See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/29678832

On various characteristics of action research

Article
Source: OAI

CITATIONS READS

7 6,049

1 author:

Pertti Järvinen
Tampere University
37 PUBLICATIONS 557 CITATIONS

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

I am working View project

All content following this page was uploaded by Pertti Järvinen on 21 December 2015.

The user has requested enhancement of the downloaded file.

 Pertti Järvinen

On various characteristics of
action research

DEPARTMENT OF COMPUTER SCIENCES

UNIVERSITY OF TAMPERE

D‐2009‐4

TAMPERE 2009
UNIVERSITY OF TAMPERE
DEPARTMENT OF COMPUTER SCIENCES
SERIES OF PUBLICATIONS D – NET PUBLICATIONS
D‐2009‐4, AUGUST 2009

Pertti Järvinen

On various characteristics of
action research

Presented in the IRIS32 Conference in Molde, Norway, August 9 ‐ 12, 2009

DEPARTMENT OF COMPUTER SCIENCES

FIN‐33014 UNIVERSITY OF TAMPERE

ISBN 978‐951‐44‐7820‐8
ISSN 1795‐4274
 1

On various characteristics of action research

Pertti Järvinen
University of Tampere, Finland
[email protected]

Abstract

Action research is normally started from practitioners’initiative. When it supports both

science and practice, hence rigor and relevance, we can highly recommend it. Iivari already
in 1991 saw action research one of the ideographic research methods. Later both Cole et al.
(2005) and Järvinen (2007) paid attention to the similarities between action research and
design research. In his paper (2007a) Iivari presented new claims on differences between
action research and design research. He continued with Venable (2009) his debate with paper:
“Action research and design science research –seemingly similar but decisively dissimilar”.
We in this paper try to understand and partially solve that contradiction. In addition to that we
also provide new information about the following two important issues: Should an action
research project be a user-guided or researcher-guided? Which kind of outcomes can action
research provide?

Introduction

Rapoport (1970) defined action research (AR) as the method which “aims to contribute both
to the practical concerns of people in an immediate problematic situation and to goals of
social science by joint collaboration within mutually acceptable ethical framework”(p. 499).
Hult and Lennung (1980) performed a wide literature survey and then formulated a new
definition of actions research: Action research simultaneously assists in practical problem-
solving and expands scientific knowledge … as well as enhances the competence of the
respective actors … being performed collaboratively … in an immediate situation … using
data feedback in a cyclical process … aiming at an increased understanding of the totality of
a given social situation … primarily applicable for the understanding of change processes in
social systems … undertaken within a mutually acceptable ethical framework”. Hence, action
research seems to satisfy the requirements of both relevance and rigor much discussed, e.g.,
in the special issues of MIS Quarterly (March 1999) and Communications of AIS (March
2001), in Information Systems.
 2

Susman and Evered (1978) described the cyclical process of action research called canonical
action research (CAR) (Figure 1).

Figure 1. The cyclical process of action research (Susman and Evered, 1978).

The cyclical process with the five steps (1.diagnosing, 2. action planning, 3. action taking, 4.
evaluating and 5. specifying learning) resembles a general problem solving process and some
information systems design methodologies. It is therefore interesting to consider how similar
action research and design research are.

An acceptance of action research in information systems was originally problematic. Ives et

al. (1980) developed a comprehensive framework for research in Management Information
Systems (MIS). The new framework is validated by mapping 331 MIS doctoral dissertations
into its research categories. The dissertations are also classified by research methodology
employed. “As part of the analysis of dissertations, the authors also classified the research
strategies employed using Van Horn’s (1973) taxonomy of MIS research methods –case
studies, field studies, field tests, and laboratory studies. Another method, action research, has
been suggested as an MIS research approach by Keen (1974) and Gibson (1975). Action
research includes the researcher as an active participant rather than a passive observer. …
Only one dissertation employed action research as part of a case study strategy.”(p. 927)

In the next ten years the situation was not much improved, because Orlikowski and Baroudi
(1991) examined 155 information systems research articles published from 1983 to 1988 and
found one action research study only in their sample. Iivari (1991) recognized action research
as a research methodology and classified it into idiographic research methods. Chen and
Hirschheim (2004) examined 1893 articles published in eight major IS publication outlets
between 1991 and 2001. They found that the five most common research designs were:
survey (41 %), case study (36 %), laboratory experiment (18 %), action research (3 %) and
field experiment (2 %), and they classified survey, laboratory and field experiments as
positivist research methods.

Chen and Hirschheim provided a historical analysis of positivist and interpretivist research
paradigms and methodologies but they consciously excluded critical studies. Richardson and
Robinson (2007) investigated critical IS research over the period of 1991-2001. They found
31 papers classified as critical research. All types of the IS studies were not, however,
identified, because design research (DR) or design science was not recognized as a separate
 3

paradigm as it could be (Järvinen 2004, 2008) nor a research method (Iivari 1991, Iivari,
Hirschheim and Klein 1998). Because two latter references belonged to the Chen and
Hirschheim’s sample and were especially known to Hirschheim, we can conclude that
constructive research methods needed in design research were not most popular in the
sample.

According to Iivari (1991, 2007a) action research is an ideographic research method and
according to Chen and Hirschheim (2004) an interpretivist one. But Cole et al. (2005) and
Järvinen (2007) saw action research similar to design research. We clearly have two
conflicting views. We must therefore ask: Could we solve this conflict and how?

Already Rapoport (1970) wrote that “in service-oriented action research conception places
the initiative with the client who states, conventionally, that he has a problem that needs
solution and approaches the social scientist who then responds by undertaking such studies as
seem required to help to solve the problem”(p. 508). Rapoport continues: “This contrasts
with the whole ethos of the academy, where protections have been erected and maintained to
take the immediate ebb and flow of practical pressures off the scholar so that he may conduct
the disinterested pursuit of knowledge with minimal interference. Initiatives in the latter case
ideally emerge purely from the internal logics of a discipline. If they are referred to the ‘real
world’outside, it is to test hypotheses or verify analyses.”(p. 508) Rapoport (1970) outlined
some advice by saying that “emphasis is placed on the collaborative nature of the enterprise,
the client providing the problem and the wherewithal to solve it and the social scientist the
concepts and methods to effect the solution. … The presenting problem might not be the most
important one with which work had to be done. The client would have to shift his perception
somewhere along the way from the initial formulation to another one as the social scientist
conducted his diagnostic work and feasibility studies.”(p.509) Thinking above raises at least
two important questions: 1) Should an action research project be a user-guided or researcher-
guided? 2) Which kind of outcomes can action research provide? After Rapoport (1970) we
have found some new material from separate sources. Hence, it is important to gather and
present that information.

Action research seems to create many questions that are different kinds. To summarize, we
are interested in consideration whether action research and design research are similar and
why and why not, or when and when not? Should an action research project be a user-guided
or researcher-guided? Which kind of outcomes can action research provide? In the next
sections we shall follow that order.

Are action research and design research similar, why and why not, when and when not?

About similarities

In order to first consider the similarity question we found that Baskerville and Wood-Harper
(1998) might be the first ones who implicitly paid attention the similarities between action
research and design research. They find “the ten action research forms: canonical action
research (Susman and Evered 1978), information systems prototyping, Soft Systems
Methodology (Checkland 1981), action science (Argyris et al. 1985), participant observation,
action learning, Multiview, ETHICS (Mumford 1986), clinical field work and process
consultation”. To our mind, information systems prototyping, Soft Systems Methodology,
Multiview (Avison and Wood-Harper1991) and ETHICS (Mumford 1986) belong to design
research, too.
 4

The explicit similarity claim is based on our study (Järvinen 2007) where we compared some
features of action research (AR) and design research (DR) (Table 1).

Table 1 Similarities of the fundamental characteristics of action research and design science

Action research Design science

AR-1: Action research emphasizes the DS-4: Design science’s products are
utility aspect of the future system from the assessed against criteria of value or utility.
people’s point of view.
AR-2: Action research produces DS-2: Design science produces design
knowledge to guide practice in knowledge (concepts, constructs, models
modi cation. and methods).
AR-3: Action research means both action DS-3: Building and evaluation are the two
taking and evaluating. main activities of design science.
AR-4: Action research is carried out in DS-5: Design science research is initiated
collaboration between action researcher by the researcher(s) interested in
and the client system. developing technological rules for a
certain type of issue. Each individual case
is primarily oriented at solving the local
problem in close collaboration with the
local people.
AR-5: Action research modi es a given DS-1: Design science solves construction
reality or develops a new system. problems (producing new innovations) and
improvement problems (improving the
performance of existing entities).
AR-6: The researcher intervenes in the DS-5: Design science research is initiated
problem setting. by the researcher(s) interested in
developing technological rules for a
certain type of issue. Each individual case
is primarily oriented at solving the local
problem in close collaboration with the
local people.
AR-7: Knowledge is generated, used, DS-6: Knowledge is generated, used and
tested and modified in the course of the evaluated through the building action.
action research project.

The first version of our paper was presented in the EURAM conference early May 2005.
Simultaneously Cole et al. (2005) prepared their paper with the supporting results: “To
examine the similarity between AR and DR, we applied the AR criteria developed by
Davison et al. (2004) to an exemplar DR paper, applied the DR criteria by Hevner et al.
(2004) to an exemplar AR paper. The exemplars selected for this cross-application were cited
by other researchers as high-quality instances of their respective research approach. For the
DR exemplar, we chose Markus et al. (2002). This study was reviewed by Hevner et al. and
found to strongly adhere to the guidelines of DR as defined by them. For the AR exemplar,
we chose Iversen et al. (2004), which, according to the editors of the September 2004 special
issue of MIS Quarterly demonstrates adherence to action research standards and serves as a
model for future action research projects (Baskerville and Myers 2004).”(p. 328)
 5

Cole et al., however, were not very lucky. Firstly, it is true that Markus et al. (2002) is one of
the first papers trying to apply Walls et al.’s (1992) ideas of design science to their research.
But what Markus et al. found was no real kernel theory but merely the six characteristics of
human being (Järvinen 2005, p. 19). Secondly, we have in our doctoral seminar every month
read and evaluated three new articles. When we read Iversen et al. (2004) I wrote in my
evaluation: “To my mind, this study is not the typical action research (Järvinen 2004, Section
5.3) but more close to the field test (Järvinen 2004, Section 3.2) of the frameworks (Table 1,
Figure 4, Table 5) and methodologies (Table 2, Figure 6) the researchers developed based on
their literature survey. I know that I over-generalize when I claim that those results the
authors reported in the article could be achieved by performing literature surveys only, and
thus without the empirical part.”Thirdly, in my evaluation (Järvinen 2005, pp. 200-202) of
Davison et al.’s (2004) article I questioned it as a good measurement instrument for action
research. Although I support the Cole et al.’s final outcome, their evidence is not without
dispute.

In addition to action research and field test the Iversen et al.’s (2004) article has the third
interpretation too. Gregor and Jones (2007) used Iversen et al.’s paper as an example of
design research in their article. Hence, Gregor and Jones slightly support that action research
and design research are similar.

Iivari and Venable (2009) analyzed the overlapping activities between AR and DR and they
identified three different cases: No overlap will take place when interests of design science
research are either ‘solving a purely technical problem by developing and evaluating a new
solution technology’or ‘solving a socio-technical problem in a non-action research context
by developing a new solution technology, but evaluating it by means other than action
research’. A slight overlap will take place when interest of design science research is
‘evaluation of a solution technology developed separately’. A significant overlap will take
place when interest of design science research is ‘solving a socio-technical problem by
developing a new solution technology and evaluating it in an organizational context’.

We can conclude that in the literature there are some other researchers that see similarities
between AR and DR. It is important that also Iivari and Venable (2009) accept that in the
socio-technical case there are similarities or overlaps between action research and design
science research, and we must also remember that Iivari and Venable restrict their design
science research to building and evaluating a technical artifact or subsystem.

Some comments on ontological and epistemological assumptions

In this sub section we first present Iivari’s (2007a) views and then we describe how some
other researchers see those aspects. Iivari (2007a) presented in his paradigmatic analysis that
“many authors associate design science with action research (Burstein and Gregor, 1999;
Järvinen, 2004; Cole et al., 2005). This is understandable, since both attempt to change the
world. Yet I [Iivari] wish to emphasize that they are historically, practically, ontologically,
epistemologically and methodologically quite different and that in my view they should be
kept conceptually clearly separate. As is well-known, action research has its roots in Kurt
Lewin and the socio-technical design movement (Baskerville and Myers, 2004), whereas
design science research has its roots in engineering. In terms of van Aken (2004), action
research has addressed more improvement problems than construction problems. It has been
much more focused on ‘treating social illnesses’in organizations and other institutions.
Technology change may be part of that ‘treatment’, but the focus has been more on adopting
 6

technology than building it. Design science research, especially in engineering and medicine,
has focused on the construction of artifacts, most of them having material embodiment. 1
Even though it may be informed by practical problems, design science research, both the
construction of new artifacts and their initial evaluation (testing), is usually done in
laboratories that are clearly separated from potential clients.”(p. 53)

I agree with Iivari that “action research has addressed more improvement problems than
construction problems”, because action research problems mostly are initiated by clients. But
this does not prevent the possibility that during the action research project, some construction
will be initiated and implemented.

Iivari (2007a) continued that “most design science research in engineering and medicine, for
example, adopts a realistic or materialistic ontology, whereas action research at least accepts
a more nominalistic, idealistic and constructivist ontology (Burrell and Morgan, 1979; Iivari
et al., 1998; Niiniluoto, 1999). As Niiniluoto (1999) points out, materialism attaches primacy
to World 1 in Popper’s classification and idealism to World 2. Action research, however, is
interested in institutions of World 3, which is socially constructed. As a consequence, design
science research, especially in engineering and medicine, has reflected a positivistic
epistemology (Burrell and Morgan, 1979) both in terms of knowledge applied from reference
disciplines (such as physics, chemistry and biology) and knowledge produced (design product
knowledge, technology norms and technology rules), whereas action research is very strongly
based on an anti-positivistic epistemology. Actually, one can claim that the very idea of
action research is anti-positivistic in its epistemology. Each client organization is unique,
with its own problems, and therefore one cannot treat all organizations using the same
medication.”(p. 53) When Iivari (2007a) claimed that “artifacts developed in design science
should first be tested in laboratory and experimental situations as far as possible”he at the
same moment selected the positivist epistemology and did not consider developers and
practitioners equal.

Braa and Vidgen (1999, p. 27) give another explanation by differentiating positivism,
interpretivism and the third view as follows: “Typically, positivism is concerned with
reducing the area of investigation in order to be able to make reliable predictions and
explanations, while interpretivism is concerned with making a reading of a situation in order
to gain understanding. We argue that in both positivist and interpretivist approaches the
researcher is making an intervention, despite aspirations to being an objective outsider (or
indeed a subjective insider), and that there will therefore be unexpected outcomes. In some
forms of research, such as action research, the aim is to gain learning and knowledge through
making deliberate interventions in order to achieve some desirable change in the
organizational setting.”

By referring to Braa and Vidgen above it seems to me that the Iivari’s view above is based “a
realistic or materialistic ontology”concerning human being, and “action research at least
accepts a more nominalistic, idealistic and constructivist ontology”concerning human being.
To my mind, the latter view on human being is more truthful than the Iivari’s one. The same
difference might also explain the different views on epistemology. I can therefore better
understand Cole et al. (2005) than Iivari who writes: “As a consequence, my conclusion is

1 In the case of medicine design science research refers to the development of new drugs and
treatments. (Iivari 2007a)
 7

just the opposite to that of Cole et al. (2005), who maintain that design science and action
research share important assumptions regarding ontology and epistemology.”(p. 53)

To summarize, we have demonstrated that if the constructivist ontology and the non-
positivist epistemology are accepted for people, Iivari does not then succeed to deny that
action research is similar to design research. It is another question whether a new artefact will
be only composed of technology and data or does it also contain people as is usual in
information systems.

Nomothetic or ideographic or constructive (design research) research methods?

In this subsection I like to analyze why I cannot support Iivari’s (1991) decision to consider
action research as an ideographic method. In my analysis I will use both the research question
and the nature of action research project to demonstrate my views.

Iivari (1991) claimed that “Burrell and Morgan (1979) distinguish two extremes in the case of
methodology: nomothetic methods and idiographic ones. Taking into account the special
character of IS and computer science as applied sciences, we can identify one more category
of constructive methods:

I. Constructive research methods

•conceptual development
•technical development
II. Nomothetic research methods
•formal-mathematical analysis
•experiments (laboratory and field experiments)
•field studies and surveys
III. Idiographic research methods
•case studies
•action research”(Iivari 1991, p. 257)

Iivari (1991, p. 257) explained nomothetic and ideographic methods by citing Burrell and
Morgan: “Nomothetic methods are ‘epitomized in the approach and methods employed in the
natural sciences, which focus upon the process of testing hypotheses in accordance with the
canons of scientific rigor’while ideographic methods place ‘considerable stress upon getting
close to one’s subject and exploring its detailed background and life history’(Burrell and
Morgan 1979, p. 6)”. Burrell and Morgan (1979, p. 6) also wrote that “the ideographic
method stresses the importance of letting one’s subject unfold its nature and characteristics
during process of investigation”.

Iivari did not question the methodological similarities of action research and design research.
He agreed with Niehaves (2007) “that design science is not primarily an epistemological
position. To me [Iivari] it is more a methodological category.”(Iivari 2007b, p. 114).

When Iivari (1991) positioned action research as an ideographic method he at the same
moment emphasized that in those action research efforts the truth (not utility) was
endeavored. When Hevner (2007) supporting Iivari (2007a) proposed that “the output from
the design science research must be returned into the environment for study and evaluation in
the application domain”he (and Iivari) could think the similar evaluation as Carlsson (2009)
proposed. The latter namely presented that in the evaluation study “the aim [is] to produce
 8

ever more detailed answers to the question of why an IS initiative - IS, types of IS, or IS
implementation - works for whom and in what circumstances”(Carlsson 2009, p. 12). He
calls this kind of evaluation “truth-driven”. The real design research Carlsson calls “solution
driven”.

When Iivari (1991, 2007a) considers action research as an ideographic research method he
then also has a certain type of research question in his mind. To concentrate on evaluation
studies and to explicate the conclusion above we can say that, in general, the ideographic
research methods are intended to reply to such a question as: Is a certain claim true or not and
why? In the ideographic evaluation studies the formulation of a question could be: Why does
a certain new information system work? On the contrary, the design research methods, in
general, are intended to reply such a question as: Can we build a certain artifact? In the
design research evaluation studies the formulation of a question could be: How good (in the
utility sense) is a novel information system? Hence, concerning evaluation studies the
differences in research questions seem to explain the conflicting views on action research.

Concerning the use of action research for evaluation of the IS initiative in the truth-driven
sense the Iivari’s classification of action research is understandable but not in the solution-
driven sense. In the latter Cole et al.’s (2005) and Järvinen’s (2008) interpretations of action
research seem understandable.

Concerning the nature of the action research project we repeat that Rapoport (1970) outlined
some advice by saying that “emphasis is placed on the collaborative nature of the enterprise,
the client providing the problem and the wherewithal to solve it and the social scientist the
concepts and methods to effect the solution”. I agree with Rapoport and argue that a
researcher can bring her or his knowledge to help in solving the client’s problem. Knowledge
a researcher provides concern some causal laws concerning relationships between two or
more changeable variables. Those relationships must be taken from either type III or IV
theories (Gregor 2006), i.e. it is possible either to predict or to predict and explain some
relationships. The prediction is needed when something is planned to be changed in action
planning phase. In a certain action research a researcher will then test the selected theory or
theories.

There are two possibilities that we shortly analyze. First, if the action research case supports
the selected theory/theories then the scientific contribution is not new but replication, and
“replications of previously published work … rarely offer enough of a contribution to warrant
publication”(Colquitt and Zapata-Phelan 2007, p. 1303). Second, if the action research case
falsifies the selected theory/theories, then we can utilize Lee’s (1989) advice for the theory-
testing case study. If all the strong requirements stated by Lee (1989) are satisfied, the
selected theory/theories can be falsified and this result is a very important scientific
contribution. Normally Lee’s requirements are not met, and the researcher might then modify
the selected theory/theories. We can conclude that if the selected theory/theories do help, the
scientific contribution is minimal if they do not help the falsification case is worth to publish.
Our characterization the nature of the action research project refers to the approach that
resembles nomothetic research methods, not ideographic ones as Iivari asserted.

In the case of falsification, the client’s problem will not be solved, and hence the next cycle
of action research is needed or the project will be finished. If the new cycle will be initiated,
the client and the researcher must diagnose the reasons of the failure and find out new
knowledge to be tried to solve the problem. The researcher can suggest the second best theory
 9

to be used, or the client can provide some “theory”based his/her experiences in practice.
Schneberger et al. (2009) call the latter a “little t”theory. The former case, the use of the
second best theory suggested by the researcher, can again be success or failure, and the
methodology and results can be evaluated as in the previous paragraph.

In the latter case, the use of the client’s theory, the researcher’s role will be changed. The
researcher will be an observer and partly an outsider in the real context and can report about a
new ‘little t’theory and its use in a real case where a part of reality was tried to change. In the
successful case the approach the researcher then applies to the object under study can be
called both ideographic one (the new client’s theory can be published) and nomothetic one
(the new theory receives support). I guess that Iivari hardly meant that action research is this
kind of ideographic method.

We can conclude that the different formulations of the research question and the nature of
action research project can explain differing views on research methods in action research.
Iivari’s view can often lead to nomothetic research method and my view to design research.
Also the consideration of the client’s and researcher’s roles might support the conclusion that
in the action research project the researcher often applies a nomothetic and rarely an
ideographic approach to.

Evaluation

Iivari (2007a) ended his criticism against action research as design research as follows:
“Despite the differences between design science and action research, I [Iivari] do not claim
that they are mutually exclusive. Action research may well be used to evaluate artifacts
developed in design science, and it may also provide information on how to improve those
artifacts. We have ample examples of the application of such action research in the context of
the developing of IS development methods (ETHICS and ISAC, for example).2 My [Iivari’s]
claim is, however, that artifacts developed in design science should first be tested in
laboratory and experimental situations as far as possible. One should not start with testing in
the real situations, except perhaps in very exceptional, special situations.”

Iivari is not alone with his view. In his comment paper on Iivari’s (2007a) article, Hevner
(2007) presented the three cycle model of design science research (relevance, design and
rigor cycles) and described that “the output from the design science research must be returned
into the environment for study and evaluation in the application domain. The field study of
artifact can be executed by means of appropriate technology transfer methods such as action
research (Cole et al. 2005; Järvinen 2007).”(p. 89).

I have some comments on the citation above. Evaluation is the fourth step in canonical action
research, but the cycle of action research also covers the three steps: diagnosing, action
planning and action taking. Iivari did not tell who will initiate the action research project in
the evaluation case, developers/researchers or clients. We shall return to this question in the

2 Recognizing the nature of systems development methods and approaches as “ways of thinking”,
“ways of control”, “ways of modeling”, “ways of working”and “ways of support”(ter Hofstede and van
der Weide, 1992), one can claim that a research approach that combines design science research
and action research is particularly appropriate when developing systems development methods.
Even in this case I am not sure that the systems development method should primarily be constructed
in an action research context. Action research may be used in the evaluation (testing) and refinement
of the method. (Iivari 2007a)
 10

next section. Hevner’s proposal that evaluation of the artifact should be executed as the field
study might be impossible, if the utility of the artifact is evaluated, not an truth value
concerning the artifact.

User-guided vs. researcher-guided action research

In this section my aim is to explore different consequences of who is an initiator of a certain

action research project, client/user or researcher. We repeat Iivari’s (2007a) claim that
“artifacts developed in design science should first be tested in laboratory and experimental
situations as far as possible”. If design research is similar to action research, I then interpret
Iivari’s view in such a way that researchers perform both design and evaluation, and users are
either totally excluded or kept as laboratory (test) animals. To my mind, this is an extreme
end (researcher-guided) in the continuum between user-guided vs. researcher-guided studies.

Baskerville and Pries-Heje (1999) described that “the [research] program consisted of a
survey and an action research study. … The client was approached with the results from the
interview survey and expressed an interest in experimenting with a prototype of the ‘missing’
problem structuring tool.”(Baskerville and Pries-Heje 1999, p. 9) Their study is an example
of the researcher-guided action research where the initiative for the study comes from
researchers.

Another example is Lindgren et al. (2004), where the researchers presented that “even though
the literature on competence in organizations recognizes the need to align organization level
core competence with individual level job competence, it does not consider the role of
information technology in managing competence across the macro and micro levels. To
address this shortcoming, we embarked on an action research study that develops and tests
design principles for competence management systems.”(p. 435) Clark (1976) recommends
that the client’s problem is primary and the researcher’s problem is secondary. From the
Lindgren et al.’s paper I received such a view that the researchers were active and went into
organizations with their ideas. To this end, I came to such a conclusion that the researchers’
research intent was primary, although the six organizations during the first cycle and two
ones during the second cycle participated in the action research projects, and with partially
financing the projects. During the second cycle the researchers themselves built two
prototypes and brought them into the two organizations for demonstrations and tests.

During our phone call between I and Iivari (2009) he explained that in action research there
are two problems: “a client’s problem that rarely produces scientific knowledge, and a
researcher’s problem that does it”. To my mind, this differentiation might explain our
conflicting views. When I has been describing similarities between action research and
design research in Järvinen (2001, 2004, 2007, 2008), I have based my assertion to a client’s
problem of action research. When Iivari has emphasized dissimilarities, he might base his
assertion to a researcher’s problem in action research.

To my mind, the client’s problem in action research is primary and it concerns some desire to
solve a certain improvement problem. In few cases a construction of a certain artifact or
innovation is also possible (cf. van Aken 2004). The goodness of the potential solution can be
measured by using some goal function (Järvinen 2008). Hence, action research and design
research are similar. Based on Clark (1976) I see the researcher’s [research] problem as a
secondary, and the researcher cannot always even study his/her research problem in
 11

connection with the action research project. If she/he can do that it will happen in the
circumstances of the action research project.

According to my understanding, Iivari emphasizes either the truth by using the ideographic
approach during the action research project or the goodness of the artifact in evaluation. As I
showed above seeking the truth might lead scientifically rather modest results and mainly
with nomothetic approaches. The idea to first develop the technical artifact in a laboratory
and then to evaluate it as the action research project means that the developers/researchers
initiate the project. - According to personal experience, those action research projects where
we as researchers presented an initiative failed, but those action research projects where
clients presented an initiative to us succeeded.

To summarize, both differentiations (user-guided vs. researcher-guided action research, and

primary vs. secondary problem) seem to at least partially explain my and Iivari’s conflicting
views on action research. We do not speak about the same problem; Iivari is speaking about
the researcher’s problem and I about the client’s problem. Clark (1976) keeps the latter as a
primary one and the former as a secondary one.

There is a certain case where the whole problematic situation, user- vs. researcher-guided or
primary vs. secondary, will vanish. That happens when a user and a researcher is one and the
same person, as in cases Coghlan (2001) and Lallé (2003). Coghlan described the special
characteristics of action research when he was both researcher and manager. “Insider action
research has its own dynamics, which distinguish it from an external action researcher
approach. The manager-researchers are already immersed in the organization and have a pre-
understanding from being an actor in the processes being studied. Challenges facing such
manager-researchers are that they need to combine their action research role with their
regular organizational roles and this role duality can create the potential for role ambiguity
and conflict. They need to manage the political dynamics, which involves balancing the
organization's formal justification of what it wants in the project with their own tactical
personal justification for the project. Manager-researchers' pre-understanding, organizational
role and ability to manage organizational politics play an important role in the political
process of framing and selecting their action research project. In order that the action research
project contribute to the organization's learning, the manager-action researcher engages in
inter-level processes engaging individuals, teams, the inter-departmental group and the
organization in processes of learning and change. Consideration of these challenges enables
manager-action researchers to grasp the opportunities such research projects afford for
personal learning, organizational learning and contribution to knowledge.”(p. 49) Lallé
(2003, p. 1097) complemented the Coghlan’s view by describing “some of the
epistemological and methodological implications involved in positioning the ’actor-
researcher’, permitting him or her, on the one hand, to play a directly useful role in an
organization, and on the other hand, to generate new scientific knowledge”.

Finally, I am willing to pay attention to a particular borderline. When Eikeland (2006a)

studied ethics and action research he found that “the normal social research ethics of
‘informed consent’is both nice and necessary, but it is still based on a fundamental
distinction between the researchers and the researched. It is a condescending ethics. Informed
consent regulates ‘our’relations to ‘them’and vice versa, but neither does it regulate their
internal relations, nor our internal relations as research colleagues. The researched are still
treated as objects subjected to observation, manipulation, or use.”
 12

When Eikeland (2006b) studied different types of knowledge he found that it depends on
dialogue between the researchers and the researched and he recommended by saying
normatively: “First of all, I [Eikeland] am not suggesting that we as researchers simply
organize dialogues for the others. I am suggesting that we enter into a dialogue with them as
practitioner colleagues in learning and research. I have suggested more about how in a recent
article on validity in action research. Secondly, we need to say more about what dialogue is
not. It is not negotiations, it is not rhetoric, etc. Thirdly, although I am all in favour of
democracy, like most people in the Nordic countries, this does not take us very far. We need
learning communities based on dialogue, not necessarily democratic structures where
rhetoric is hard to prevent. We need to pursue insights, learning, and competence
development, individually and collectively, not necessarily all-encompassing discussions
about decisions of different kinds. We also need to be clear about what kind of knowledge we
are seeking for what purposes.”Compared to Davison et al.’s (2004, p. 65) “first principle
that relates to the development of an agreement that facilitates collaboration between the
action researcher and the client”Eikeland’s recommendations are more balanced and
sympathetic.

Potential outcomes of action research

We earlier referred to Braa and Vidgen (1999, p. 27) who emphasized that “in some forms of
research, such as action research, the aim is to gain learning and knowledge through making
deliberate interventions in order to achieve some desirable change in the organizational
setting.”

Concerning knowledge produced Eikeland (2006b) “presented some thoughts on knowledge

forms or ways of knowing extracted and borrowed from Aristotle. … Aristotle did not start
out with abstract knowledge as such, however. And he did not think about knowledge from
the standpoint of specialized academic institutions. His thinking about knowledge is
fundamentally relational. This is important. In his way of thinking there is always a knower
and something or somebody known who relate to each other in specific and different ways.”

“The knowledge forms in Table 2 are, however, not all of them found as explicitly in the
writings of Aristotle.”

Table 2. Different knowledge forms based on Aristotle

Basis Way of knowing Associated rationality English equivalent

Aísthesis Theoresis = Deduction Spectator speculation
(perception) episteme2
?? Being affected passively
Páthos from the outside
Empeiría Khresis Tékhne (calculation) Using
(practically Poíesis Making, manipulating
acquired Praxis2 Phrónesis (deliberation) Doing: virtuous
experience) performance
Praxis1 Dialectics / dialogue. The Practice, training for
way from novice to expert, competence development
from tacit to articulate and insight (theoría)
Theoría = episteme1 Dialogue, deduction, Insight
deliberation
 13

Eikeland (2006b) starts “by saying something about the concept of episteme, often but rather
confusingly, he think, translated as ‘science’. As you can see from Table 2, there are two
concepts of theory and episteme, one on the top and one at the bottom. … But the two forms
of theory are still fundamentally different and need to be kept apart.”

“The first one, called theoresis, episteme2, or ‘spectator speculation’in Table 2, is based on
observation at a distance. Theoresis relates to external objects. The relation, or the ethics
implied between the knower and the known, is difference, distance, separation, non-
interaction, and non-interference. In its deductive form, astronomy has served as a paradigm,
and for social and historical reasons, this astronomical model conquered the whole field of
science from the 17th Century on. … Still, the people studied, the people known, are the
others, not the knowers themselves. … It is important not to intervene, and to neutralize any
unintended effects of observation or of the research activities at all.”

Eikeland (2006b) used “grammar as the paradigm example for this other kind of episteme1 or
theoría in the table. I have translated theoría with ‘insight’. In grammar the relation between
the knower and the known is quite different from the corresponding relation in astronomy.
Grammar is about ourselves as native speakers of a language. It expresses and organizes
certain aspects of our linguistic practice, the more or less stable patterns that repeat
themselves in certain ways in our performance. Grammar is descriptive and analytical, but it
is also normative, since it delivers standards for correct speech and writing. … There is no
distance between the knower and the known as in astronomy.”

Eikeland (2006b) continued that “grammar also exemplifies what is called praxis knowledge
in the table, where the relationship between the starting point, the means, and the end or
objective for our actions is one of formal equality. Praxis knowledge is the primary base for
theoría, the episteme-form at the bottom of the table. … For Aristotle, praxis knowledge
represents a relationship between colleagues sharing common standards for how to go about
their professional activities. … Praxis knowledge regulates the relationships between equals.
It constitutes a ‘we’literally as a community with common standards (as in grammar), and it
regulates relations among ‘us’, not between ‘us’and ‘them’.

Eikeland (2006b) described that “by searching patterns, sorting similarities and differences in
our accumulated practical experience, and in how we use language, dialog helps to articulate
what we carry with us as tacit knowledge, and it helps us on our way from novices to experts
and to virtuoso performers. This is what is called praxis1 on the second lowest row in the
table.”

According to Eikeland (2006b) “with grammar the practical enactment is often immediate
and spontaneous in proficient speakers. … But in other fields where the practice is not
equally standardized and ‘automated’, … we need deliberation or phrónesis trying to find out
how to act in the most just or fair way towards someone we must act in relation to right here
and now. This is what is called praxis2 on the third lowest row in the table.”

Eikeland (2006b) only wrote that “páthos is knowledge created from being passively affected
by external sources. … Khresis is competence in using external or reified things for purposes
of the user. It is independent competence, as for example in driving a car. … Khresis is
possible and even prevalent in human relations too, of course. Some are even very good at it.
But the ethics of using other people as instruments for achieving your own interests is hard to
 14

defend generally … as an ethics between un-equals. … In addition, work life seems to be full
of these kinds of relationships in apparently legitimate and institutionalized forms.

With poíesis the case is similar. It is competence in manipulating external objects according
to the manipulator’s own plans and intentions, making something out of them as materials.
Poíesis intervenes artificially in its material. I think the conventional experiment is a variant
of poíesis. … The art of manipulating others is hard to defend ethically on a general basis.”

“This presentation of different ways of knowing now makes it easier to return to the subjects
I discussed in the article on condescending ethics in Eikeland (2006a), and to work life
challenges. It makes it easier to understand why many forms of action research do not
communicate very well or easily with institutional review boards controlling the ethical
quality of research projects. As I hope has become clear, some forms of condescending ethics
are inherent to all the knowledge forms in the table, except to praxis and theoría. Things,
animals, or other human beings observed, manipulated, or used are not one of ‘us’the
observers, manipulators, or users. But as practitioner researchers or co-researchers in action
research projects, they are among us the researchers. For review boards this creates
confusion. We can discuss exactly how much of modern social research that produces
knowledge in the form of either theoresis, khresis, or poíesis. It may not be all of it, but I
think we have to say that most of it does. … At least Table 2 of Aristotelian knowledge forms
suggested that ‘othering’ways of doing social research are not necessary, natural, or the only
possible ways.”

Eikeland clearly recommend that we should only report such action research knowledge that
both researchers and practitioners together accept and support. In addition to scientific new
knowledge, action research can also produce new instantiations or new artifacts, if we accept
that action research is similar to design research.

Discussion

We found that action research and design research are especially similar in improvement
studies. The constructivist ontology and the non-positivist epistemology seem better than
other alternatives to be applicable to people. These ontological and epistemological views can
partly explain why Iivari does not see action research similar to design research. The
methodological difference seems to disappear when the use of action research is taken in the
truth-driven sense and for evaluation of the IT artefact.

Based on Carlsson (2009) we identified two types of evaluation studies, both truth-driven and
solution-driven. This result might a bit change our classification (Järvinen 2004, 2008) of
research methods. This also arouses a question: It is possible to find two different
formulations for another main activity of design research, namely for building?

The differentiation between the client’s problem as a primary and the researcher’s problem as
a secondary seems to mainly explain differing views that Iivari and I have on action research.
According to personal experience, those action research projects where we as researchers
presented an initiative failed, but those action research projects where clients presented an
initiative to us succeeded.

For characteristics of research outcomes, we found Eikeland’s views most suitable. To our
mind, Eikeland’s ethically sound finding is he does not separate “we”as researchers and
 15

“they”as practitioners, but prefers to consider all the participants, both “us”and “them”as
researchers, and hence at the equal level. This democratic view is very typical for our
Scandinavian community, and we should always remember to support it in all our research
efforts.

It was in concordance of Eikeland’s ideology that he recommends that we should only report
such action research knowledge that both researchers and practitioners together accept and
support. We supplemented his idea that in addition to scientific new knowledge, action
research can also produce new instantiations or new artifacts, if we accept that action
research is similar to design research.

References:

Argyris C., R. Putnam and D. Smith (1985), Action science, Jossey-Bass, San Francisco.
Avison and Wood-Harper (1991), Information systems development research: An exploration
of ideas in practice, The Computer Journal 34, No 2, 98-112.
Baskerville R. and M.D. Myers (2004), Special issue on Action Research in Information
Systems: Making IS research relevant to practice –Foreword, MIS Quarterly 28, No 3, 329-
335.
Baskerville R. and J. Pries-Heje (1999), Grounded action research: A method for
understanding IT in practice, Accounting, Management and Information Technology 9, No 1,
1-23.
Baskerville R. and A.T. Wood-Harper (1998), Diversity in information systems action
research methods, European Journal of Information Systems, 7, 90-107.
Braa K. and R.T. Vidgen (1999), Interpretation, intervention and reduction in the
organizational laboratory: A framework for in-context information systems research,
Information and Organization 9, No 1, 25-47.
Burrell G. and Morgan G. (1979), Sociological paradigms and organisational analysis,
Heinemann, London.
Burstein, F. and s. Gregor (1999), The systems development or engineering approach to re-
search in information systems: An action research perspective, in Hope, B. and Yoong, P.
(eds.), Proceedings of the 10th Australasian Conference on Information Systems, Victoria
University of Wellington, New Zealand, 122-134.
Checkland P. (1981), Systems thinking, systems practice, Wiley, Chichester.
Chen W.S. and R. Hirschheim (2004), A paradigmatic and methodological examination of
information systems research from 1991 to 2001, Information Systems Journal 14, No 3, 197-
235.
Clark A.W. (1976), Experimenting with organizational life, action research approach,
Plenum, New York.
Coghlan D. (2001), Insider action research projects - Implications for practising managers,
Management Learning 32, No 1, 49-60.
Cole R., S. Purao, M. Rossi and M.K. Sein (2005), Being proactive: Where action research
meets design research, In Proceedings of 26th International Conference on Information
systems (ICIS), AIS, , 325-336.
Colquitt J.A. and C. P. Zapata-Phelan (2007), Trends in theory building and theory testing: A
five-decade study of the Academy of Management Journal, Academy of Management Journal
50, No 6, 1281-1303.
Eikeland O. (2006a), Condescending ethics and action research, Action Research 4, No 1, 37-
47.
 16

Eikeland O. (2006b), Ethics and action research, key note speech at Finnish work research
days, Oct 6-7, 2006 at Tampere, 17p.
Gibson C.F. (1975), A methodology for implementation research, in Schultz and Slevin
(Eds.), Implementing operations research/management science, American Elsevier, New
York.
Gregor S. (2006), The nature of theory in information systems, MIS Quarterly 30, No 3, 611-
642.
Gregor S. and D. Jones (2007), The anatomy of a design theory, Journal of the Association
for Information Systems 8, No 2, 312-335.
Hult M. and S.-Å. Lennung (1980), Towards a definition of action research: A note and
bibliography, Journal of Management Studies 17, 241-250.
Iivari J. (1991), A paradigmatic analysis of contemporary schools of IS development, Eur. J.
Inf. Systs 1, No 4, 249-272.
Iivari J. (2007a), A paradigmatic analysis of Information Systems as a design science,
Scandinavian Journal of Information Systems 19, No 2, 39-64.
Iivari J. (2007b), Nothing is as clear as unclear –Iivari’s response, Scandinavian Journal of
Information Systems 19, No 2, 111-117.
Iivari (2009), A phone call, May 14, 2009.
Iivari J., R. Hirschheim and H.K. Klein (1998), A paradigmatic analysis contrasting
information systems development approaches and methodologies, Information Systems
Research 9, No 2, 164-193.
Iivari J. and J. Venable (2009), Action research and design science research –seemingly
similar but decisively dissimilar, a paper accepted to ECIS 2009, June 8 - 10, 2009 in Verona,
Italy.
Ives B., Hamilton S. and G.B. Davis (1980), A framework for research in computer-based
management information systems, Management Science 26, No. 9, 910-934.
Järvinen P. (2001, 2004), On research methods, Opinpajan kirja, Tampere.
Järvinen P. (2005), IS Reviews 2005 (as an editor)
http://www.cs.uta.fi/reports/dsarja/D-2005-7.pdf
Järvinen P. (2007), Action research is similar to design science, Quality & Quantity Vol. 41,
No 1, 37-54.
Järvinen P. (2008), Mapping Research Questions to Research Methods, in IFIP International
Federation for Information Processing, Volume 274; Advances in Information Systems
Research, Education and Practice; David Avison, George M. Kasper, Barbara Pernici, Isabel
Ramos, Dewald Roode; (Boston: Springer), pp. 29-41. http://dx.doi.org/10.1007/978-0-387-
09682-7-9_3
Keen P.G.W. (1974), Towards a behavioral methodology for the study of OR/MS
implementation, Unpublished paper, MIT.
Lallé B. (2003), The management science researcher between theory and practice,
Organization Studies 24, No 7, 1097-1114.
Lee, A.S. (1989), A scientific Methodology for MIS case studies, MIS Quarterly 13, No. 1,
33-50.
Lindgren R., Henfridsson O. and U. Schultze (2004), Design Principles for Competence
Management Systems: A Synthesis of an Action Research Study, MIS Quarterly 28, No 3,
435-472.
Mumford E. (1986), Using computers for business success: The ETHICS method,
Manchester Business School, Manchester.
Niehaves B. (2007), On epistemological pluralism in design science, Scandinavian Journal
of Information Systems 19, No 2, 93-104.
Niiniluoto, I., 1999, Critical Scientific Realism, Oxford University Press, Oxford.
 17

Orlikowski W.J. and J.J. Baroudi (1991), Studying information technology in organizations:
Research approaches and assumptions, Information Systems Research 2, No 1, 1-28.
Rapoport R.N. (1970), Three dilemmas in action research, Human Relations 23, No 6, 499-
513.
Richardson H. and B. Robinson (2007), The mysterious case of the missing paradigm: A
review of critical information systems research 1991-2001, Information Systems Journal 17,
No 3, 251-270.
Schneberger S., C. Pollard and H Watson (2009), Theories: For academics and practitioners,
Information Systems Management 26, No 1, 52-60.
Susman G.I. and R.D. Evered (1978), An assessment of the scientific merits of action
research, Administrative Science Quarterly 23, No 4. 582-603.
ter Hofstede A.H.M. and T.P. van der Weide (1992), Formalization of techniques: Chopping
down the methodology jungle, Information and Software Technology 34, No 1, 57-65.
van Aken J.E. (2004), Management research based on the paradigm of the design sciences:
The quest for field-tested and grounded technological rules, Journal of Management Studies
41, No 2, 219-246.
Van Horn R.L. (1973), Empirical studies of management information systems, Database 5,
No. 2 & 3, 172-180.

View publication stats