificate from AQIS.
Health Certificates generated by AQIS are the official means by which the Australian Government
certifies to an importing country that a product
meets that country’s import standards and regulations. The Phytosanitary Certificate is a type
of Health documentation testifying to the health
status of the certified product. AQIS procedures
ensure products meet Australian legislative and
importing country standards and requirements;
and EXDOC is an integral part of these procedures, providing greater certainty in certification
through the standardisation of documentation and
the consequently enhanced integrity of Australia’s
certification systems. The type of product and the
destination determine which (if any) certificates
are required. In December 2002, the move to harmonisation of regulatory systems and standards
gained impetus when. Food Standards Australia
New Zealand (FSANZ) assumed responsibility
218
for standards setting in the primary production
sector in Australia, so establishing for the first
time a single standards-setting body for the whole
of the food chain.
The system operated by AQIS has been in production since August 1992. Originally designed
for Meat exports, it was then redeveloped for use
by non-meat commodities and, by April 2000, had
been made available for exports from the Dairy,
Fish, Grain and Horticulture sectors. Key events
in the development of EXDOC by AQIS are set
out in Table 2.
The original version of EXDOC had been
developed by 1990. EXDOC relied on open EDI
systems at the data communication, application
system and document translation levels, in line
with the latest trends at the time. Internal AQIS
systems and processes had to be redesigned around
electronic trading and a number of organisational
changes were identified – such as the need to
replace the physical signature of an authorised
veterinarian with an electronic authorisation –
which necessitated process redesign at a number
of levels, including that of Human Resources.
In spite of significant problems, the phased
implementation of EXDOC survived and developed over time. This AQIS-led project stimulated
uptake of new technology, supporting improved
food industry supply chain management within
and across food sectors and increasing Australian competitiveness on world markets. The case
study also revealed that coordination of innovative technology implementations was strongly
affected by existing industry culture. Hence it
appears that policy maker interventions are most
successful where there is a real and recognised
need for policy change
A cross-disciplinary review of the literature
enabled the authors to synthesise key features of
EDT , develop four axioms and then apply them
to case study analysis. Table 3 sets out four EDT
axioms with their associated elements in order
to present a theoretical basis for analysing the
AQIS case study data from the EXDOC implementation.
Evolutionary Diffusion Theory
Table 2. Key events in the AQIS Development of EXDOC3
EXDOC
Development
Date
Description of Activity
1990 July:
Development of original EXDOC by AQIS (end date)
1991 Jan:
AQIS Reforms: Commercialisation of Services
1992 August:
EXDOC in production
1997 March:
Development work starts on new EXDOC and extra commodities
1997
The Nairn Report: 109 recommendations on improving quarantine – ‘a shared responsibility’
1998 July:
MEAT EXDOC implementation begins
1998 Nov:
DAIRY EXDOC implementation begins
1999 Nov:
FISH EXDOC implementation begins
2000 April:
GRAIN EXDOC & HORTICULTURE EXDOC begin
2002 March:
Minter Ellison Post-Meat Implementation Review reports
2003 June:
WOOL EXDOC & SKINS & HIDES EXDOC begins
(source: N.Scott, EXDOC administrator AQIS, October 2004).
Table 3. Applying EDT Axioms to the AQIS EXDOC implementation
Evolutionary Diffusion Theory
Explanatory Contribution to Case Analysis
Rejection of Optimisation
No predictable pattern of uptake.
Unexpected and lengthy delays delay diffusion across sectors after initial swift uptake
and industry support
Unpredictable (and costly) processes that frequently occur throughout the implementation
Innovation and diffusion from a whole-ofmarket perspective
Gradualism of internal adoption within firms in a sector
Differing rates of uptake across sectors
Impact of external conditions on sectoral responses
Holistic View of Contributions to Knowledge
Development
Importance of ensuring consistency and compliance with international standards
across sectors
Need to improve supply chain management across sectors in a globalised environment
Government as policy maker
Plays key role in:
Coordinating stakeholder institutions within innovation system
Seeking an acceptable business solution to ensure diffusion across sectors
Intervening with context specific solutions sensitive to local path dependencies
Analysis of the EXDOC case study confirmed
that it was impossible to predict a single ‘optimal’
policy development path. No single solution
or ‘magic bullet’ emerged for groups of firms
targeted for uptake of innovative technology in
the AQIS implementation. Uptake and diffusion of
innovative technology in the food industry sectors
took unpredictable and non-linear paths. A wholeof-market perspective, however, served to explain
the apparently confused and confusing patterns of
uptake which occurred during EXDOC’s phased
implementation. A holistic view of contributions
219
Evolutionary Diffusion Theory
to knowledge development and consideration
of the role government can play as policy
maker contributed to the conclusion that rapid
dissemination of innovative G2B technology in the
Australian food sector depends on compatibility
with existing commercial practice, cooperation
with local institutions and accommodation of
key stakeholders
cAse sTuDy TwO:
ImplemenTATIOn Of An
enTeRpRIse wIDe DOcumenT
AnD RecORDs mAnAgemenT
sysTem (eDRms)
In our second case study we move from a cross
sectoral document delivery application implemented at Australian federal government level,
to a review of an enterprisewide system selected
and implemented by a single public sector entity
at local government level in South Australia. An
Electronic Document and Records Management
Systems (EDRMS) implementation with fully
scanned records is a type of technology change
that affects all users in an organisation, taking
them from a manual, self-managed style of work
to an organisational systems approach with limited
personal customisation. The acceptance required
of users is several orders of magnitude greater
than in an update of an existing computer-based
system, which represents changing “like for
like” and hence does not truly test technology
acceptance.
Corporate governance requirements and growing pressures for legislative compliance have
stimulated national and global uptake of content
management systems that can mirror internal
corporate approval processes, store information
and retrieve it in an accurate and timely manner
(Government Exchange, 2004). These external
pressures, together with the availability of funding
and support for systems underpinning open and
accountable government, have resulted in local
220
government instrumentalities becoming fertile
ground for full EDRMS implementations.
The selection and Implementation
of an eDRms, city of charles sturt,
south Australia
Local government in Australia consists of 629
Councils and 100 community governments. In
South Australia, 68 councils employ around
10,000 people. One of the largest of these councils, covering approximately 10% of metropolitan
Adelaide, the state capital, is the City of Charles
Sturt (http://www.charlessturt.sa.gov.au/). The
City of Charles Sturt (CCS) employs just over 400
full time equivalent staff in 10 business portfolios
or departments. Council employs seven Records
staff members to register over 300,000 records
per annum.
In 2002, the process of replacing the Council’s
paper records system began with a series of staff
workshops designed to examine the issues for staff
in moving from paper to electronic documentation. They found that the paper system limited the
ability of staff to track and share documentation
and the speed at which information could travel
through the organisation. The increasing volume
and usage of email correspondence in recent times
added to these problems. The discussion from this
series of workshops culminated in the creation
of a Records Management Strategic Directions
document that clearly established the need for an
Electronic Document and Records Management
System (EDRMS).
The selection project began with establishing
and charging a project team with the responsibility
‘to replace the records section of the GEAC TCS
and other localized manual and indexed systems
used for the same purpose’. The initial document set out six objectives: replacement of the
system within a set budget limit, set time frame
for implementing the replacement system, work
process and efficiency gains, minimized need for
new or different hardware and operating systems
Evolutionary Diffusion Theory
within the organisation, minimized changeover
effects on running the business and ensuring the
new system was implemented consistently with
the Records Management Strategy. At this stage
the size and budget of the project still had to be
determined.
In October 2002 the request for tender was
advertised; the decision was based on knowledge
that the project was now very likely to move
forward. A full proposal for the EDRMS was
put together in November 2002. Final approval
was given by Council in June 2003. Following
tender evaluation and demonstrations, TRIM—a
product from Tower Software—emerged as the
CCS preferred product. By June 2003 the CEO
at Council had signed contracts for the purchase
of the product.
The EDRMS project team was keenly aware
that communication about the project needed planning – otherwise there was likely to be significant
resistance from staff when it came to training.
Clearly, electronic record-keeping would require
a changed understanding of records administration: staff members had to be able to relate the
changes to their individual work requirements.
Effective communication of these changes was
important to project outcomes. Integrating a review of records-keeping with other required topics
subsequently led to the vendor’s half day training
program being extended to a full day.
Around two thirds of the CCS workforce (280
staff members) were involved with the selected
EDRMS solution: TRIM. However, unless the
EDRMS was fully and formally integrated into
business processes, it was likely no one would
use the system. The implementation team consequently decided to institute an analysis of all
council business processes that touched records.
Business process analysis would serve to assist
the integration process and also help to identify
flagship projects for each area. These flagship
projects would champion the benefits of uptake and
so accelerate adoption and compliance. Undertaking analysis of well over 100 business processes
required the involvement of many additional staff
apart from the members of the three teams. The
process accelerated the integration of records
with each team’s activities, but also ensured that
significant communication about the project took
place not only within, but across teams.
Moving from paper to TRIM required a
number of technology decisions. Containers for
records had to be put in the EDRMS. Finding and
choosing the containers required considerable
fine-tuning. In August 2003, design of the container structures was undertaken by interviewing
all business units and ascertaining their needs.
Although time-consuming it was important to
consult with staff regarding the types of files being
used for classification purposes. This process led
to the creation of a draft list of containers and the
record types that they would hold. The list was
distributed to and revised in conjunction with
business units in an iterative process.
Workflows in the EDRMS constitute a benefit
above and beyond core records management. All
business units would use TRIM. The workflow
development process was designed to develop the
EDRMS as a system people wanted - not just one
they had to use. The aim was to achieve business
process gain so business and records would come
to be viewed as one process - not two. All new
technology and systems needed to be designed
around the EDRMS. Otherwise there was a real
risk that new information silos would form.
The documentation of business process workflows at CCS began by identifying each business
unit and process. An early iteration of this process
found some 191 business processes. Each process
was then ranked for its suitability for conversion
to electronic work flows on TRIM. Ranking was
undertaken according to the suitability of workflows to the TRIM system as well as by its priority
as assessed by the relevant business unit. Each of
the documented workflows was developed over
multiple iterations. Whilst feasibility of conversion
to TRIM was ultimately an executive decision,
there was considerable consultation with staff
221
Evolutionary Diffusion Theory
prior to the final decisions. As part of this exercise interest in specific processes for conversion
within individual business units was recorded
and ranked. The consultation process proved to
be a valuable resource as the rankings were of
great assistance to management when it came to
selecting ‘champions’ to promote uptake within
a number of CCS business units. Some 10 to 15
of the 191 processes identified as candidates for
conversion were then implemented at start-up.
A key challenge facing user acceptance testing
early in the development process is the difficulty
of conveying to users what a proposed system will
consist of in a realistic way. It is at this point that
people form general perceptions of a system’s
usefulness – perceptions that are strongly linked
to usage intentions. Hence, introductory training,
particularly its perceived relevance and integration with business processes, can have a strong
impact on usage intentions (Laeven T, 2005).
These intentions are significantly correlated with
future acceptance of the system (Davis, Bagozzi
and Warshaw, 1989 p.1000). It was vital to ensure
that staff understood the meaning of a Record
and related procedures once the EDRMS went
live. The decision was therefore made to extend
the mandatory training for all CCS computer
users from the proposed half day to a whole day
to cover Records Management procedures. Ten
staff members were trained each day over a 7.5
week period.
Continuing steady growth in registration of
Council records clearly indicates that processing
staff at all levels of the organisation appreciate
the benefits of the implementation for finding,
searching and viewing documents (Figure 2). Familiarity with the EDRMS at CCS increased staff
confidence in using electronic records systems and
other software – skills that are increasingly valued
in the job market. All records have now been on
the TRIM system since 8th March 2004. Staff
members understand the system and feel secure.
Low compliance remains a problem in certain
areas; for example decisions to ‘TRIM’ (add to
records) and the time required entering each item
are difficult to resolve in the case of email
The success of systems integration at CCS,
achieved for a project budget of $150,000, has
been a notable feature of the EDRMS implementation. Transactional integration at CCS has meant
that all systems integrate information. Although
not part of the original project, the Document
Assembly process within TRIM is now widely
used for internal forms and document templates.
Most importantly, the integration between TRIM,
Proclaim One and GIS means that all of these
systems can now be viewed at desk level.
The EDRMS replaced a paper-based system
that represented a mainstay for staff members’
day-to-day operations and one which had been
established over many years. For some employees
the fact that there is a new baseline for ‘business
Figure 2. Year on year registrations 2006 – 2007 (City of Charles Sturt, South Australia)
222
Evolutionary Diffusion Theory
Table 4. Application of EDT Axioms to the EDRMS case study
Evolutionary Diffusion Theory
Explanatory Contribution to Case Analysis
Attention to discontinuing ineffective practice & slowing uptake of inappropriate
technologies via initial staff workshops
Decision to articulate dynamic mechanisms underlying uptake of innovation by instituting Business process analysis & documentation of business process work flows
Rejection of Optimisation
Recognition that an organisational decision to adopt does not automatically translate
into individual or group acceptance without resistance results in use of monthly usage
reports by work sectors to stimulate uptake across organisation
Recognition that subjective norms can be shaped by influence. Hence EDRMS uptake
and usage are strongly supported and encouraged by CEO and senior management
Innovation and diffusion from a whole-ofmarket perspective
N/A
System wide perspective by management underpins successful integration of EDRMS
as indicated by:
Holistic View of Contributions to Knowledge
Development
Extra training for all staff to ensure understanding of meaning of a record and so
encourage user acceptance of the new system
Documentation of all business process workflows to encourage identification of business
and records as one process and so avoid information silos
Continual building on the base system to indicate that the EDRMS is a core technology
assisting staff become more productive
Plays key role in:
Recognition that subjective norms can be shaped by influences such as pressures for
legislative compliance and corporate governance requirements
Government as policy maker
Coordinating stakeholder institutions within innovation system by making funding
available
Seeking an acceptable business solution and intervening with context specific solutions
sensitive to local path dependencies to ensure diffusion across sectors by providing
support for systems underpinning open and accountable government
as normal’ continues to be a demanding concept.
For any organisation in the public or private sector, new systems integration ultimately means
finding ways to maintain the impetus for cultural
change (Gregory 2005; Laeven 2005; Maguire
2005; Kotter and Cohen 2002).
The successful outcome of EDRMS uptake in
this local government instrumentality has been
attributed to a number of factors including:
•
•
•
•
The vital role played by senior executive
support especially that of the CEO
The culture of open communication\supportive of staff involvement and risk-taking
throughout the project.
A clear understanding by key players of the
benefits of the project and their energy in
pursuing outcomes
The emphasis on consultation from the
earliest days of the project
223
Evolutionary Diffusion Theory
•
The pressure for uptake from fellow employees.
Since the initial introduction of TRIM as the
EDRMS at the City of Charles Sturt, development has continued in the areas of integration to
other systems, complementary smaller software
applications to address business issues, work-flow
of business records and many other advances associated with the technology. Continual building
on the base system clearly demonstrates to staff
within CCS that the EDRMS is a core technology
and assists them in being more productive. Closely
coupled to these on-going advances is sensitivity
to the fact that the organisation as a whole must
continue to perceive that there is benefit gained
for the work that is required to maintain a full
EDRMS.
In Table 4 we set out the four axioms we developed from EDT and apply them to this second
case study. The four axioms with their associated
elements (see Table 1) were applied to analysing
our findings from this EDRMS implemented by a
local council in South Australia. Table 6 indicates
how closely the four axioms could be applied to
our single organisation case study analysis.
Understanding the EDT axiom, Rejection of
Optimisation provided real insight into how a
technology implementation unfolds. It addresses
many of the factors that an organisation must deal
with when facing technological change such as
understanding how choices are made between
systems and even whether to choose a system at
all. Management appreciation of the importance
of the axiom A Holistic View of Contributions to
Knowledge underpinned the successful integration of the EDRMS in every aspect of Council
business. The new system was deliberately built
into every aspect of business undertaken at the
City of Charles Sturt, with productivity linked to
its success at every opportunity.
The EDT axiom, Government as policy maker
also proved highly applicable to the implementation. In many instances, Australian local
224
government has been a leader in implementing
EDRMS due in no small part to factors closely
associated with government acting in the role of
policy maker by exerting pressures for legislative
compliance, via corporate governance requirements and the availability of funding for uptake
of EDRMS. These three EDT axioms have provided excellent indicators to the project manager
and a clear explanation of why the organisation
chose a particular path. Only one of the four EDT
axioms - Innovation and diffusion from a wholeof-market perspective - was not applicable to our
second case study.
cOnclusIOn
IS implementation studies lack a widely accepted
theoretical framework to satisfactorily explain
how complex and networked technologies diffuse
in practice. The two detailed case studies described
in this chapter applied EDT to analyse how a technology implementation unfolds. We investigated
how choices were made between systems or even
whether to choose a system at all. EDT provided
reliable indicators and an explanation of why
the organisation chose a particular path. Three
of our EDT axioms Rejection of Optimisation, A
Holistic View of Contributions to Knowledge and
Government as Policy Maker offered particularly
useful tools for analysing key aspects of both
implementations (see Table 6). The only EDT
axiom which did not fully apply to our second
case study was Innovation and Diffusion from a
Whole-Of-Market Perspective.
EDT axioms thus proved useful for analyses
at both sectoral and organisational levels: they
were only marginally less applicable as an analytical tool for investigating a single enterprise
wide system implementation. When applied to
the AQIS implementation in our first case study,
the broader IS investigative approach of Evolutionary Diffusion Theory effectively explained
the impact(s) of diffusion in the more complex
Evolutionary Diffusion Theory
environment of a market, in a way the more
restricted alternative diffusion of innovation
theories could not match. Our second case study
featured uptake of an enterprise-wide EDRMS
solution by a local government entity. Here, EDT
provided real assistance in explaining how a
single – but complex – organisation can manage
a major change in its understanding of activities
and sources of information.
Our chapter, therefore, provides not only an
introduction to Evolutionary Diffusion Theory, but
also sets out how EDT can be effectively applied
to explain influences on uptake of innovation in
an Information Systems environment. EDT offers
an explanatory basis for many of the complex
and interrelated issues that apply to instances of
e-business uptake in ways which are not matched
by either Classical DoI theory, or by the networkbased extensions to that approach. We believe that
EDT and the axioms we derived from the theory,
together offer a richness of analytic approach and
a powerful explanatory capability rendering it
particularly appropriate to discussing the diffusion
of information technology. Since the majority of
real world e-business implementations are cross
sectoral, confirmation of the validity of EDT for
such applications presents exciting possibilities
for improved understanding of issues affecting
acceptance levels of new technologies in the IS
field.
RefeRences
Andersen, E. S. (1994). Evolutionary economics: Post-Schumpeterian contributions. London:
Pinter Publishers.
Arthur, W. B. (1994). Increasing returns and path
dependency in the economy. Ann Arbor, MI: The
University of Michigan Press.
Baskerville, R., & Pries-Heje, J. (2001). A multiple-theory analysis of a diffusion of information
technology case. Information Systems Journal,
11(3), 181-212.
Bijker, W. B. (1995). Of bicycles, bakelite and
bulbs: Toward a theory of sociotechnical change.
Cambridge, MA: MIT Press.
Boudreau, M-C., & Robey, D. (1999). Organizational transition to enterprise resource planning
systems: Theoretical choices for process research.
In Proceedingss. of the Twentieth International
Conference on Information Systems, Charlotte,
NC (pp. 291-299).
Clarke, R. (1999, June 7-9). Electronic services
delivery: From brochure-ware to entry points. In
Proceedings of the Twelfth International Electronic Commerce Conference, Bled, Slovenia
(pp. 8-9).
Clarke, R. (2001). Innovation diffusion resources.
Open site. Retrieved September 25, 2005, from
http://www.anu.edu.au/people/Roger.Clarke/
SOS/InnDiffISW.html
Constant, E. W. (1980). The origins of the turbojet
revolution. Baltimore: Johns Hopkins University
Press.
Damsgaard, J., & Lyytinen, K. (1996). Government Strategies to promote the diffusion of
electronic data interchange: What we know and
what we don’t know. Information Infrastructure
and Policy, 5(3), 169-190.
Davis, F. D., Bagozzi, R. P., & Warshaw, P. R.
(1989). User acceptance of computer technology:
A comparison of two theoretical models. Management Science, 35(8), 982-1003.
Dopfer, K. (Ed.). (2001). Evolutionary economics:
Program and scope. Dordrecht, The Netherlands:
Kluwer Academic Publishers.
Dosi, G., & Nelson, R. R. (1994). An introduction
to evolutionary theories in economics. Journal of
Evolutionary Economics, 4, 153-172.
Dosi, G., Freeman, C., Nelson, R. R., Silverberg,
G., & Soete, L. (Eds.). (1988). Technological
225
Evolutionary Diffusion Theory
change and economic theory. London: Pinter
Publishers.
Downs, G. W., & Mohr, L. B. (1976). Conceptual
issues in the study of innovations. Administrative
Science Quarterly, 21(Dec), 700-14.
Edquist, C. (Ed.). (1997). Systems of innovation:
Technologies, institutions and organisations.
London: Pinter Publishers.
Fichman, R. G. (1992, December 13-16). Information technology diffusion: A review of empirical
research. In Proceedings of the Thirteenth International Conference on Information Systems
(ICIS), Dallas, TX, USA (pp. 195-206).
Galliers, R. D., & Swan J. A. (1999). Information
systems and strategic change: A critical review of
business process re-engineering. In W. L. Currie
& R. D. Galliers (Eds.), Rethinking MIS (pp. 361387). Oxford, UK: Oxford University Press.
Gregory, K. (2005). Implementing an electronic
records management system: A public sector
case study. Records Management Journal, 15,(2),
80-85.
Holbrook, A., & Salazar, M. (2004). Regional
innovation systems within a federation: Do
national policies affect all regions equally? The
International Journal of Innovation Research,
Commercialization, Policy Analysis and Best
Practice, 6(1), 50-64.
Jones, N., & Myers, M. D. (2001). Assessing three
theories of information systems innovation: An
interpretive case study of a funds management
company. In Proceedings of the Pacific Asia
Conference on Information Systems, Seoul, Korea
(pp. 1005-1019).
Kowol, U., & Küppers, U. (2003). Innovation networks: A new approach to innovation dynamics. In
M. van Geenhuizen, D. V. Gibson, & M. V. Heitor
(Eds.), Regional development and conditions for
innovation in the network society. Lafayette, IN:
Purdue University Press.
226
Kotter, J. P., & Cohen, D. S. (2002). The heart of
change: Real-life stories of how people change
their organizations. Boston: Harvard Business
School Press.
Kwon, T. H., & Zmud, R. W. (1987). Unifying
the fragmented models of information systems
implementations. In J. R. Boland & R. Hirschheim
(Eds.), Critical issues in information systems research (pp. 227-252). New York: John Wiley.
Laeven, T. (2005). Competencies – the asset that
counts most: On developing human talents as a
prerequisite for successful EDRM changes. In C.
Hare & J. McLeod (Eds.), Managing electronic
records (pp. 129-148). UK: Facet Publishing.
Lambooy, J. G., & Boschma, R. A. (2001). Evolutionary economics and regional policy. Annals
of Regional Science, 35, 113-131.
Lissoni, F., & Metcalfe, J. S. (1994). Diffusion
of innovation ancient and modern: A review of
the main themes. In M. Dodgson & R. Rothwell
(Eds.), Handbook of industrial innovation (pp.
106-144). Cheltenham, UK: Edward Elgar Publishing Limited.
McKelvey, M. (1997). Using evolutionary theory to
define systems of innovation. In C. Edquist (Ed.),
Systems of innovation—technologies, institutions
and organizations (pp. 200-222). London: Pinter/
Cassell Publishers.
Maguire, R. (2005). Lessons learned from
implementing an electronic records management
system. Records Management Journal, 15(3),
150-157.
Markus, M. L., & Robey, D. (1988). Information
technology and organizational change: Causal
structure in theory and research. Management
Science, 34(5), 583-598.
Metcalfe, J. S. (1994). The economics of evolution
and the economics of technology policy. Economic
Journal, 104, 931-944.
Evolutionary Diffusion Theory
Mokyr, J. (1996). Evolution and technological
change: A new metaphor for economic history?
In R. Fox (Ed.), Technological change (pp. 63-83).
London: Harwood Publishers.
Saviotti, P. P., & Metcalfe, J. S. (1991). Evolutionary theories of economic and technological
change: Present status and future prospects.
Reading, UK: Harwood Academic Publishers.
Moore, G. C., & Benbasat, I. (1991). Development
of an instrument to measure the perceptions of
adopting an information technology innovation.
Information Systems Research, 2, 192-222.
Van de Ven, A. H., Polley, D. E., Garud, R., &
Venkatraman, S. (1999). The innovation journey.
New York: Oxford University Press.
Nelson, R. R. (1995). Recent evolutionary theorizing about economic change. Journal of Economic
Literature, 33, 48-90.
Nelson R. R. (2002). Bringing institutions into
evolutionary growth theory, Journal of Evolutionary Economics, 12(1/2), 17-29.
Nelson, R. R., & Winter, S. G. (1982). An evolutionary theory of economic change. Cambridge,
MA: Belknap Press.
Norgren, L., & Hauknes, J. (1999). Economic rationales of government involvement in innovation
and the supply of innovation-related service. Oslo,
Norway: Rise Project, European Commission.
Orlikowski, W. J., & Hofman, J. D. (1997). An
improvisational model for change management:
The case of groupware technologies. Sloan Management Review, 38(2), 11-21.
Reich, B. H., & Benbasat, I. (2000). Factors that
influence the social dimension of alignment
between business and information technology
objectives. MIS Quarterly, 24(1), 81-113.
Rogers, E. M. (1995) The diffusion of innovation
(4th ed.). New York: The Free Press.
Rosenberg, N. (1969). The direction of technological change: Inducement mechanisms and focusing devices. In N. Rosenberg (Ed.), Economic
development and cultural change (Vol. 18, No.
1, Part 1) (pp. 1-24). Chicago: The University of
Chicago Press.
Saviotti, P. P. (1996). Systems theory and technological change. Futures, Dec., 773-785.
Wilkins, L., Castleman, T., & Swatman P. M. C.
(2001). Organisational factors in the diffusion of
an industry standard. Electronic Markets, 11(4),
222-230.
Wilkins, L., Holt, D., Swatman, P. M. C., & Chan,
E. S. K. (2007, June 4-6). Implementing information management strategically: An Australian
EDRMS case study. In Proceedings of the 20th
Bled International E-Commerce Conference,
Bled, Slovenia.
Wilkins, L., Swatman, P. M. C., & Castleman, T.
(2003). Electronic markets and service delivery:
Requisite competencies for virtual environments.
In Proceedings of the 11th European Conference
on Information Systems ECIS 2003, Milan, Italy
(pp. 1-13).
Wilkins, L., Swatman, P. M. C., & Castleman, T.
(2002). Government sponsored virtual communities and the incentives for buy-in. International
Journal of Electronic Commerce, 7(1), 121-134.
Wilkins, L., Swatman, P. M. C., & Castleman, T.
(2000, June 19-21). Electronic commerce as innovation – a framework for interpretive analysis.
In Proceedings of the 13th Bled International
Electronic Commerce Conference, Bled, Slovenia
(pp. 150-121).
key TeRms AnD DefInITIOns
Axioms: An axiom is any starting assumption from which other statements are logically
derived. An axiom can be a sentence, a proposi-
227
Evolutionary Diffusion Theory
tion, a statement or a rule that forms the basis of
a formal system.
Corporate Governance: The system/process
by which the directors & officers of an organization are required to carry out & discharge their
legal, moral & regulatory accountabilities &
responsibilities.
Electronic Document and Records Management System: An EDRMS aims to enable
businesses to manage documents throughout the
life cycle of those documents, from creation to
destruction.
a process that moves through an initial phase of
generating variety in technology, to selecting
across that variety to produce patterns of change
resulting in feedback from the selection process,
to the development of further variation.
enDnOTes
1
Evolutionary Diffusion Theory: A classic
evolutionary model of technological change first
developed by Nelson and Winter (1982). The model
demonstrates the possibility of collating a wide
diversity of elements (including the processes of
transmission; variety creation and selection) and
integrates them into a process which can then be
applied to understanding the uptake of innovative
technology.
Market Focus: The development and diffusion of new variety or innovation in an economic
system.
2
Rejection of Optimisation: Evolutionary
Diffusion Theory rejects the assumptions that:
individual firms behave optimally at the micro
level and stress the gradualism of internal adoption
of innovative technology between firms and over
time. Successful innovations are rarely predictable and represent the outcome of multiple and
contingent variables and do not always have to
be the best ones.
Rogers’ Classical Diffusion of Innovation
Theory: Rogers focused attention on the shape
of the diffusion curve, describing innovation as
228
3
Path dependence is a key concept of Evolutionary Economics, usually explained
in association with an industrial process
e.g. the QWERTY keyboard. It describes
the natural trajectory of an unpredictable
original selection where, although many
outcomes are possible, under increasing
returns the process becomes focused on a
particular path (Arthur 1994; Nelson and
Winter 1982). Path dependence suggests
that it is often costly to change technologies
in production processes. When the costs of
change are large, it is possible that firms will
continue to use sub-optimal technologies.
A simple overview of the EXDOC system
can be accessed at the AFFA website link,
under AQIS publications (see http://www.
affa.gov.au/content/publications.cfm). This
description of the EXDOC system draws on
the following sources: an earlier case study
of AQIS and EXDOC as part of a major
research project on EDI systems integration (Swatman, 1993); the Minter Ellison
2002 Report; and information provided
by the EXDOC administrator Mr N Scott
2002/4.
Aspects of the EXDOC case study have
already been reported in the proceedings of
BLED 2000, ECIS 2000, ECIS 2003; and in
IJEC 2002 and EM 2001 journal articles