Knowledge Creation Through Interactive Research:

A Learning Approach1
Per-Erik Ellström
HELIX VINN Excellence Centre
Linköping University
SE - 581 83 Linköping
e-mail: per-erik.ellstrom@liu.se

1. Introduction

Knowledge creation through the development of mutually productive forms of collaboration

between research and practice, has for a long time been an important issue for researchers as
well as practitioners in different fields of working life. In the social sciences there is also a
long tradition of criticism of traditional research models, and a corresponding interest in
different models of practice oriented and collaborative research2 (e.g. Adler, Shani & Styhre,
2004; Argyris, 1980; Clark, 1976; Reason & Bradbury, 2001). Traditional models of research
has been be criticized for objectifying the participants, and for being too inflexible, closed,
specialized, and of little practical relevance etc. According to many of its critics, one major
problem with traditional research models is a too strict distinction between basic and applied
research, and in line with this, an assumed linear relationship between theory and practice,
and a role model of the researcher as basically detached from practice (Brulin, 1998; Gibbons
et. al. 1994; Svensson et al, 2002; Toulmin & Gustavsen 1996).

Conversely, from a perspective of traditional (mainstream) social science research, different

varieties of collaborative research have often been characterized by a number of difficulties
and alleged flaws, and the scientific value of collaborative research has for a long time been a
matter of debate within the social science research community. A classical criticism concerns
the potential threats to the validity of the research results, due to the involvement of the
researcher in practical activities. On the other hand, there are also counter arguments to the
effect that a collaborative relationship between researchers and practitioners may contribute,
first of all, to a better access to important processes, but also to more valid data compared to
traditional research models with their typically more authoritarian relations between the
researchers and the subjects of research (Argyris, 1980; Eikeland, 2006).

A perhaps more serious criticism against collaborative research models concerns the risk that
maximizing the contribution to practice, tends to minimize the contribution to research, and
vice versa (Sandberg, 1981; Seashore, 1976; Svensson, Ellström & Brulin, 2007). As argued
by the latter authors, action research is simply often too close to practice and too practically
oriented with limited scientific ambitions. The objective to change things has often taken
precedence over the interest in producing general and critical knowledge. By implication,
there is the risk, as stated by Seashore (1976), that the notion of action research becomes a

1
Paper to be presented at the ECER Conference, Gothenburg, September 10–12, 2008.
2
The notion of collaborative research is used here to refer to an umbrella concept covering a wide
range of variants, including action research, interactive research and participatory research. In this
paper there will be a focus on the concept of interactive research.
 2

justification for practical development work masquerading as research, and, conversely, for
research being reduced to a more or less trivial service role.

Among proponents of action-research a not too uncommon reaction to this kind of criticism
has been to try to circumvent and avoid it by claiming that action research represents a
different and ”alternative” epistemology compared to mainstream research. One recent
example of this type of argumentation is pursued by Reason & Bradbury (2001). According to
these authors, any concerns about a trade-off between research and practice in the conduct of
action research is just another expression of a ”positivist” research tradition. However, such a
denial of a trade-off between research and practice usually has a price. For Reason and
associates this price means that they have to give up the fundamental aim of research, that is,
to create new knowledge in the sense of concepts, models, and theoretical frameworks. In
order to argue their case, they propose what they call an ”extended epistemology”. In essence,
this means a shift in emphasis: ”from the traditional emphasis on propositional knowledge and
the written word to practical knowledge and the manifest deed.” (Heron & Reason, 2001). Of
course, this shift in emphasis may be viewed as an illustration of the original concern that a lot
of collaborative research tends to sacrify traditional aims of research for more practical
concerns. According to my mind, this type of argumentation does not solve the basic problem,
that is, the trade-off between the aims and interests of research and those of practice. This
issue is still a compelling one, and a dilemma that has to be directly dealt with in one way or
the other (rather than avoided) if we are to become successful in conducting good
collaborative research.

How, then, can this dilemma be resolved or at least handled in a satisfactory way? Is it at all
possible to carry out collaborative research in a way that is productive both from a practical
and from a scientific perspective? Although much has been written about different forms of
collaborative research, there is a lack of empirically grounded studies concerning the use of
such a research approach in practice. In this paper, a model of what is called interactive
research is presented and discussed. Thus, rather than using the concept of action or
collaborative research, the notion of interactive research will be used.

In line with Svensson, Ellström & Brulin (2007), the concept of interactive research as used
here refers to a research approach which position itself in contrast to mainstream academic
research on the one hand and action research on the other hand. The ideas and reflections
presented in this paper has been developed through a series of research projects carried out in
co-operation between researchers at the CMTO (”Centre for Studies of Humans, Technology
and Organization”), Linköping University (www.liu.se/cmto) and a number of industrial
companies and public sector organizations for a number of years (see Ellström et al, 1999).
The notion of interactive research as presented in this paper is also applied as an important
part of the research strategy of the recently established HELIX VINN Excellence Centre at
the same university (www.liu.se/helix).

What, then, could be meant by the notion of interactive research? How to organize interactive
research? What trade-offs may be involved? The purpose of this paper is to – based on the
experiences mentioned above – try to shed some light on these questions. This is done, first,
by proposing a definition of the concept of interactive research, and, secondly, by giving an
outline of an emerging model of interactive research. In addition, the ideas and arguments
presented below are also attempts to deal with some of the classical criticisms against
collaborative research (and, by implication, in many cases also against interactive research)
that were previously mentioned.
 3

2. On the Concept of Interactive Research

With certain ups and downs collaborative models of research have been part of the social
sciences at least since the 1930s and onwards (see e.g. Chein, Cook & Harding, 1948), with
perhaps its most intensive and creative period during the 1960s and 1970s (Clark, 1976).
However, in spite of its long history, it seems as if the collaborative research approach never
really got off the ground. Still, the separation of social science and practice is strongly
institutionalized.

However, there are signs of ongoing changes (and proposals for change) towards a stronger
integration of research and practice (e.g. Aagaard-Nielsen & Svensson, 2006; Etzkowitz,
1998; 2003; Gibbons et. al. 1994; Nowotny et. al 2001; Sörlin, 1996). Specifically, in a
Swedish context, there is a growing interest in interactive research. This is attested to, among
other things, by the establishment of the Swedish Association for Interactive Research (SIRA)
a few years ago. This organization involves a number of research milieus with a focus on
interactive research. Furthermore, there is a legal requirement on the universities to engage in
co-operative efforts concerning education, research and development with companies,
agencies, different kinds of associations, and other organizations (the so called third task of
the universities). It has also been argued that this new legal requirement on the universities
may prove to be an impetus to a new model for research, development and innovation based
on an interactive model of knowledge creation and use (Brulin, 1998).

What, then, could be meant by the concept of interactive research? A general definition is
given by Caswill & Shove (2000), which states that interactive research means a research
approach:

”where researchers, funding agencies and ’user groups’ interact throughout the entire research process, including
the definition of the research agenda, project selection, project execution and the application of research
insights.” (Caswill & Shove, 2000).

More recently Svensson, Ellström & Brulin (2007) has argued that interactive research can be
seen as a continuation of action research in some respects, but also with quite distinct ideas
concerning, among other things, the tasks and objectives of research and the researcher’s role.
Specifically, these authors argue that there are a number of problems and difficulties with
action research that need to be attended to, and, if possible, eliminated or at least alleviated.
For example, it is necessary to be aware of and try to deal with at least the following
problems, which have tended to characterize much action research in the past:

• The strong involvement of the researcher in practical development work, which may decrease
the attention and time that it is possible to devote to research activities (e.g. to the writing of
research papers), and, thereby, also lead to an decrease in scientific productivity as usually
measured;
• A risk that the strong proximity to practice and the personal involvement of the researcher
makes it more difficult to conduct a critical analysis;
• A focus on local understanding and the production of practical knowledge at the expense of a
more general analysis and the production of theoretical knowledge;
• A lack of acceptance and legitimacy in the general academic scientific community.

What then do we mean by the concept of interactive research? As used here, this concept is
defined as constituted by two basic ideas, called the idea of the threefold task and the idea of
knowledge creation through collaboration between researchers and practitioners.
 4

The Idea of the Threefold Task

What is here called the idea of the threefold task refers to the idea that interactive research
aims to contribute both to practical concerns, for example, how to handle practical issues in
relation to organizational or technological change, and to the creation of scientifically valid
knowledge, for example, new concepts, theories, and models. In addition, however, a third
task needs to be included, namely the educative task of enhancing the knowledge and
competencies of the parties involved in the research process through processes of individual
and collective learning. Indeed, it could be argued that this third task is fundamental to the
possibility to accomplish the other two tasks, and that therefore interactive research is
essentially about joint learning between the participants and the researchers throughout the
entire research process (Svensson, Ellström & Brulin, 2007).

As argued above, a common problem, or rather dilemma, of much action research is the
tendency to emphasize one of these tasks at the expense of the others. For example,
maximizing the interests of practice at the expense of the interests of research or vice versa.
Depending on which of the tasks that is emphasized, action research may, in practice, be
reduced to a research strategy, a strategy for practical problem solving and change, or an
educational strategy. The position taken in this paper is that the comparative advantage of
interactive research is its potential for combining and integrating the concerns of research,
development work (social change) and learning. Of course, this is also where the problems
and challenges start.

The Idea of Knowledge Creation Through Collaboration Between Researchers and

Practitioners

The second basic idea of the concept of interactive research as used here, is its emphasis on
knowledge creation through an egalitarian co-operation between researchers and practitioners
with a focus on a common problem or research object (Brulin, 1998; Rapoport, 1970; Van de
Ven & Johnson, 2006; Woolgar, 2000). This idea means essentially that there is a certain
degree of sharing of responsibility and power between the parties involved in the research
process.

Sometimes this and related ideas of interaction between researchers and practitioners is
interpreted in quite idealistic terms, emphasizing a subject-to-subject relationship and a
personal encounter, while ignoring differences in identities, competencies, responsibilities and
roles between the actors. Differences that may have their roots in cultural differences between
the two spheres of research and practice, respectively (e.g. differences with respect to
autonomy), but which also mirror the division of labour in society at large. In line with such a
view, we rather emphasize the importance of a clear division of labour between researchers
and practitioners based on their different interests, responsibilities, and competencies. Thus,
rather than trying to erase the differences between the spheres of research and practice we
believe that it is very important to respect and preserve these differences (see also Brulin,
1998; Edquist & Flodström, 1997).

At the same time, co-operation between researchers and practitioners is likely to bring to the
fore a number of problems concerning authority, power, and participation, which need to be
carefully analyzed and taken into account in the planning and conduct of research. In addition,
there are important ethical issues that need to be dealt with. In order to establish a mutually
acceptable ethical framework it appears to be necessary to discuss and establish a consensus
 5

concerning the values and goals underlying the research effort, the methods to be used, and
the expected outcomes of the research at the outset of the research process.

Of course, co-operation between researchers and practitioners also have a number of

methodological implications. At a general level, it is a rather well established fact that the
relationship between researchers and subjects affects the outcomes of research (Argyris, 1980;
Rosenthal, 1966; Eikeland, 2006). Concerning the more specific question of in what way a co-
operative relationship between researchers and practitioners affects the reliability and validity
of research data, there is little or no direct evidence. As already mentioned in the introductory
section, there exists, however, rather general assumptions and speculations of about such
effects. These assumptions range from those emphasizing co-operation as a threat to the
validity of research, to those stressing the importance of co-operation for getting access to
organizational processes and obtaining valid data. However, taken together available
methodological analyses seem to indicate that collaborative research designs, when used
under favourable conditions and competently conducted, is not inferior to traditional forms of
research with respect to the validity of the research results (Argyris, 1980; Eikeland, 2006).
Although crucial to the advancement of interactive research, these methodological problems
and the underlying problems of epistemology will not be further dealt with in this study.
Rather, the focus of the remainder of this paper will concern how to conceive of and organize
interactive research. More specifically, there will be a focus on exploring, and, if possible,
expanding the two defining ideas of interactive research that were introduced above.

3. Organizing Interactive Research: The Case of HELIX Research

and Innovation Centre

The HELIX3 Research and Innovation Centre was formally established as a so-called Centre
of Excellence at Linköping University the 1st of January 2006. The establishment of the
Centre means a concentration of resources for research and innovation in the field of working
life research. More specifically, the research programme has a focus on the management of
mobility – of people and ideas – in relation to working life development (see below for a short
description of the research programme; see also www.liu.se/helix). A central task of the
HELIX Centre is to carry out research and innovation activities that contribute to research-
based knowledge and theory and, at the same time, add value to practice.

The HELIX Centre is based on a Triple Helix logic (Etzkowitz, 2003), and is organized as a
regional partnership between research (represented by Linköping University) companies,
public sector organizations and labour market organizations. The Centre has a long term
funding – ten years – from the Swedish Governmental Agency for Innovation Systems
(VINNOVA), from Linköping University, and from companies, public sector organization,
and labour market organizations. The funding is divided equally between the Governmental
Agency (VINNOVA), Linköping University, and the other members of the partnership.

In general terms, the activities of the HELIX Centre includes: (a) research; (b) education and
training – a graduate programme (“HELIX Graduate School”), but also contributions to
relevant undergraduate courses and programmes within the university; and (c) innovation

3
The HELIX acronym could be read as Health, Learning, and Innovation (E)Xellence. However, the intent is
also to allude to a helix – a spiral – vindling towards exellence in our fields of research and innovation, and, of
course, to the idea of Triple Helix as an organizing principle behind the centre.
 6

activities with respect to process and product innovations (improvements) as well as

commercial and public service development. An important part of the goals and research
strategy of the Centre is to attain significant synergy effects between these three areas. An
important part of HELIX’ research strategy is the interactive research approach. However,
before this approach is presented, an outline will be given of the HELIX research programme
and the organization of the Centre.

A Multi-Disciplinary Research Programme

The HELIX research area concerns mobility – and the management of mobility – in relation
to working life development. The notion of mobility is used in a broad sense, and as a
generative concept including horizontal (e.g. job rotation) as well as vertical mobility (e.g.
from lower to higher levels of responsibility and authority); mobility between production and
development; mobility within and between sectors of the labour market (e.g. between the
private and the public sector); within and between organizations and occupations (e.g.
between traditionally male and female occupations); and mobility from unemployment to
employment, or from being on sick leave to getting a job.

Our focus on mobility means an emphasis on exploring and developing conditions and
processes that will promote what has been called “the good mobility". Thus, the notion of
mobility does not refer to labour market flexibility in a narrow sense, emphasizing for
example numerical flexibility by making use of part-time, temporary or short-term contracts.
Rather, we put a strong emphasis on measures attempting to integrate mobility (flexibility),
participation, gender equality, and job security. This research orientation has been fuelled not
least by observed negative effects of a low degree of mobility – that is, employees being
"locked into" an unfavorable working situation (Aronsson et al., 2000). Other studies indicate
that mobility, under certain conditions, is likely to have a positive impact on individual
learning and development (Brown, Green & Lauder, 2001); on knowledge flows between
firms in regional clusters (Malmberg, 2002); and on innovative activities at an organizational
level (Michie & Sheehan, 2003).

The overall research programme (for a description of the research programme and projects,
see www.liu.se/helix) can be described in terms of three interrelated sub-programmes of
research and innovation focusing on:

• Mobility in relation to workplace learning and competence development

• Mobility in relation to health promotion.
• Mobility in relation to idea development and entrepreneurship.

These sub-programmes as well as ongoing and planned research projects emerged from a
series of dialogue seminars within the HELIX partnership (see below). Three thematic issues
are common to all three sub-programmes, creating a matrix of interrelated knowledge
interests (see Figure 1): (a) participation in development and other organizational processes;
(b) gender equality; and (c) leadership with respect to the promotion of learning, health, and
innovation.
 7

Research Programmes
Mobility in Relation to:
Idea Development
Thematic Issues: Learning Health Promotion and Entrepreneurship

Participation

Gender Equality

Leadership

Figure 1: The Matrix of Interrelated Knowledge Interests Behind the HELIX Research and Innovation Centre

The idea behind the matrix is that the columns and rows of the matrix represent different
perspectives on mobility, and that there is a two-way road between mobility and the aspects
that are emphasized in the matrix. On the one hand, there is a focus on mobility as a driving
force for learning, health, and innovation – moderated by aspects related to employee
participation, gender/equality and leadership. Furthermore, there is an interest in how
different organizational aspects – including aspects related to employee participation,
gender/equality and leadership – together with available systems for learning/competence
development and health promotion may promote or hinder mobility.

The research programme has a multi-disciplinary basis. Thus, the HELIX researchers
represent different disciplines (business administration, education, health sciences, innovation
and entrepreneurship, sociology, and work science) with a basis in more or less divergent
theories and traditions. This multi-disciplinary orientation creates a strong potential for
innovative research, but also for increasing the robustness and validity of research results. The
multi-disciplinary approach involves creating an environment for research where there is a
potential for discovering the strengths and weaknesses of one’s own disciplinary matrix for
studying certain problems, and also for supporting innovative thinking and creativity in
understanding and analysing mobility. This is facilitated through regular interaction, seminars,
discussions and systematic work in common integrative projects that cut across the sub-
programmes (see www.liu.se/helix).

A Partnership Approach

A basic characteristic of the organization of the HELIX Centre is the establishment of a

regional partnership of researchers and representatives from companies, public sector
organizations, and labour market organizations. The partnership meets on a regular basis 4 to
5 times a year. Furthermore, different constellations of partners co-operate in different
research and innovation projects.

The partnership has de facto been of outmost importance in the initiation of the HELIX
programme: by being involved in the process of developing the research programme; by being
involved in the process of generating projects; by giving access to workplaces; by putting in
resources; by giving legitimacy for the different projects etc. In addition, the partnership has a
number of other important functions in the R&D process, specifically: (a) to take an active
ownership of the outcomes of the R&D process, that is, to be involved in the implementation
and dissemination of research results; (b) to take part in the joint learning processes of
knowledge production in order to make them more innovative and to increase the validity of
 8

the findings; and (c) to take part in the public debate of how to organize research and
developmental processes on a regional level. Thus, the partnership is assumed to be able to
have a key role in the production and transfer of knowledge.

In theoretical terms, the partnership is viewed as a developmental coalition (Asheim, 2001;

Gustavsen, Finne & Oscarsson, 2001). In line with this concept, an important task for the
HELIX Centre is to create arenas and process support that create the potential for not only
individual leaning, but also for collective learning within the partnership. Such collective
learning processes could be expected to result, for example, in a common language and a
common culture that in turn could be expected to facilitate innovations (Lawson & Lorenz,
1999) with respect to issues of mobility, learning, health, and innovation.

Management and Organization of the Centre

The HELIX Centre has the status of a special work unit within Linköping University, with its
own economy, its own premises, and led by a Board and a Director. Formally HELIX has an
administrative coupling to a so called ”host department”, which from the start has been the
Department of Behavioural Sciences. The relations between the HELIX Centre and the “host
department” and other concerned departments or units within Linköping University are
regulated by a written contract.

The HELIX organization is built on the principle that the researchers, doctoral students and
administrative staff keep their positions at their "mother departments" within the university.
Thus, there will be a kind of "staff-union" between HELIX and other departments/disciplines
within the university. In practice, this means that there are no formal positions for researchers,
administrators or doctoral students within the HELIX organization. Although the post-
graduate programme for doctoral students is planned and carried out on a multidisciplinary
basis, the students will obtain their degrees in the subject field to which they were originally
admitted; therefore, the programme needs to be planned and realized in close cooperation
with concerned departments within the university.

In addition to the tasks of the partnership mentioned above, the partnership has an important
role in the nomination of members of the Centre Board. The Board comprises 12 regular
members, representing the members of the partnership, that is, the companies, the public
sector organizations, the labour market organizations, and research. The tasks of the Board
have an emphasis on the strategic management and development of the Centre, which
includes decisions concerning annual and long-term planning, the budget, and an annual
report. However, the tasks also include deciding on priorities among different research areas,
to decide on the initiation of new programmes/projects, allocation of resources, follow-up and
evaluation of the quality, results and effects of the research and innovation programmes and
projects. The Centre Director reports to the Board and has the main responsibility for
successful implementation of its decisions, including budget decisions. The latter
responsibility includes tasks in relation to the management and follow-up of projects and
other research and innovation activities. In addition, the Director has the important overall
task to act as a driving force for the development of the Centre.
 9

4. Interactive Research in Practice: An Emerging Model

The HELIX’ research programme is carried out within an interactive research framework. In
line with the definition proposed in section 2, this means a strong emphasis on contributing to
long-term, critical theory development as well as to learning and innovation processes in
partner organizations. Interactive research as understood here differs from action research in
stressing the independence of the researcher, the necessity of having a critical distance and the
focus of long term theory development (see also Svensson, Ellström & Brulin, 2008).

The commitment to interactive research should not, however, be interpreted as a belief in the
existence of one research strategy that represents in some sense the single best approach for
all types of research problems. On the contrary, according to our view of interactive research
there is a need to – depending on the problem – work with a mix of studies ranging from
descriptive-analytical studies (e.g. surveys or qualitative case studies) to different types of
evaluation or follow-up studies, and to studies based on different forms of interventions. In
line with this, a wide range of methods of data collection is employed, including interviews,
focus groups, questionnaires or surveys. Independent of the type of study, a strong focus on
reflective, theoretical work is a common core.

The overall model for interactive research to be used (see Figure 2 below) is developed on the
basis of several years of experience by the research group in carrying out research based on
interactive research strategies in different contexts (Ellström et al., 1999; Svensson, Ellström
& Brulin, 2007).

Figure 2: Interactive Research as a Two-Way Flow of Problems and Knowledge

The model presented in Figure 2 is proposed as an ideal typic model of knowledge creation
through interactive research. The model depicts two interacting systems, called the research
system and the practice system, respectively. Both systems are viewed as cyclical in character
and driven by problems/issues originating in research or practice. The basic activities in both
 10

systems, that is, research activities (e.g. data collection and analyses) and different kind of
organizational actions are assumed to be informed by theoretical and practical knowledge
based on previous research and/or practical experiences. Theoretical knowledge refers to
conceptual, codified, and explicit knowledge that is more or less valid across different
contexts. The practical knowledge assumed to guide organizational action (called local
theories in Figure 2) is seen as more contextual (situated), experience-based and difficult to
articulate, that is, tacit or implicit (cf. Ellström, 2006 a). Of course, this distinction should not
be seen as in some sense an absolute divide between two qualitative different forms of
knowledge. On the contrary, as generally agreed, theoretical knowledge has tacit (implicit)
dimensions, and practical knowledge is assumed to be more or less dependent on theoretical
assumptions (cf. the expression local theories).

Furthermore, in order to avoid a too rationalistic interpretation of the learning loops in Figure
2, it is important to underline that the activities in the two systems are assumed to be shaped
not only by cognitive-theoretical factors. A range of other factors related to the participating
individuals as well as to organizational and societal conditions are assumed to influence the
activities that are undertaken as part of an interactive research process.

It should also be underlined that the two interacting systems can refer to and be defined at
different system levels. Thus, the research system may, at a micro level, refer to a single
researcher or a project team; at a higher system level to a research programme, a research
department, or a centre like HELIX; and at still higher system levels to a university or even a
national or multinational research system. Correspondingly, the practice system my refer to
different system levels from a single organization or a department or work group within an
organization to clusters of organizations (e.g. a certain business) at a regional, national or
global level. In the text below, I will try to comment on some of the core elements of this
model of interactive research.

Two Interlocked, Collective Learning Cycles

A basic characteristic of the model depicted above, is that the process of interactive research
is assumed to produce conceptualizations and interpretations of a research object that are fed
back as cognitive input into the next cycle of practical problem-solving activities, but also into
the next cycle of research activities. This is indicated in Figure 2 by the shaded circle in the
intersection between the two systems that is assumed to ”move” along the horizontal arrows
representing a time axis. Considering this cyclical process of knowledge creation and use, the
two activity systems may be seen as two interlocked, collective learning cycles that produce
successive versions of common conceptualizations of the research object. In this way,
researchers and practitioners are assumed to co produce common knowledge and
understanding of the research object through joint exploration and analysis of data.

As argued already in the introductory section, the model of interactive research outlined in
this paper claims to represent an alternative both to conventional more linear models of
research and to different models of action-research. What, then, do these claims mean more
specifically? First, the model presented here is interactive in the sense of attempting to build a
two-way flow of problems and knowledge between research and practice. This in contrast to
the common assumption made by traditional models of research, that there are predefined
practical problems that are waiting to be solved by theories, results, and methods derived from
research, and disseminated to the potential users. This may certainly be true in a few cases,
but usually you cannot expect to find workable solutions to predefined practical problems.
 11

Furthermore, many real world problems are too complex to be captured by any one actor or
perspective (Caswill & Shove, 2000; Van de Ven & Johnson, 2006). Rather, the problems
typically have to be analysed and redefined through an interactive process between
researchers and practitioners. In fact, in many situations an interactive process of problem
finding and diagnosis is required before possible solutions could be considered and applied.

In line with this, the model presented here starts with a joint process of diagnosis and problem
finding, where the researchers and practitioners together try to conceptualise and define the
problems and issues that should be addressed. This problem-finding process may sometimes
require an empirical study of the problem area (”a pilot study”) in order to be able to define
the problems. Thus, as argued also by Van de Ven & Johnson (2006), the practice system (for
example, a workplace) is viewed not only as a site for data collection, but also as a site for
idea generation and learning where researchers and practitioners co produce and test
alternative ideas about of a common research object.

The joint processes of individual and collective learning that are assumed to be essential to
interactive research can be viewed as an interplay and integration between theoretical and
practical knowledge, that is, an integration between explicit, codified knowledge and
experiential, situated knowledge. Several frameworks from the field of organizational
learning could be helpful in understanding this interplay and integration. One such framework
would be the model proposed by Nonaka & Takeuchi (1995). This model assumes that
knowledge creation could be understood as a cyclical process of knowledge conversion based
on the interaction between tacit and explicit knowledge. Four modes of knowledge conversion
are distinguished: socialization, externalization, combination, and internalization. Other
relevant ways of understanding learning in this context are the distinction between learning as
exploration versus exploitation proposed by March (1991), and the model of expansive
learning proposed by Engeström (1999). In line with these latter conceptualizations, the
present author would view the knowledge creation process as a process of developmental
learning with an emphasis on exploring and questioning existing conditions, solving
ambiguous problems, and developing new solutions. In this view, conflicts and ambiguity are
not potential threats to learning or to efficient performance, but rather potentials for triggering
developmental learning, and break out of routinized and taken for granted ways of thinking
and acting (Ellström, 2006 b).

Defining an Integrative Task

In what ways, then, do the model outlined above differ from typical forms of action research?
Could not in fact much of what has been said so far equally well be said about many kinds of
action research? One main difference between this model and many models of action research
concerns what we above called the threefold task of interactive research and how this task is
handled. As discussed in section 2, a common way of handling the threefold task within the
tradition of action research is to subordinate the research task to that of practical problem
solving and change, that is, to reduce interactive research to a strategy for research-based
developmental work.

Contrary to this way of handling the threefold task, we have tried another strategy based on
the idea of identifying and negotiating a common, and potentially integrative task between the
research system and the practice system, that is, to identify a kind of common denominator. In
this way it is, at least in principle, possible to integrate different perspectives (e.g. perspectives
of researchers and practitioners) on the research object in the initial formulation of the
 12

research task or problem. To the extent that such integration is successful, it is likely to
increase both the quality and the relevance of a research problem (cf. Pettigrew, 2001).

In one research project focusing on change programmes in an industrial enterprise (Ellström et

al, 1999), the integrative task was formulated in general terms as the task of: ”reaching a
common understanding of the ongoing change process that could be viewed as significant
both from the perspective of practice and from the perspective of research”. In accordance
with this task, the outcomes of the research activities were fed back at particular ”feed-back
seminars” to the practice system and used as cognitive input in the management of the change
programme. This integrative task as stated at the beginning of the research process should be
open to re-examination and redefinition during the whole research process. Furthermore, it is
of course also necessary to jointly specify this general task, and to define more specific and
operational research tasks already at the beginning of the research process.

Of course, it is difficult to know to what extent one in practice succeeds in creating such a
common understanding of the research object. However, there is evidence that it is at least
possible to create an interest in searching for such a common understanding (Ellström et al,
1999). One indication of this was the comment made by one production manager concerning
the company's experiences of participating in an interactive research project for some time:

“…before we believed that there were two kinds of companies; companies that make things happen and
companies that watch things happen. Our goal was to be a company of the first kind. Now we understand that
there is a third type of company. Our goal is now to become the company that makes things happen and try to
understand why they happen”.

Similar observations concerning the possibility for researchers and practitioners alike to meet
in a search not only for empirically based instrumental knowledge or practical solutions, but
also in a search for new conceptualizations, perspectives or ideas of a common research object
is also attested to by others (e.g. Caswill & Shove, 2000; Van de Ven & Johnson, 2006). As
stated by the former authors, practitioners are ”often more attracted by new ideas and
concepts than by empirical material.” (p. 221).

An Alternation Between Research- and Practice-Oriented Activities Over Time

Although the strategy of defining an integrative task in certain respects have proved to be able
to handle the tensions created by the so called threefold task of interactive research, it is
hardly sufficient for reconciling the different aspects (subtasks) of this task, that is, the
subtasks of carrying out scientific knowledge production, practical, change-oriented activities,
and learning activities. In practice it has proved necessary to use the strategy of creating an
alternation and interplay between research-oriented and practice-oriented activities over time.
Thus, while practice-oriented activities are allowed to predominate during certain periods of
time, more purely research-oriented activities (e.g. analyses of data, writing research reports)
must be allowed to predominate during others. In other words, one needs to alternate between
practice-oriented activities and activities oriented towards critical reflection and knowledge
production.

This tension between different subtasks of interactive research is also addressed by Eikeland
(2006), and he reaches a conclusion similar to the one indicated here. As argued by Eikeland
(2006), it is useful to make a distinction between ”on-stage-performance” at the workplace,
that is, activities related to data collection or practical activities, and ”back-stage-reflections”
including learning activities, analysis, and critical reflection. Using this distinction as a point
 13

of departure, one of his main conclusions is the necessity to build capacity to systematically
alternate between performing ”on-stage” and engaging in critical analysis and reflection
”back-stage” (Eikeland, 2006, p. 232). Similar to the idea of the three-fold task advanced
above, such an alternation is assumed to be fruitful from the perspective of the validity of
research results, but also from the practical perspectives of organizational learning and
change.

A Clear Division of Labour

Another important characteristic of the model outlined above in relation to many models of
action-research, concerns the character of the co-operation between researchers and
practitioners. The co-operation between the researchers and the practitioners involved in the
research process should, according to our experiences, be based on a division of labour in
which traditional academic values are kept intact, that is, the right to be critical, the
development of new theories and general knowledge, a long-term perspective on knowledge
production etc. Thus, also in these respects, interactive research differs from action research
(Svensson, Ellström & Brulin, 2007).

From the first contacts with the practice system, a clear division of labour between the
researchers and the practitioners is agreed on. Thus, rather than trying to erase the differences
between the spheres of research and practice we believe that it is very important to respect
and preserve these differences. This means that the role of the researchers is clearly separated
from the role of being a consultant or change agent in the sense of an advisor or expert that
prescribes certain solutions to "given" problems. Rather, as researchers we define our primary
task as the task to contribute to a common conceptualization and interpretation of the research
object (see Figure 2), and to participate in joint arrangements (e.g. seminars) in order to give
feed-back, discuss and explore different conceptions of this object (e.g. a process of
organizational change).

The primary task of the practitioners is to manage, carry out and take responsibility for the
process of action within the organization. Decisions concerning possible implications for
practical action, and the management of the organizational changes that may be undertaken on
the basis of the research is the task of the managers of the practice system. Of course, this
does not mean that the researchers are avoiding practical issues. On the contrary, the task of
the researchers is to support in different ways (e.g. as dialogue partners) the management of
change and innovation processes for example through arranging joint seminars with a focus
on interpretation and implementation of research findings. Finally, the common primary task
of both researchers and practitioners is to engage in a mutual effort to conceptualize, reflect
on and learn from the ongoing change processes.

A Negotiated Agreement Concerning Research Problems and Research Design

It is important to reach an initial agreement between researchers and practitioners concerning

research problems and the general design of the research project. This early negotiation phase
should result in a project plan. As argued above, it is also very important to minimize role
ambiguity regarding the researchers and practitioners by defining these roles quite early in the
research process. The joint problem definition phase can in this perspective also be seen as a
role definition phase, where questions of responsibility and division of labour in the co-
operative process are important ingredients.
 14

In some projects a project steering group has been created with management and employee
representatives. Here, issues and problems are analysed in a process of diagnosis and problem
finding in interaction between the researchers and practitioners. Areas of interest could be
raised from the researchers as well as from the practitioners in the organization. This phase
has proved to be important in several ways:

• It involves researchers and practitioners in a mutual effort to express different expectations on

the planned research process;
• It provides researchers and practitioners with opportunities to discuss what roles initially, and
over time the parties should have in the research process; that is, it becomes a process of
interpretation and sense-making regarding roles and expectations in relation to the research
process;
• It has the potential to create a mutual understanding of questions and issues at stake;
• It could be regarded as a supportive recurrent activity if the project definition, problems,
strategies etc. need to be refined and/or redefined.

From a research perspective, it is important that the research project has the explicit aim to
contribute not only or primarily to practical concerns, but also to the development of
theoretical knowledge, that is, new concepts, theories, and models. Without an explicit and
firm orientation towards theoretical development, there is an obvious risk that practical
concerns drive out more theoretically oriented interests. To a certain extent this is simply a
function of time restrictions. The practical activities are often very time consuming, and there
is the risk that the immediate and often short-term practical concerns are automatically given
priority over more long-term research activities (Adler, Shani & Styhre, 2004). An important
means for avoiding this is that the research project and the problems to be approached are
founded on an explicit theoretical framework, and based on results from previous research in
the problem area.

The main function of an articulated theoretical framework is of course to guide the collection,
analysis, and interpretation of data. However, in interactive research projects the theoretical
framework might, in addition, offer conceptual tools that can be used both in the initial
discussions and negotiations concerning what problems to address, and as common tools of
reflection and analysis during feed-back seminars. Furthermore, in order to be able to use
interactive research projects as a basis for engaging undergraduate students and PhD-students
to do their theses, it is essential that the research work has a firm theoretical grounding, and
uses the more or less established methods of research within a certain field. Not least in this
perspective, giving up on ”the written word” as a main product of interactive research as
advocated by Reason & Bradbury (2001), would hardly be recommendable if we want to
increase the legitimation of interactive research within universities.

The Importance of Broad Support

HELIX projects are generated in a process involving the regional partnership and the Board
with representatives from the university, companies, public sector and labour market
organizations (see section 3 above). Thus, the projects are discussed and scrutinized from the
perspectives of a number of parties representing different knowledge interests and ideologies.

Also at the organizational level (e.g. at a company) it is important to assure that there is a
broad support for a certain project. Our experiences indicate that support only from a certain
department or from certain persons within an organization is not enough for securing access
to the research object and other necessary preconditions for carrying out a certain project.
 15

Specifically, the importance of management support for the successful conduct of interactive
research projects can hardly be overestimated. However, equally important is to have strong
support from concerned unions within the organization, and, of course, from concerned
groups of employees that are expected to participate in the project. This is vital not only from
the perspective of getting access to the organization, but also for the possibility to have an
impact on change or innovation processes within the organization. Thus, in this sense,
interactive research assumes a certain degree of mutual trust and co-operation between the
management at different levels and the employees and unions in the organizations that are
supposed to participate in a research project. Without this mutual trust, it is doubtful if it is at
all possible to try to engage the organization in a process of interactive rsearch. By
implication, it is important to explore during the first contacts with an organization in the
planning of a research projects to what extent these conditions prevail.

The wide range of participants with different knowledge interests and perspectives that are
involved in the generation and conduct of an interactive research project is significant also in
other respects. First, the involvement of practitioners in the choice of research object and
formulation of research questions is likely not only to increase the relevance of research, but
also to sharpen and increasing the quality of the research questions. Second, there is also
evidence suggesting that the involvement of practitioners representing different perspectives
in the research process is likely to increase the validity (impartiality) of the findings (Van de
Ven & Johnson, 2006).

Of course, the diversity of interests and perspectives involved in this type of interactive
research can also be expected to generate conflicts, not only among the different partners
involved (e.g. partners representing employer interests versus union interests), but also among
the researchers coming from different disciplines. Depending on how such conflicts are
experienced and managed they can be creative or destructive for the research process. It is
important to interactive researchers to be aware of the risks and possibilities related to these
types of conflict, and a challenge to develop methods of constructive conflict management.

5. Concluding Remarks

The interactive research approach presented in this paper should be viewed, not as fixed, but
rather as an emerging model of knowledge creation through interactive research that need to
be continually modified and improved. However, the model – in its general outlines – seems
so far to be promising. As argued already at the beginning of this paper, interactive research
differs from action research in its alleged capacity to promote both practical concerns and
traditional academic values, that is, the right to be critical, the development of new theories
and general knowledge, a long-term perspective on knowledge production etc. This might be
true at philosophical level, whether or not it is true in practice remains still an open – and
challenging – question.

Among the difficulties experienced in the application of an interactive research approach is

that it is quite resource demanding in terms of time for data collecting and analysis, followed
by feedback seminars and workshops involving different groups within the organization. As
mentioned above – if this issue is not handled in an appropriate way – the relatively strong
resource demands in terms of time and effort may constitute a threat to the possibilities for the
researchers to engage in long-term theoretical development.
 16

An important dimension of the so called threefold task of interactive research (see section 2)
concerns the practitioners opportunities to learn and benefit from research in a longer
perspective. In line with this presumption, several practitioners involved in interactive
research projects within the HELIX Centre have expressed that the interactive approach has
created an arena for reflection and conceptualization, which did not previously exist within
their organizations. Participation in this type of research project is also reported to provide an
opportunity to get in contact with researchers that can give an outside perspective and update
with current research on certain problems. In the same spirit, several partners have expressed
their interest in participating in interactive research projects compared to engaging consultants
for addressing certain problems. As one production manager put it: “many consultants tend to
give too many good models and answers, but too few good questions to connect these answers
to".

A final comment. The focus in this paper has concerned what may be called ”the macro
design” of interactive research projects, that is, the process from the initial choice of research
object and definition of research questions, over data collection and analysis, to dissemination
and implementation of research findings. The attention given to issues related to ”the micro
design”, that is, methods for interacting with practitioners in the analysis, interpretation and
conceptualization of data, has been quite limited, if dealt with at all. Although these issues
were considered as outside the scope of the present paper, they are certainly central for
developing the methodology of interactive research.

6. References

Aagaard Nielsen, K., & Svensson, L. (2006). (Eds.). Action Research and Interactive
Research. Beyond Theory and Practice. Maastricht: Shaker Publishing.
Adler, N., Shani, A. B. (Rahmi), & Styhre, A. (2004). Collaborative Research in
Organizations. Foundations for Learning, Change, and Theoretical Development.
Thousand Oaks, CA: Sage Publications
Argyris, C. (1980). Inner Contradictions of Rigorous Research. San Diego, CA:
Academic Press.
Aronsson, G., Dallner, M. & Gustafsson, K. (2000). Yrkes- och arbetsplatsinlåsning. En
empirisk studie av omfattning och hälsokonsekvenser. Arbete och Hälsa, 5, 1–24.
Asheim, B. T. (2001). Learning regions as development coalitions. Partnership as governance
in European workfare states? Concepts and Transformation, 6, 1, 73–101.
Brown, P., Green, A. & Lauder, H. (2001). High Skills. Globalization, Competitiveness, and
Skill Formation. Oxford: Oxford University Press.
Brulin, G. (1998). Den tredje uppgiften. Högskola och omgivning i samverkan.
Stockholm: SNS Förlag & Arbetslivsinstitutet.
Brulin, G., Ellström, P.-E., & Svensson, L. (2003). Interactive Knowledge Formation –
A Challenge for Swedish Research and Higher Education. Paper presented at HSS–03,
May 14–16, 2003, Ronneby, Sweden.
Caswill, C., & Shove, E. (2000). Postscript to special issue on interactive social science.
Science and Public Policy, 27, 220–222.
Chein, I., Cook, S. & Harding, J. (1948). The Field of Action Research. The
American Psychologist, 3, 43–50.
Clark, A. W. (Ed.) (1976). Experimenting with Organizational Life: The
Action Research Approach. New York: Plenum Publ. Co.
 17

Edquist, C. & Flodström, A. (1997). Den tredje uppgiften: universitet och samhälle i
samverkan. I: S. Dahl (red.) Kunskap så det räcker? Arton debattinlägg om utbildning
och forskning. En debattbok från SACO och SULF. Stockholm: SACO.
Eikeland, O. (2006). Validity of Action Research and Validity in Action Research. In: K.
Aagaard Nielsen & L. Svensson (Eds.). Action Research and Interactive Research.
Beyond Theory and Practice. Maastricht: Shaker Publishing.
Ellström, P.-E. (2006 a). The Meaning and Role of Reflection in Informal Learning at Work.
In: Boud, D., Cressey, P. & Docherty, P. (Eds.), Productive Reflection. An Anthology on
Reflection and Learning at Work. London: Routledge.
Ellström, P.-E. (2006 b). Two Logics of Learning. In: E. Antonacopoulou, B. Elkjær, & P.
Jarvis (Eds.), Learning, Working and Living. Mapping the Terrain of Working Life
Learning. London: Palgrave Macmillan.
Ellström, P-E., Eklund, J., Kock, H., Lindström, L., & Melin, U. (1999). Knowledge Creation
Through Collaborative Research: An Emerging Model. Paper presented at HSS –99
Conference, March 16-18 1999, Falun. Also in: CMTO Research Papers, CMTO,
Linköping University.
Engeström, Y. (1999). Innovative learning in work teams: Analyzing cycles of knowledge
creation in practice. In Engeström, Y., Miettinen, R. & Punamäki, R. L. (eds.)
Perspectives on activity theory, pp 19–38. New York: Cambridge University Press.
Etzkowitz, H. (1998). The Norms of Entrepreneurial Science: Cognitiv Effects of the New
University-Industry Linkages. Working Paper. New York: Science Policy Institute, State
University of New York.
Etzkowitz, H. (2003). Innovation in innovation: the Triple Helix of university – industry –
government relations. Social Science Information, 42(3), 293-337.
Gibbons, M. et.al. (1994). The New Production of Knowledge. London: Sage.
Gustavsen, B., Finne, H. & Oscarsson, B. (2001). Creating Connectedness. The Role of Social
Research in Innovation Policy. Amsterdam: John Benjamin.
Heron, J., & Reason, P. (2001). The Practice of Co-operative Inquiry: Research ‘with’ rather
than ‘on’ people. In: P. Reason & H. Bradbury (Eds.). Handbook of Action Research.
Participative Inquiry and Practice. London: Sage Publications.
Lawson, C. & Lorenz, E. (1999). Collective Learning, Tacit Knowledge and Regional
Innovative Capacity. Regional Studies, 33, 4, 305–317.
Malmberg, A. (2002). Klusterdynamik och regional näringslivsutveckling –
begreppsdiskussion och forskningsöversikt. Stockholm: Institutet för tillväxtpolitiska
studier, A2002:008.
March, J. G. (1991). Exploration and exploitation in organizational learning. Organization
Science 2, 71–87.
Michie, J. & Sheehan, M. (2003). Labour market deregulation, 'flexibility', and innovation.
Cambridge Journal of Economics, 27, 123–143.
Nonaka, I. and Takeuchi, H. (1995). The knowledge-creating company: How Japanese
companies create the dynamics of innovation. Oxford: Oxford University Press.
Nowotny, H., Scott, P. & Gibbons, M. (2001). Re-thinking Science. Knowledge and the
Public in an Age of Uncertainty. Cambridge: Polity Press.
Pettigrew, A. M. (2001). Management research after modernism. British Journal of
Management, 12, 61–70.
Rapoport, R. N. (1970). Three Dilemmas in Action Research. Human
Relations, 23, 499-513.
Reason, P., & Bradbury, H. (2001). Introduction: Inquiry and Participation in Search for a
World Worthy of Human Participation. In: P. Reason & H. Bradbury (Eds.). Handbook
of Action Research. Participative Inquiry and Practice. London: Sage Publications.
 18

Rosenthal, R. (1966). Experimenter Effects in Behavioral Research. New York: Appleton-

Century-Crofts.
Sandberg, Å. (1981). Om arbetslivsforskningens metoder och förutsättningar. I: Å. Sandberg
(red.) Forskning för förändring. Om metoder och förutsättningar för handlingsinriktad
forskning i arbetslivet. Stockholm: Arbetslivscentrum.
Seashore, S. E. (1976). The Design of Action Research. In: Clark (1976).
Svensson, L., Brulin, G., Ellström, P-E., & Widegren, Ö. (2002). Interaktiv forskning – för
utveckling av teori och praktik. Stockholm: Arbetslivsinstitutet.
Svensson, L., Ellström, P.-E., & Brulin, G. (2007). Introduction – On Interactive Research.
International Journal of Action Research, 3, 3, 233–249.
Sörlin, S. (1996). Universiteten som drivkrafter. Globalisering, kunskapspolitik och den nya
intellektuella geografin. Stockholm: SNS Förlag.
Toulmin, S. & Gustavsen, B. (1996). Beyond Theory: Changing Organization through
Participation. Amsterdam: John Benjamin Publishing Company.
Van de Ven, A. H., & Johnson, P. E. (2006). Knowledge for theory and practice. Academy of
Management Review, 31, 4, 802–821.
Woolgar, S. (2000). Social basis of interactive social science. Science and Public Policy, 27,
165–173.