Documente Academic
Documente Profesional
Documente Cultură
2009
Vol. 21, No. 1
Established 1989
ISSN 0905-0167
Scandinavian Journal of Information Systems
Publishing
Copyright © 2009 Scandinavian Journal of Information Systems. The IRIS Association, Aal-
borg University, Department of Computer Science, Selma Lagerlöfs Vej 300, DK-9220 Aal-
borg, Denmark.
ISSN: 0905-0167
ISBN: 9788799142576
Production team: Peter Axel Nielsen (director) and Annette Moss (editing).
See www.e-sjis.org for scope, submission of articles, and subscription to the printed journal.
Editorial Board
• Bjørn Erik Munkvold, University of Agder, Norway, Coordinating Editor.
• Rikard Lindgren, IT University of Gothenburg & Viktoria Institute, Sweden, Editor.
• Keld Bødker, Roskilde University, Denmark, Editor.
• Samuli Pekkola, Tampere University of Technology, Finland Editor.
Advisory Board
• Richard Boland, Case Western Reserve University.
• Tone Bratteteig, Oslo University
• Susanne Bødker, Aarhus University.
• Bo Dahlbom, IT University of Göteborg.
• Jan Damsgaard, Copenhagen Business School.
• Jukka Heikkila, University of Jyväskylä.
• Ola Henfridsson, Viktoria Institute.
• Karl Kautz, Copenhagen Business School.
• Finn Kensing, University of Copenhagen.
• Karlheinz Kautz, Copenhagen Business School.
• Kalle Lyytinen, Case Western Reserve University.
• Lars Mathiassen, Georgia State University.
• Eric Monteiro, Norwegian University of Science and Technology.
• Joe Nandhakumar, Warwick Business School.
• Ojelanki Ngwenyama, Ryerson University.
• Peter Axel Nielsen, Aalborg University.
• Sandeep Purao, Penn State University.
• Matti Rossi, Helsinki School of Economics.
• Ulrike Schultze, Southern Methodist University.
• Jesper Simonsen, Roskilde University.
• Susan Leigh Star, Santa Clara University.
• Erik Stolterman, Indiana University.
• Carsten Sørensen, London School of Economics.
• Pål Sørgaard, Telenor Research & Innovation.
• Virpi Tuunainen, Helsinki School of Economics.
• Geoff Walsham, University of Cambridge.
• Youngjin Yoo, Temple University.
Editorial
Keld Bødker
Rikard Lindgren
Samuli Pekkola
The first issue of the SJIS volume 21 features one research article and a special issue on challenges
and opportunities of distributed participatory design. The research article by Taina Kaapu and
Tarja Tiainen presents a phenomenographical analysis of consumers’ interpretation of privacy in
e-commerce, identifying a number of different privacy conceptions. Their results illustrate how
the consumers’ view of privacy is situated and dynamically constructed based on new informa-
tion and experience. The guest editors of the special issue were Karin Danielsson Öberg, Dorina
Gumm and Amir M. Naghsh. We thank them for their contribution to the journal, and the
considerable efforts they have put into serving as guest editors.
With this issue, Bjørn Erik Munkvold has taken over from Matti Rossi as coordinating edi-
tor for SJIS. We are grateful to Matti for his valuable service for the journal and the community,
and welcome Samuli Pekkola as the new Finnish representative on the SJIS editorial board.
Taina Kaapu
Department of Computer Sciences, University of Tampere, Finland
taina.kaapu@uta.fi
Tarja Tiainen
Department of Computer Sciences, University of Tampere, Finland
tarja@cs.uta.fi
Abstract. Information privacy protection and invasion of privacy in e-commerce have be-
come important topics in both everyday activities and scientific discussions. The aim of
this study is to understand how consumers regard privacy in business-to-consumer e-com-
merce. As this study focuses on consumers’ own interpretations of privacy, the research
approach is empirical, rather than theoretical. Based on a phenomenographical analysis
of consumer interviews, we identify different layers of understanding by focusing on the
referential objects and the structural components of information privacy. The result in-
cludes 25 different privacy conceptions, showing that consumers’ view of privacy is situated
and constantly under construction as the consumer gets new information or experiences.
1 Introduction
Understanding consumer behaviour is vitally important in online shopping. One precondition
for the growth of e-commerce is that the consumers use online channels. These channels are
chosen at each stage in the purchase process: requirements determination, vendor selection,
purchase, and after-sales service (Choughury and Karahanna 2008). The process is connected
to the consumers’ trust in e-purchase and e-vendors, and to the consumers’ perceptions of risk
(Choughury and Karahanna 2008; Verhagen et al. 2006). The number of online consumers has
grown; at the same time, the fears regarding information privacy have also increased (Malhotra
et al. 2004). The biggest concerns to Internet users are viruses, spam, spyware and hackers (Paine
et al. 2007). If these problems are not solved, the consumers whose privacy concerns have not
3 Methodology
Our aim is to understand how consumers see privacy and to describe the differences in their
views in e-commerce. For studying people’s own interpretations of a concept (information pri-
vacy in this case) a qualitative method which focuses on people’s narration is needed. Indi-
viduals’ views are socially constructed; however, the individuals’ own background, including
their education and experience, can also have effect on their views. Methods such as discourse
analysis, grounded theory (GT), and phenomenography can thus be considered. In choosing the
research method, we discarded discourse analysis, since it has its focus on social interaction, such
as shared (communal) views and argumentations (Alvesson and Karreman 2000).
A decision between GT and phenomenography was made based on their different study
aims. Although they both itemize individuals’ talk to its elements, the target of that process
Otherwise, the interviewing progress was alike in both of the interview sets. The individual in-
terviews were open-ended, and only the topics were decided beforehand. The interviews started
from a general discussion about the interviewees’ backgrounds (as consumers in e-commerce)
and were followed by a discussion about the main issue (which concerned, in the first set,
privacy issues in e-commerce and, in the second set, e-services in the form of e-journals and e-
commerce). The duration of the interview situations varied from thirty minutes up to two hours
and thirty minutes. The interviewer’s role was to follow the interviewees’ ideas and explore their
narration.
Object A: Use and misuse of customer information. In the interviews, the consumers mostly
discuss their personal information as customer information. This is information which e-ven-
dors collect by asking it from consumers. Some of the interviewees use the concept “customer
information”, others underline their own viewpoint by talking about “my information that my
e-vendor has or knows”. In general, the interviewees are reluctant to give their information
and they are afraid of misuse of their personal information – but only if they do not know the
e-vendor beforehand. Nevertheless, the consumers interviewed also understand the benefits of
getting personalized offers from e-vendors. One problem is that giving information is compli-
cated (Conception 2A):
Object B: Monitoring consumers. The interviewees describe two kinds of monitoring. They
tell that some one may watch when they are using the computer. This is the same as traditional
monitoring: the one who is monitored and the one who monitors are in the same physical space.
Besides of the traditional monitoring the interviewees are concerned if there exists virtual moni-
toring which happens over the Internet. For example, the interviewees state that some e-vendors
keep an eye on consumer purchase behaviour or that the police monitors the Internet. Virtual
monitoring means that the visits to websites are monitored secretly and information about In-
ternet use is added to visitors’ personal information. The interviewees claim that it should be a
fundamental right (of Internet privacy) to visit web pages anonymously.
Some of the interviewees are more afraid of monitoring than others. The most careless ones
say that monitoring occurs only on a small scale or that their personal information is not rel-
evant to strangers. Some interviewees admit that they do not know enough – and they also do
not like giving their personal information, because they do not actually know what happens to
the information or what is possible to do over the Internet. One of the interviewees gives an
example where her own level of knowledge about technology affects her concerns with consumer
monitoring (Conception 2B):
I’ve heard about user tracing. When I hadn’t used the net for a very long time I got a
notification saying ‘you are running out of ink’. Oh my god, I thought, did I run out of
ink so fast? Then I realized that it was an advertisement. I’m still wondering if they could
know whether I was running out of ink.
Object C: Threat of spam. All the interviewees agree that unwanted e-mail is annoying and
most of them see a conflict between spam and privacy. However, they clearly differentiate be-
tween spam and other uses of customer information (object: use and misuse of customer infor-
mation); especially when the unsolicited mail or directed ads originate from their own e-vendor,
it seems acceptable. We use the word “spam’ because the interviewees cite the term; in addition,
“garbage” and “junk” were also used. Spam usually refers to unwanted e-mails, which causes
harm at least by filling, and occasionally choking (malicious attacks), one’s mailbox. Still, it is
difficult to determine whether an e-mail is wanted or not; for defining that, the recipients’ inter-
pretation is needed. Besides of defining what spam is, protection against it, and the consumers’
own actions are important, as seen in the next quotation (Conception 3C):
Of course, one factor is that if lots of spam start coming. -- But I haven’t done any busi-
ness with unfamiliar vendors and I have indeed avoided foreign firms - any contacts to
them.
Object D: Danger of hackers and viruses. The concept of hacker is used here because the
interviewees use it. They usually add that hackers use viruses to get their personal information
Object E: Risk with payment. This object refers to how issues related to making payment on
the Internet entail privacy problems: the interviewees are concerned that credit card information
can be stolen or used somewhere without their permission. Only a few of the interviewees report
having used their credit cards on the web. Although most of them frequently use Internet bank-
ing, they do not talk about it in a context of risk with payments; the bank is seen as an institu-
tion that can be trusted it seems. E-shopping could be paid in two main ways according to the
interviewees: with a credit card or in a post office when the consumer receives the product. The
interviewees said that now it is also possible to pay e-shopping directly via an Internet bank.
The interview quoted below shows that even though some of the consumers interviewed
may feel afraid using their credit cards, they have nevertheless used them because of the benefits
perceived (Conception 1E):
When you use some trustworthy and large vendors you get (the ordered product) in a
week or so… Then you will get the right product at the right time, and the bill comes at
the same time (with the product). When you pay with a Visa (credit card) it is chancy. I
wouldn’t otherwise use it (Visa credit card), but you have to use it (for paying) abroad.
The products are cheaper there.
Component 1: Product and e-vendor. When privacy is understood as related to the compo-
nent of the product and the e-vendor, all interviewees express very similar opinions. They say
that it is easier to order and give information if you know exactly what you are getting and who
the e-vendor is. First, the interviewees underline that they trust well-known e-vendors. They
mentioned many familiar Finnish brands – such as Veikkaus (a betting company, also working
online) and NetAnttila (an e-shop for clothes and household goods) – as places whereto it is safe
Component 3: Societal norms. The interviewees describe that differences in national laws and
behavioural norms affect on their actions in e-commerce. The third component is called Societal
norms since it concentrates on others’ expected behaviour based on images of commerce habits
and technological expertise in certain countries. The interviewees have very strong views on
security in different countries and the opinions are similar among all interviewees. They note
that ordering from abroad is not safe; in other words, they would not give their personal infor-
mation abroad. The interviewees emphasize their perception that information security is better
in Finland, and that it is easier to give personal information to Finnish vendors than to others
(Conception 3D):
I have been interested in high quality artistic tools. And there (abroad) are plenty of
them (artistic tools) that are not available here (in Finland). Sometimes I have tried to
order them. However, there is still this problem of privacy... Some hacker may steal my
information and I don’t receive my packet at all.
Component 4: Consumer him/herself. The interviewees can also understand the implications
of their own actions regarding privacy in e-commerce. The interviewees describe problems and
how their own behaviour can provide solutions. The problems mentioned in this respect are
Component 5: Fellow men. Besides of the interviewee’s own actions there are other people
whose actions affect the interviewee’s privacy situation. Focusing on the whole interviewing situ-
ation the interviewees also described what problems other people could cause to them. In the
category of Fellow men, the people are known by name and they belong to the same community
as the interviewee. They can be family members, work colleagues, or friends – also from a virtual
community. These known persons can help in the use of e-commerce, but they can also cause
harm, as in the next interview quoted, in which the interviewee discusses monitoring which
injures privacy (Conception 5B):
When I am using the computer, there can be so-called distractions. I find them disturb-
ing. (-) Sometimes my husband is trying to peek at what I am doing.
5 Discussion
Our study dealt with how consumers view information privacy. The result is a categorization,
which reveals consumers’ anxiety in the referential objects of privacy (use and misuse of cus-
tomer information, monitoring consumers, threat of spam, danger of hackers and viruses, risk
with payment). When we discussed privacy with the interviewees, they pointed our several risks
and threats. In the categorization, only one object, use and misuse of customer information,
contains both positive and negative sides of privacy; all other objects of privacy are regarded as
rather negative. Besides of the objects, we also identified the structural components of thought
about privacy. These were: product and e-vendor, technology, societal norms, consumer him/
herself, and fellow men.
As we described earlier, there are contradictory views about consumers’ fears of information
privacy. While several studies (Cassidy and Chae 2006; Chung 2003; Liu et al. 2005) state
that consumers are worried of their personal information use in e-commerce, Cheung and Lee
(2006) state that consumers are not afraid of invasions of privacy. Our study explains one reason
for the contradictory findings. Cheung and Lee studied the fears of privacy concerns in the con-
6 Conclusion
The aim of this study is to describe the differences in how consumers view privacy in e-com-
merce. In privacy issues we need more understanding from a consumer viewpoint (Cassidy and
Chae 2006; Dinev et al. 2006; Malhotra et al. 2004). The result of our study is a categorization
of consumers’ views on information privacy. It includes different layers of understanding by
focusing on referential objects and structural components of information privacy.
7 Acknowledgment
The authors would like to gratefully acknowledge the helpful comments on earlier versions of
this paper from the Associate Editor, Professor Bjørn Erik Munkvold, the three anonymous
reviewers, Dr Hannakaisa Isomäki, Professor Pertti Järvinen, Dr Minna-Kristiina Paakki, Dr
Tero Vartiainen, and the members of Dr Viveca Asproth’s working group in IRIS28 - 2005,
Kristiansand, Norway. We thank Steve Legrand for making our English more readable.
8 References
Alexandersson, M., Metod och medvetande, (Method and Awareness, in Swedish), Acta Universita-
tis Gothoburgensis, Göteborg, 1994.
Alvesson M., and Karreman, D., “Varieties of discourse: On the study of organizations through
discourse analysis”, Human Relation, (53:9), 2000, pp. 1125-1149.
Awad, N. F. and Krishnan, M. S., “The Personalization Privacy Paradox: An Empirical Evalua-
tion of Information Transparency and the Willingness to be Profiled Online for Personaliza-
tion”, MIS Quarterly, (30:1), 2006, pp. 13-28.
Bellman, S., Johnson, E. J., Kobrin, S. J., and Lohse, G. L., “International differences in infor-
mation privacy concerns: a global survey of consumers”, The Information Society, (20), 2004,
pp. 313-324.
Bergkamp, L., “EU data protection policy. The privacy fallacy: adverse effects of Europe’s data
protection policy in an information-driven economy”, Computer Law & Security Report,
(18:1), 2002, pp. 31-47.
Booth, S. A., Learning to program: A phenomenographic perspective, Acta Universitatis Gothobur-
gensis, Göteborg, 1992.
Bygrave, L. A. “Data Protection Pursuant to the Right to Privacy in Human Rights Treaties”,
International Journal of Law and Information Technology, (6), 1998, pp. 247-284
Camp, L. J., and Osorio, C., Privacy Enhancing Technologies for Internet Commerce, Trust in the
Network Economy, Springer-Verlag, Berlin, Germany, 2003.
Cassidy, C. M., and Chae, B., “Consumer Information Use and Misuse in Electronic Business:
An Alternative to Privacy Regulation”, Information Systems Management, Summer 2006,
pp. 75-87.
Dorina Gumm
effective WEBWORK GmbH, Germany
gumm@effective-webwork.de
Amir M Naghsh
Sheffield Hallam University, UK
a.naghsh@shu.ac.uk
Participatory design (PD) has been one of the important contributions of Scandinavian re-
searchers to information systems (Floyd et al. 1989). For a considerable amount of time, the
importance and usefulness of the approach has been illustrated during the design of computer
systems for work. The Scandinavian Journal of Information Systems (SJIS) serves as an impor-
tant contribution to the history and development of the approach. Already in its first volume
published in 1989, PD research was represented. The PD approach focuses on the collaboration
between designers and users, supporting direct involvement of users in the design process itself
while allowing designers to participate in the users’ world (cf., Kyng 1998).
Globalisation and organisational networks—and the implicit distribution coming along
with this—increasingly determine software development (Audy et al. 2004). The distribution of
collaboration in design and development processes has become a key challenge for PD, espe-
cially when one considers participation that was originally developed with a focus on co-located
design activities. This view of activity was developed to allow users and designers to collaborate
face-to-face, building on their mutual learning through the design process. However, the in-
volvement of a large number of users and other stakeholders who may be distributed across time,
space and organisational structures, has challenged developed PD research and activities.
Distributed design teams are by no means a new aspect in participatory design. A variety
of research work reports on projects that face some kind of distributedness (e.g., Divitini et al.
2000; Irestig et al. 2002, Bødker et al. 2004). Moreover, support of such distributed projects,
1 References
Audy, J., Evaristo, R. & Watson-Manheim, M. B., “Distributed analysis: The last frontier?”, in
HICSS ’04, IEEE Computer Society, 2004, p. 10010.2.
Barcellini, F., Détienne, F., Burkhardt, J.M. “Users and developers mediation in an Open Source
Software Community: boundary spanning through cross participation in online discus-
sions,” Int. J. Human-Computer Studies, 66, 2008 pp. 558–570.
Bannon, L. CSCW: An initial exploration, in Scandinavian Journal of Information Systems, 1993,
Vol. 5, pp. 3-24.
Bødker, K., Pors, J. K. & Simonsen, J., ‘Implementation of web-based information systems in
distributed organizations’, in Scandinavian Journal of Information Systems, 2004, Vol 16,
85–116.
Knut Staring
Department of Informatics, University of Oslo, Norway
knutst@ifi.uio.no
Jørn Braa
Department of Informatics, University of Oslo, Norway
jbraa@ifi.uio.no
Abstract. Through the study of a long term, globally targeted effort to design health in-
formation systems in the Global South, we explore challenges to distributed participation
within and across countries, and describe efforts at addressing these. Networked action
research projects can enable pooling of resources, skills, best practices and tools, and cross-
country collaboration does not have to preclude local ownership, as illustrated by the case
material in this article. We highlight specifically the need for circulation of people, artefacts,
and standards, to both support local practices and foster the capacity of all stakeholders to
take active part in the design and implementation of information systems. The deep effects
of global technological change call for a multilevel approach bridging local implementations
with global research and participatory design efforts and co-evolution of standardised tools.
3 Method
Our empirical investigations are grounded in the Scandinavian action research tradition, which
emphasises engagement in the field. A long-term distributed action research effort, the approach
followed in HISP builds on the literature on networked action research (Elden and Chisholm
1993; Engelstad and Gustavsen 1993), and has elsewhere been termed Networks of Action (Braa
et al. 2004a). In essence, it is based on the premise that action should not take place in isolated
units, but only as part of a greater community where sharing of experiences, learning processes,
and support are facilitated. The Networks of Action approach is specifically designed for the
resource limited conditions in the Global South. The 15 years time-span of HISP exceeds tra-
ditional ‘projects’ and is more akin to social movements or programmes (Elden and Chisholm
1993).
All three authors are actively engaged in the Health Information Systems Programme on a
daily basis. One of the authors was the originator of the project in the mid-1990’s, and the two
other authors have been engaged in the action research project for about five years, working on
The empirical focus of this paper is on the trajectory of participatory design in the project,
from its beginnings following a traditional PD approach in a few districts in South Africa, to an
increasingly globally dispersed processes of design and implementation taking place around the
network today. Taking Gumm’s (2006b) four axes of distribution (spatial, temporal, organisa-
tional, and stakeholder related) as an organizing principle, we identified a number of important
distributed interactions impacting on the design, both of the software and of the overall proc-
ess, including implementation. These interactions were then grouped according to similarities
in how they were affected by distribution. Surveying these groups, the following main themes
emerged:
• Design taking place at multiple levels, and the interaction between local and global
design processes
Technological transition Two branches (v1 & v2), infrastructure for sharing, but
2004-2007
fragmented processes, isolated modifications
Integration Multiple local teams, travelling, local developments
2008->
contributing to global software
Sub-sections 4.3 and 4.4 describe in more detail how the organisation of both people and tech-
nologies was decisive in the design of DHIS during the stages depicted above.
A shared toolbox
We use the term global toolbox for systems design to refer to what is being distributed and ap-
plied in the various in-country design processes in the global HISP network. The toolbox con-
tains software (the DHIS), best practices and recommended strategies for overall development
of in-country systems, as well as training material and other documentation.
DHIS is a generic tool rather than a pre-configured database application, with an open
metadata model and a flexible user interface that allow the user to design the contents of a spe-
cific information system without the need for programming. This also allows for rapid prototyp-
ing: for example, during a short initial visit in Sierra Leone, DHIS v2 was quickly customised
with the local organisational hierarchy, and data was imported from various fragmented systems.
The implementation team could then present the rapidly assembled system to major stakehold-
ers in the project, visualizing their own data in indicators and charts. By displaying concretely
what the final integrated system could provide, both technically and in terms of information, the
DHIS served as a powerful interactive tool supporting early discussions.
In a typical implementation sequence, the initial system designed at the national level un-
dergoes a validation process at selected pilot sites, where local managers and health workers play
a central role in modifying and further adapting the system to their needs. Being able to com-
municate through a visual and easily editable tool facilitates local PD processes. For example,
Hi, we have some problem in calculated Data element. I tried both in DHIS 1.4.0.37
and 40 but both didnt work. I tried to add some calculated data element and then tried
to export to data mart but the field show blank. And also there is no DataelementCalcalt-
edfield table in datamart file. Acutally, i was going through the code but all the steps for
export to data mart are about aggreation and indicators and others but not the anything
about caluclated data element. I dont know whether i am doing any mistake in setting
up the dataelement.
The Zanzibari implementers were recruited from local technical universities and had no prior
experience with either the DHIS or the health information field. They were immersed in an in-
tensive learning by trying process, working under the guidance of the visiting implementers on
designing data sets and indicators and customising the slowly maturing software. The local team
gradually increased their overall command of the domain and the maintenance of the system,
and gained the capacity to conduct extensive training workshops and facilitate on-the-job learn-
ing by doing in district offices. Fluent in the local language and culture, they could more easily
understand the feedback from local users and engage in often heated design discussions with
key stakeholders. Such intensive engagement with users often led to queries or suggested im-
provements which the implementers then raised with the international team, and subsequently
discussed on the global mailing list or directly with the core developers, in effect forming a
multi-level chain of mediation. However, as they became more familiar with the core team, the
local implementers also communicated directly via email or instant messaging.
5 Discussion
The stages in the HISP case—Pilot and national system, Expansion, Technological transition,
and Integration—illustrate several aspects of the relationship between DSD and DPD. We now
proceed to analyze the distribution of design activities through the interplay between local and
global levels, and the crucial role of implementers and networks.
Global standards
Maintaining the focus on PD as the network expanded necessitated distributed teams of de-
velopers engaging in locally relevant design. In time, and through training and the circulation
of people and artefacts, these teams took part in the design of generic modules which had de-
veloped from their own or similar contexts, thereby forming a DSD network. While the DSD
literature mainly discusses virtual teams and offshore outsourcing for the commercial sector in
the US or Europe, the HISP case turns the setting on its head and looks at “outsourcing” of large
The platform-like character of the software means that it is simultaneously part of an end
product which includes reports and local data standards, as well as scaffolding used by boundary
spanners in collaboration with local users to make systems function in their work settings. We
have seen from the initial phase in Sierra Leone how the flexibility of the platform enabled rapid
production of relatively elaborate boundary objects that facilitated communication between
users and developers.
The HISP variety of DPD is one where local implementers fill crucial roles as intermediaries
with end users. While the project has seen instances of enthusiastic health managers provid-
ing excellent feedback to core developers, such direct communication with end users are very
much the exception. The South African user mailing list shows some promise in this regard,
but strong “champions” and other boundary spanners are needed in each country to enable
anything similar to strong FOSS communities (Barcellini et al. 2008). The long term global
support to capacity building for local design has been crucial to providing sustainability. In most
of the HISP countries implementers were enrolled in graduate programs in health information
systems. However, emphasis was placed on maintaining strong links with field activities, and
all students were encouraged to link their studies to ongoing design projects, as illustrated i.e.
through the long term involvement of Tanzanian graduates in the Zanzibar project.
6 Conclusion
The contributions of this paper consist of adding perspectives to the study of distributed partici-
patory design, by 1) extending it to the Global South and 2) emphasizing the role of implement-
ers as boundary spanning mediators, 3) enabling active generification of innovations that arise in
local settings and 4) collaboration on the DPD of a common toolbox which in turn 5) affords
learning-by-trying through scaffolding.
In the perspective of the PD literature, parallels emerge between the early Scandinavian
projects and the current challenges facing information systems in the Global South. The HISP
case shares many features with for example the influential UTOPIA project (Bødker et al. 1987),
in terms of both practical methods and political aims. The approaches are similar in their focus
on empowerment through participatory capacity building and on the generation of useful tools
for and by people at the margins in global development, technological and otherwise. However,
while the UTOPIA project stranded as commercial technologies improved, the timing is bet-
ter for HISP. Today, FOSS technologies are at the cutting edge of technological development,
and huge corporations are embracing them. For example, IBM is using FOSS for competitive
advantage (Eclipse Foundation 2007). These practices are potentially of immense importance
in the Global South (Weerawarana and Weeratunga 2004). However, as discussed in this paper,
much scaffolding and learning by trying is needed before developing countries will be able to
master them “as their own language”.
Still, as we have shown, the HISP approach to institutional scaffolding of DPD contains
elements of a model that can plausibly be extended beyond its origins in the public health care
sector. Moreover, the major contribution of the HISP case lies in the demonstration that the
learning engendered by use of the system in the Global South can make considerable contribu-
tions to generic software design. In other words, emphasis should be put on facilitating two-way
learning and innofusion processes. Despite the challenges, the design of global software poten-
tially stands to benefit from wide distribution of use.
7 References
AbouZhar, C., and Boerma, T., “Health information systems: the foundations of public
health,” Bulletin of the World Health Organisation, (83:8), 2005, pp. 578-583.
Hartmut Obendorf
C1 WPS, Germany
hartmut.obendorf@c1-wps.de
Monique Janneck
University of Hamburg, Germany
monique.janneck@uni-hamburg.de
Matthias Finck
effective webwork, Germany
matthias.finck@effective-webwork.de
Abstract. Most studies of participatory design examine the development of a single, cus-
tomized software system that supports typical workflows within a single client organization.
To cope with other use contexts and new forms of work – such as communities and virtual
networks – the traditional repertoire of PD methodology needs to be expanded to deal with
distribution and diversification of users. Based on a ten-year case study, we describe experi-
ences with PD in the development of a groupware system that initially targeted a single use
context, but was continually extended and adapted to new contexts of use with new require-
ments and work practices. To enable distributed participation across contexts, new methods
had to be established: inter-contextual user workshops bring users from different contexts
and developers together to reflect on the usage and design of the software and its further de-
velopment. Commented case studies make this face-to-face interaction persistent, providing
a written documentation of distributed use experiences and design decisions. In the process
of building an inter-contextual community of users, the PD focus shifts from custom soft-
ware development to empowering users in assessing their own practice and technology use.
Table 1: Number of users in different contexts using the CommSy installation hosted by the
primary application service provider, effective webwork GmbH (April 2009).
In the spring of 2001, some of the CommSy developers joined a research project dealing
with computer supported cooperative learning (CSCL) and participatory software development
(cf., Pape et al. 2002). As a consequence, the CommSy development was no longer a volun-
tary, unpaid activity, but grew into a professional software development project, following the
STEPS (Software Technology for Evolutionary and Participative System Development) software
engineering framework (cf., Floyd et al. 1989). STEPS is a methodological framework for evo-
lutionary development in the PD tradition, stressing the importance of close cooperation with
users as well as an extensive use of prototyping to facilitate the embedding of software systems
in meaningful work processes in the user organization.
As a result of the three-year research project, CommSy had evolved into a sophisticated
groupware tool that had been extensively field-tested and refined in the context of university
education. As the CSCL research project ended in 2003, development of CommSy continued
in form of an Open Source project. However, hosting and support services could no longer be
offered free of charge for the universities using CommSy, as this had been the case while develop-
ment was funded by the research project. Instead, a spin-off company began to provide applica-
tion services (ASP) for anyone wishing to use CommSy without hosting their own webserver.
This new commercial perspective led to an exploration of new contexts of use.
One use context that seemed promising commercially were secondary schools. During its
development, CommSy had been used to a small extent in secondary education by schoolteach-
ers who had been introduced to CommSy during their studies at the University of Hamburg.
Table 2: Comparison of Communities of Practice and Interest (cf., Fischer 2001, p. 9).
To this end, two new techniques were introduced in the design process: inter-contextual user
workshops and commented case studies. Inter-contextual user workshops bring users from dif-
ferent contexts and developers together to reflect on the usage and design of the software and its
future development. Commented case studies make this type of face-to-face interaction persist-
ent – providing a written documentation of use experiences and design decisions.
Participants
Usually, about ten to fifteen users and three to five developers or members of current research
teams attend the one-day workshops. Over the past years, more than 50 users took part in the
workshops.
All moderators of CommSy workspaces—typically teachers or leaders of project groups –
receive individual invitations. Workshops are also announced publicly, e.g., on the CommSy
website, to reach a wider audience.
Participants are typically ‘heavy users’, representing a certain institution or group, who will
communicate the workshop results back to their colleagues. Their motivation is to partake in the
future development of CommSy and to benefit from the experience of others’ ‘lessons learned’.
Typically, the workshops start with an extensive round of introductions to help participants
to get to know each other and their respective contexts of use. Then, developers report about
upcoming developments. This is usually the starting point for an extensive discussion of use
experiences, problems, and requirements. The moderators visualize the contributions for clus-
tering later on. Sometimes the participants wish to work on different issues emerging from the
discussion in smaller groups. Topics often address use problems or phenomena that are shared
by many participants, e.g., how to increase active participation in online communication. A
feedback round concludes the workshop.
Since these workshops are not primarily research activities and need to provide an informal
and productive working atmosphere, they are not audio- or videotaped for detailed analysis. The
workshops are evaluated using the records and documentation (posters, etc.). A lot of evaluation
and documentation is actually carried out during the workshops, as participants, e.g., cluster
Example
In the following section, a concrete example of an inter-contextual user workshop is described.
The workshop took place from 10.30 A.M. to 5 P.M. at a secondary school. It was organized
by two schoolteachers who had introduced CommSy to their school and provided support for
their colleagues regarding CommSy use. They had already attended earlier workshops and had
offered to host the workshop when a location was needed.
Ten users and three members of the development team attended the workshop. Five us-
ers were also secondary school teachers, three were university lecturers and two were from an
Employers’ Liability Insurance Association which used CommSy as an intranet. Two of the
participants hosted their own CommSy servers while the others made use of the application
services provided by the CommSy development team. Figure 3 reproduces an excerpt from the
workshop record.
Reflection
So far, the development team received thoroughly positive feedback by the workshop partici-
pants regarding the benefits and usefulness of inter-contextual exchange. Being confronted with
perspectives originating in different contexts makes users reflect their own usage; this is pos-
sible from an angle not present within their own community of practice. Challenged by other
participants to explain why certain features are important to them, and contrasting this with
experiences from other backgrounds, they start to think through and sometimes question their
use routines. An example: The demand for highly differentiated access rights, which was voiced
by several users, turned out to be grounded in the transfer of practices participants knew from
other software products. In the workshop, this boiled down to a discussion of trust, hierarchies,
and authority and the value of equality that is inherent in the design of CommSy.
Nevertheless, establishing a dialogue between users from different ‘cultures’ (in the sense of
different working contexts, disciplines, values), speaking different ‘languages’ (conveying very
different experiences and beliefs) is not always easy and requires time. For example, at one oc-
casion entrepreneurs, teachers and university lecturers all spoke of ‘projects’—while addressing
completely different settings. A lengthy argument arose on the correct usage of the term, which
in the end had to be settled by the moderators.
Other conflicts arise when more ‘powerful’ user groups (who are able and willing to pay
for specific developments, for example) threaten to dominate the development of the software.
Inter-contextual user workshops help to set these conflicts at rest, as a compromise is worked out
in direct interaction with other communities of practice. However, sometimes the compromise
does not work out in practice and the problem surfaces again.
Furthermore, developers use the workshops to validate their own design decisions and ideas:
If features are approved by users from different contexts and with different backgrounds, they
are more likely to be helpful for the user community as a whole.
Setting
Commented case studies require some time for completion as they loosely follow the publishing
process of a scientific journal.
Invitations for contributions were made by e-mail and included a description of the current
CommSy version that the reports would refer to. Also a common format and structure was
proposed to make it easier to both comprehend and compare the user reports. As a side effect,
the development team used the thematic structure to moderate the writing process and guide
attention to topics they needed information on, much as in the workshop format. In the first
edition, the structure included section titles such as initial configuration, introduction into use
context, use conventions and so forth. This structure was not directly imposed on the authors,
but the template document that was sent out included the section subtitles. As a result, the
structure was accepted and used by almost all authors (some omitted sections they felt were not
relevant for their report).
Contributed reports were subjected to a review cycle that was mainly focused on under-
standibility and spelling, while care was taken not to influence the content of the article. These
reports formed the main body of the resulting case study volume. To preserve the value of the
contributions over time, a software description was prepended to the report collection.
The development team then grouped the different reports and created several classifications
of the practices and problems described in the case studies. An excerpt of the classifications used
in the case study are listed in Table 3: Apart from descriptive and quantitative data character-
izing the individual use case, the reports were analyzed regarding similarities with other use
cases and typical activities such as configuration of the workspace, documentation and support
given to users, etc. This offered pointers to similarities and differences, providing the basis for
developing a common language and the negotiation of requirements that also takes place in the
workshops.
Finally, the last section of a volume consists of comments by the development team, discuss-
ing the different use experiences reported in the case studies and the feature requests resulting
from them, and explaining software features and design decisions on the background of these
use experiences. Also, an outlook on the upcoming version of CommSy was added to the vol-
ume to indicate how the development would react to the requirements voiced by the different
user communities.
Example
In the following section, one edition of commented case studies is described in detail.
In the 2006 edition, contributions were made by users from VIRKON, a research project
on virtual networks, Mikropolis, an information management research group, Branta, JUMP
and Consulting Netzwerk, three networks of independent consultants, HR-Verband, a human
resources association and C1 WPS, an IT-consulting firm. All these organizations use CommSy
to support their internal communication and documentation needs. Furthermore, MQ21, a
European management development network, reported on their use of CommSy for organizing
a conference, and a member of the Branta network described his use of CommSy as a commu-
nication platform for a single customer-related project. Finally, the eCampus report detailed how
CommSy was introduced and supported in a University department.
Three different categories were created to group the case studies: preparation of use, appro-
priation of use, and facilitation of use. Furthermore, the case studies were categorized according
to the functionalities of the software that were used and described (Figure 4). Based on these
classifications, the case studies were indexed to enable readers to find case descriptions that
match their use context and practices.
The case descriptions are about three to four pages in length. In this volume, they are typi-
cally structured along the following paragraphs: Description of use context (name of group/
organization using CommSy, number of members, reason for using CommSy), types of us-
age (objectives and purpose of use, frequency and participation etc.), moderation of use (e.g.,
netiquettes and rules), and evaluation and conclusion (e.g., problems, lessons learned, feature
requests etc.). Figure 5 shows an excerpt of an actual case description.
The case descriptions are followed by an extensive comment (about 10 pages) by the devel-
opers, reflecting their past design decisions, answering questions and commenting on feature
requests raised in the case descriptions. For example, they describe how the CommSy design
principles were slightly altered to meet the requirements of heterogenous use contexts better,
and if and how (or why not, respectively) specific feature requests would be addressed in the
future. Furthermore, they make suggestions how specific use problems might be targeted within
the current software design.
Reflection
To sum up, commented case studies consist of indexed and annotated typical descriptions of use
written by real users, and a reflection and explanation of design decisions on the basis of these
case studies. They follow an informally proposed structure, including a short description of the
use context and purpose, the participants, the way the software was introduced and adopted
in the respective setting, and an extensive report of use experiences and ‘lessons learned’. An
extensive introduction written by the editors gives an overview and classification of the cases
presented and helps readers to select the ones that are most relevant or interesting for their pur-
poses (cf., Finck et al. 2004b, Finck and Janneck 2006).
Commented case studies satisfy several possible uses:
For developers, they bundle authentic, unedited reports from different use contexts in a
comparable way to help them develop a more thorough understanding of their requirements
beyond the observable feature requests. Furthermore, the commented case studies serve as a col-
lection of use cases for evaluation or documentation purposes.
Users are given access to the experiences of other users. If desired, they can get in contact
with them to exchange their experiences. Furthermore, by commenting the case descriptions,
developers document design decisions in a transparent way, enabling users to gain insights about
how and why CommSy is developed in the way it is, or—possibly—why design proposals us-
ers made were changed or could not been taken into account. This also makes it possible to
Figure 6: Extension of participation scope in user and developer domain and increasingly
abstract level of discussion.
Based on the different levels of abstraction that we find in both the workshops and the case
studies, we distinguish three levels where the software development process benefited from the
methods described here (Table 4): on the first level, users’ acceptance of development decisions
is increased as they learn about the rationale behind design. Reciprocally, developers obtain
legitimation as their underlying values are being confirmed by users. On the second level, em-
powerment of users comes along with a deepened understanding of the respective domain on
the developers’ side. Building upon this, on the third level, users are enabled to integrate the
software into their work practice, while developers can form a consistent vision spanning the
different use contexts.
6 Acknowledgement
First and foremost, we would like to thank all our users, especially the workshop participants
and the case study authors. We have all learned much from you, and continue to do so. This
work would not have been possible without our colleagues in the WissPro project and at the
University of Hamburg. We would like to thank Christiane Floyd and Horst Oberquelle for
laying the roots for this work, and for continuous support during the last decade. We thank
Susanne Bødker for her comments on an earlier draft, and for helping us to find our own words
for the issues described in this paper. Finally, we would like to thank the anonymous reviewers
for their patient and helpful insistence on numerous improvements to this paper.
7 References
Amaury, D., Charlier, B., France, H. and Esnault, L., “Participatory Design for Developing
Instruments for and with Communities of Practice: a Case Study,” 2nd Workshop on Dis-
tributed Participatory Design, CHI 2008, Italy, Florence (2008).
D’Andrea, V., Baskin, A., and Reinke, R., “A Distributed Software Development Team Meets a
Distributed Community of Practice: Participatory Meetings,” 2nd Workshop on Distrib-
uted Participatory Design, CHI 2008, Italy, Florence (2008).
Arias, E., Eden, H., Fischer, G., Gorman, A. and Scharff, E., “Transcending the Individual Hu-
man Mind. Creating Shared Understanding through Collaborative Design,” ACM Transac-
tion on Computer-Human Interaction, (7), 2000, pp. 84-113.
Barcellini, F., Détienne, F., and Burkhardt, J.-M., “Requirements for design participation in
Open Source Software communities,” 2nd Workshop on Distributed Participatory Design,
CHI 2008, Italy, Florence (2008).
Binder, T., Brandt, E., and Gregory, J., “Design Participation(-s),” CoDesign, (4: 1), 2008, 1-3.
Claudia Loebbecke
University of Cologne, Germany
claudia.loebbecke@uni-koeln.de
Philip Powell
Birbeck College, University of London, UK and University of Groningen,
Netherlands
p.powell@bath.ac.uk
Abstract. Participatory design (PD) and its derivative distributed participatory design (DPD)
are examples of collaborative research methods that have been successfully applied to in-
formation systems problems. Yet, there are other collaborative research methods such as
action research and design science that have also been used in the same context. This paper
argues that this trifurcation in collaborative methods is unhelpful and that the ‘walled gar-
dens’ in which these methods exist inhibit learning and the methods’ development. As PD
moves to tackle the problems that arise in distributed projects, it becomes more necessary
to look outside its own domain for solutions. This paper investigates whether collaborative
research projects that are categorized under one method also match the characteristics of
the other methods. It finds that research projects using different methods demonstrate re-
markable similarities concerning research contributions, roots, and methodological guide-
lines, but use different terminologies, and also maintain method-specific publication out-
lets and communities. Thus, insight into some of the issues raised by participatory design in
distributed contexts may arise if PD looks outside its walled garden.
Research contribution. AR contributes to both practice and theory (Baskerville 1999; Basker-
ville and Meyers 2004; Cole et al. 2005). It aims at triggering change and investigates the results
of organizational development in a constructed social system (Lewin 1946; Rapoport 1970;
Avison et al. 1999; Gummesson 2000; Davison et al. 2004). AR allows researchers to enter the
social context under consideration and conduct change in collaborative and mutually nurturing
relations with practitioners (Susman and Evered 1978; Avison et al. 1999). It targets practical
consequences and is usually organized so that each iteration of an AR process adds to the theory
(Vidgen 2002).
Roots. Until World War II, most social science research employed partial and sense data under
positivist methods. As scientists started to study behavioral issues during wartime, a call for re-
search methods respecting the socio-psychological aspects of social reality emerged. AR marked
an appropriate approach (Foster 1972; Susman and Evered 1978). The roots of AR lie in the
works of the Research Center for Group Dynamics in the US on social change and social con-
flicts (Lewin 1946; 1947; 1948) and of the Tavistock Institute of Human Relation in the UK
on socio-technical theory (Emery and Trist 1960), and in Checkland’s (1981) view of human
activity systems. The Tavistock pioneers believed that research projects should improve work
situations that were unsatisfactory in human terms and therefore fostered AR and developed
the so-called socio-technical approach (Mumford 2006). Following its genesis in this post-war
research, AR established itself as a research method suitable for academic fields as diverse as
organization studies and medicine (Baskerville 1999). An early example in the field of IS lies in
the Multiview Methodology (Avison and Wood-Harper 1990).
Generally, the action researcher’s viewpoint is an interventionist one (Baskerville 1999).
Thus, AR differs from positivist epistemological considerations and outcomes due to the impli-
cations of the researcher in the field—“prediction versus making things happen” (Susman and
Evered 1978, p. 597)—even though AR may develop positivist, critical, and interpretive forms
(Klein and Myers 1999). AR follows the main conceptual and epistemological foundations of
pragmatism and social constructivism (Baskerville and Myers 2004). Pragmatism understands
reality as based on the interactions among social selves (James 1890; Mead 1913; Haack 1976).
It demands efforts to understand how people learn and create the structures of their social sys-
tem (Dewey 1938) and paved the way for later constructivism, which stresses the importance of
Methodological Guidelines. AR requires the researcher’s participation along the entire change
process, stretching from initial reflections on the social context to implementing change in the
social system (Davison et al. 2004). AR instructs researchers to determine the requirements
for change in dialogue with the actors of the observed social systems. It expects researchers to
interpret the inter-subjective meaning of the observations (Baskerville 1999) and to contribute
to the organizational change processes and reflect upon the scientific knowledge created in the
process.
When conducting AR in IS, it is strongly suggested that the researchers start by stating the
“purpose of any action” from a conceptual and practical/contextual perspective (Baskerville and
Myers 2004, p. 333) and to follow five principles (Davison et al. 2004): (1) The ‘researcher-
client agreement’ should give clarity to both practitioners and researchers concerning the proc-
ess and the conditions of work for both parts. (2) ‘A model of cyclical process’ respects the AR
iterative tradition of research steps along a cycle of activities. (3) ‘The principle of theory’ recalls
the importance of conceptualizing the intervention. (4) ‘The principle of change through action’
puts the focus on change, claiming that any action taken should be an attempt to provide change
and therefore should not be hampered. Finally, (5) ‘the principle of learning through reflection’
assures that the research efforts lead to relevant new insights for other researchers.
As to AR guidelines, Lewin (1947) initially suggested (1) analysis, (2) fact finding, (3) con-
ceptualization, (4) planning, (5) implementation of action, and (6) evaluation. Later refine-
ments have resulted in five iterative steps: (1) Understanding and diagnosis of the situation and
its underlying dynamics, (2) action planning, (3) intervention, (4) evaluation, and (5) reflec-
tion (Susman and Evered 1978; Hult and Lennung 1980; Baskerville and Wood-Harper 1996;
Baskerville and Wood-Harper 1998; McKay and Marshall 2001; Baskerville and Myers 2004;
Davison et al. 2004; Järvinen 2005). Concerning the creation of knowledge, AR briefs research-
ers to ground interpretations in pre-existing knowledge to develop new knowledge following the
hermeneutic circle (Gadamer 1976; Gummesson 2000).
Research contribution. Answering how to questions (Walls et al. 1992), DS follows the objec-
tive “to create things that serve human purposes” (March and Smith 1995, p. 253). It aims at
designing artifacts that enhance the efficiency of the interaction between humans and technol-
ogy (March and Smith 1995; Orlikowski and Iacono 2001) and, for that purpose, applies design
theories to guide developers and reduce their uncertainties in design. DS intends to enhance IT
use and performance in organizations, striving for organizational acceptance of its outcomes
(Markus et al. 2002).
Roots. The birth of DS as a research field lies in the early efforts of governments and mili-
tary institutions to apply computing technology in the 1950s and 1960s (Banker and Kauff-
man 2004). DS is grounded in the seminal work ‘Sciences of the Artificial’ by Simon (1969)
and draws on other disciplines such as engineering, architecture, or art where problem solving
through design plays a key role (Hevner et al. 2004). “The intellectual activity that produces
material artifacts is no different fundamentally from the one that prescribes remedies for a sick
patient or the one that devises a new sales plan for a company or a social welfare policy for a
state” (Simon 1969, p. 130).
Differing from natural sciences that are based on positivism and foster insights towards
generalization, DS seeks prescriptions. It aims at improving performance through the use of IT
in a given institution, in order “to create things that serve human purposes” (March and Smith
1995, p. 253). Hence, DS enables researchers to theorize about the IT artifact itself, instead
of building variable-driven theories around the artifact (Orlikowski and Iacono 2001). With
its problem-driven orientation, it applies an engineering approach to IT research (Orlikowski
and Barley 2001). Being rooted in pragmatism (e.g., Haack 1976), where “truth lies in utility”
(Cole et al. 2005, p. 326), DS supports a situated and practice-driven vision of research, where
“truth (justified theory) and utility (artifacts that are effective) are two sides of the same coin”
and where “scientific research should be evaluated in light of its practical implications” (Hevner
et al. 2004, p. 77). Considering the interactions between technology and the members of an
Research contribution. Despite different outcome foci (social change, artifact design, and
technology design), the three methods aim at problem-solving and influencing organizational
settings in given practical situations. Referring to change in social systems, AR remains more ab-
stract than DS, which prescribes the creation of an IT artifact and studies the consequences of it.
However, the two are similar insofar as “action in action research is itself an artifact” Lee (2007,
p. 49). Referring to technology design, PD fills the gap between the other two methods.
Roots. The three methods involve system development as prime theoretical concerns, but either
have a focus on social sciences (AR), a combination of social and technically-oriented sciences
(PD), or an emphasis on technically-oriented sciences (DS). They commonly refer to the con-
cept of learning in the context of the research project. Although their respective history and
‘birth’ differ, the Tavistock Institute has influenced both AR and PD. Also, all three share epis-
temological grounds in the concept of pragmatism.
3 Research approach
This research now applies text analysis, a specific type of content analysis concerned with sys-
tematic reading of a body of texts, images, and symbolic matters. By analyzing textual data,
text analysis helps to retrieve pre-defined structures in texts and allows inferences on the basis
of retrieved structures (Krippendorf 2004). Text analysis is a widely applied qualitative method
in the social sciences and has been frequently used in the fields of anthropology and cultural
Methodological guidelines
Researcher On-site in social setting On-site in On-site at work place
intervention organization
Practitioner User contribution to Worker contribution to
Org. members’
contribution to entire artifact design full-development cycle
involvement research
Determining change/ Dialogue Observation Ethnography
design requirements
Phased, continuous, Phased, continuous, Phased, continuous,
Research process
iterative iterative iterative
studies (Bernard and Ryan 2000), business and management history (O’Connor 1999), or-
ganization studies (Prasad and Prasad 2000; Munir and Phillips 2005), organizational behavior
(O’Connor 1995; Gibson and Zellmer-Bruhn 2001), and strategic management (Vaara and
Tienari 2002; Nørreklit 2003).
In IS research, text analysis has been utilized by Järvenpaa and Ives (1990) who investigated
firms’ involvement in IT based on letters to shareholders included in business reports, by Scar-
brough et al. (2005) who reported on the role of professional media in the diffusion of knowl-
edge management, and by Gallivan and Depledge (2003) who conducted a structured content
analysis of 16 published case studies on inter-organizational systems to understand the relation
between systems use and trust and control.
Within text analysis, Lacity and Janson (1994) distinguish three approaches—positivist, lin-
guistic, and interpretivist. The positivist approach, used here, considers texts as objective data rep-
resenting the reality without interferences. The positivist researcher works primarily on content
Action research
AR √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
Lindgren
DS √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
et al. ‘04
PD √ √√ - √√ √√ √√ √ √ √√ √
AR √√ √√ √√ - √√ √√ √√ √√ √√ √√√
Salmela et
DS √√ √√ - √√ √√ √√ √√ √√ √√ √√√
al. ‘00
PD √√ √√ √ √√ √√ √√ √√ √ √√ √√
AR √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
Vidgen
DS √√ √√ - √√ √√ √√ √√ √√ √√ √√√
‘97
PD √ √√ - √√ √√ √√ √ √ √√ √
Braa & AR √√ √√ √√ √√ √√ √√ √√ √√ √√ √√√
Hedberg DS √√ √√ - √ √√ √√ √√ √√ √√ √√√
‘02 PD √√ √√ √√ √√ √√ √√ √√ √ √√ √√
Fruhling AR √ √√ √√ - √√ √√ √√ √√ √√ √√√
& De DS √√ √√ - √√ √√ √√ √√ √√ √√ √√√
Vreede
‘06 PD √ √√ - √ √√ √√ √√ √√ √√ √√
When analyzing the fifteen projects from the selected publications on the level of method
characteristics (see Table 1), 26 (of 45) combinations of project and method show a strong fit,
indicating a very strong match for most individual characteristics. Fifteen strong fit combina-
tions result from each publication following ‘its own’ method. The additional eleven strong fits
(out of 15 papers) between project and method indicate a match of a project with one of the
other two methods on the majority of individual characteristics.
Most AR projects show a strong similarity with DS characteristics; and some AR projects
also reflect PD characteristics. DS projects present some similarity with the other two methods.
Finally, PD projects indicate an almost perfect resemblance with DS characteristics and a very
strong similarity with AR characteristics (Tables 3, 4, and 5).
Four characteristics, ‘research focus’, ‘learning approach’, ‘researcher intervention’, and ‘re-
search process’ show a perfect match; at least 44 of the 45 selected projects show evidence of
PD Publications. All five PD publications show that the respective projects also meet AR char-
acteristics. For instance, Bødker (1996, p. 220) aims at “general processes of organizational
development”. Kensing et al. (1998b, p. 244) “discuss coordination in a complex organization
with multiple, different and reconfigurable groups”. Supporting a learning-by-acting approach
typical for AR, Anderson and Crocca (1993, p. 50) apply “[o]pen and continuous feedback”,
and Bødker (1996, p. 226) points to “a structured brainstorming activity meant to emphasize
critique, fantasy, and realization in three phases”. Concerning users’ involvement, Anderson
and Crocca (1993, p. 50) argue that “[t]he team recognized the need to involve the users from
the beginning in all planning and development activities”. Clement (1994, p. 57-58) finds that
“[m]anagement…released the clerical staff to participate in project activities” involving the or-
ganizational members on an ongoing basis, where “the study circles were an impetus…generally
reshaping the design process as a whole so it was more in tune with the organizational realities”
(1994, p. 58). Kensing et al. (1998b, p. 248) report that the “analytic activities conducted by
the design team involved approximately one third of the total 140 employees”. By using inter-
AR DS AR DS
PD
PD
High similarity Different terminolo-
in research focus, gies, literature and
approach, outcomes, audience
and methods
6 Notes
1. A slightly different set of iterative steps is provided by Dunne and Martin (2006) who
refer to testing artifacts in context, inducting generalizations, abducting ideas, deducting
consequences, and retesting the ameliorated artifact.
7 References
Adamides, E., and Karacapilidis, N., “A Knowledge Centred Framework for Collaborative Busi-
ness Process Modelling,” Business Process Management Journal, (12:5), 2006, pp. 557-575.
Anderson, W., and Crocca, W., “Engineering Practice and Co-Development of Product Proto-
types,” Communications of the ACM, (36:4), 1993, pp. 49-56.
Argyris, C., and Schön, D., Organizational Learning: A Theory of Action Perspective, Addison-
Wesley, Reading, 1978.
Argyris, C., Putnam, R., and Smith, D., Action science: Concepts, Methods and Skills for Research
and Intervention, Jossey-Bass, San Francisco, 1985.
Arnott, D., “Cognitive biases and decision support systems development: a design science ap-
proach,” Information Systems Journal, (16:1), 2006, pp. 55-78.
Asaro, P., “Transforming society by transforming technology: the science and politics of par-
ticipatory design,” Accounting, Management and Information Technologies, (10:4), 2000, pp.
257-290.
Avison, D., “Is IS an intellectual subject?,” European Journal of Information Systems, (12:3),
2003, pp. 229-230.
Avison, D., Lau, F., Myers, M., and Nielsen, P., “Action research,” Communications of the ACM,
(42:1), 1999, pp. 94-97.
Avison, D., Wood-Harper, A., Vidgen, R., and Wood, J., “A further exploration into infor-
mation systems development: The evolution of Multiview 2,” Information, Technology and
People, (11:2), 1998, pp. 124-139.
Avison, D., and Wood-Harper, A., Multiview: an Exploration in Information Systems Develop-
ment, McGraw-Hill, Maidenhead, UK, 1990.